LC/MS Precision Issues continue after PM

[StandridgeS]

Hi all,

I posted a while back about having precision issues with my sample (20-40% difference in peak areas). This is for my Agilent 1200 series HPLC with a 6130 MSD. Here is the link to the other post: viewtopic.php?f=3&t=49114

I just had my PM with Agilent where rotors and seals, etc. were replaced in the autosampler which I thought was the cause, but I still see the issues.

The last thing I can think of before getting Agilent repair to tear apart the autosampler is the column. I see slight tailing on my chromatograms and I think this is due to it being an old column. Can LC columns cause precision issues? Do you have any other suggestions as to what I could look at? Are there Agilent MSD standards that I could run to ensure that it is the machine and not my sample causing the issue?

[tkubowicz]

Hello

Have you got DAD detector coupled with MSD? Check if you see the same problem for DAD. If not - it could be MSD problem (ion source/ionization etc). If yes - it is probably LC itself.
If you have 1200 Sampler (G1313/G1329/G1367) check metering device, it has seal inside as well as plunger so perhaps problem is there.
Of course make sure that needle/needle seat and rotor seal are working fine - run pressure test for sampler.

And if your peak is tailing check different method where you have nice sharp, peak - just use caffeine and restriction capillary instead of column.

Regards

Tomasz Kubowicz

[StandridgeS]

I do have a DAD in series, but recently bypassed it because my sample doesn't have any chromophores. So, I know it isn't the DAD.

During the PM the LC passed all its tests (leak, pressure, etc.) and the MSD showed a good tune. The needle and needle seat were replaced, so I suspect the metering device may be the problem. I'll try caffeine and see what kind of precision I get.

Thanks for your input!

[lmh]

Tomasz is right: put the DAD back in line, run something with a chromophore, and compare what's happening when you quantify it using the DAD or the MS. If the problem is related to the metering device (or anything else in the LC system) then you will see identical behaviour by DAD and MS. If you see precision problems in the MS signal but the DAD signal remains very reliably constant, then you have isolated the problem to the MS (which would not be a great surprise: MS has a habit of throwing up strange precision issues).

[Hollow]

in addition maybe also inject 2-3 blanks after your samples to check for carryover

[BostonFSE]

UV detector is best way to isolate problem, Run caffeine through UV at 273nm isocratic and positive mode for MS for 195 ion.
Run 10 injections in a row once equilibrated
The metering head seal on the agilent sampler is not changed during a PM so that be the issue however they should last a long time
Best way to test hardware is with caffeine and keep it simple, once you have confidence in the hardware, then your method, column, sample etc

[StandridgeS]

I injected diluted acetone (1:100) with a MP of isocratic water with a union for both the DAD and MSD, and I saw about 4% RSD and 7% RSD respectively. Does this indicate that something is wrong with the LC, or should the %RSD for both be the same?

I'm probably going to go ahead and replace the parts of the metering device and buy a new column

[tkubowicz]

Hello

Just inject caffeine not acetone and put restriction capillary not an union instead of column. The reason is that you need good back-pressure to have nice sharp peaks.

Regards

Tomasz Kubowicz

[BostonFSE]

A qualified gmp agilent sampler passes at <1%RSD injector precision with a UV, that being said, your work may be fine at 1-2%RSD, depends what your doing...
MS will never be as reproducible as a UV
RSD issue is almost always sampler related, however check your configuration if any parameters got changed like loop size etc.
I have seen some customers have massive RSD issues and it turns out they have a different size loop/syringe than whats in the config. Perhaps something was changed during the PM?
Also, when inserts are used with tiny volumes, the chances of poor RSD increases