Measuring injection precision

<p><span>Hello,</span></p><p></p><p><span>I'd like a quick and easy way to measure the precision of the Acquity injection volume. Normally, on HPLC, I do repeated water injections out of a vial and determine the difference in weight.</span></p><p><span>On my Acquity, though, doing precision this way is either not the best practice (or my instrument is way out of whack after Waters performed a PM) or my expectations are too high.</span></p><p><span>I have a 20uL loop on there, using the overfill method.</span></p><p><span>I passed precision with a 5uL injection with +0.1 %</span></p><p><span>10uL injections gave +1.5% (but the injections left water drips on the top of the vial...is the injector normally this messy?)</span></p><p><span>2uL injections gave +2.6%</span></p><p></p><p><span>Any opinions?</span></p><p><span>Thanks!</span></p>

Answers

  • Lisa:

    No it should not be this messy, you have a leak, this is probably needle wash and may well be contributing to the poor RSD precision.

    We will send details on next actions.

    A comment about weighing and best pactices, Waters qualification calls for a combined injector accuarcy and linearity test which can be much quicker. Is your SOP possible to change?

    Bear with us.

    Liz

  • Hello,

    This instrument is non GxP, so we can change the procedure easily.

    In the meantime, I believe there is a leak test function for the SM, correct? I'm a bit annoyed, because Waters just serviced this instrument before I tested precision.

    Thanks,

    Lisa

  • Lisa:

    My apologies, it is most likely that the FSE may not noted the issue while he was there. You probably do have a leak of most likely needlewash, but I do not think this is not the root casue of the perfromance of your %RSD's

    It may be wash solution that is not being taken away by the vacuum system intended to do so. If it is not obvious, and I think that you should let the Field Service Engineer know, and he can assist with troubleshooting. However, this is minor.

    As for the Injection Accuracy testing by sample weight, I am told this should meet expectations. Although you might want to discuss the combined test.

    CAn you confirm that you are using PLNO and PLPA mode, 1% to 75% and 1% to 50% is the range (you will see this noted in the online help). It appears that at 5uL you were in good shape. Can you send the needle wash as well as the mode that you are using. All the emthod parameters may be helpful. Have you noted any issues with peak shape, just in case there is a loose fitting. IAn overlay of the peaks would be great).

    Let me know.

    Liz

  • HI Liz,

    I am using PLNO with a 20uL loop.

    wnw: 90:10 water:acn 600uL

    snw: 50:50 water:acn 200uL

    all drawing speeds, etc are default

    I'm using waters brand vials with a preslit cap

    My method is a gradient starting with:

    90% KH2PO4 pH 3.39 with 1% IPA

    10% 40:60 MeOH:ACN

    The peak shapes are fine. I ran 6 injections of a sample and had a strange result (bear with me, I'm just pulling this out of my brain and not actually looking it up): an early peak varied by several %, while a later peak only varied 0.15%. This was with the 5uL injection.

    I'm not sure what you mean by the 1% to 75% statement? To which online help are you refering?

    Thanks!

    Lisa

  • Lisa:

    First pass the needle wash seems good. What is your final gradient strength.

    Apologies on the earlier email. There was a typo as well as a clarity issue!

    As ACQUITY is a loop based injector, you cannot reliably (so says physics of straws) inject more than 50% of the loop or below 10%. Its too hard to suck it in or push it out beyond those limits. With PLNO you can extend that to 75%. So for example in a 20 uL loop, 10 uL in PLPA you are at the limit of the loop's perfromance. Therefore, in PLNO mode you would be fine. That is why we asked. Since you are in PLNO mode, I would tend to make the strong wash strong. You have IPA in the initial conditions so I would add into the strong wash. What is the sample diluent.

    If we are seeing peak shape issues do we have a possiblity of overload? What is the column dimensions? For a 10 uL injection - the column might be too small. Can you list your sample diluent too as that could be a contributor too.

    Regards,

    Liz

  • Hi Liz

    My final gradient strength is 45:55 buffer:organic.

    I am not having problems with peak shape. The column is a 1.5um, 50x2mm Grace Vision HT.

    I'm continuing to run at a 5uL injection. My diluent is either 50:50 ACN:H2O or ACN alone (this project is in a research stage).

    Thanks,

    Lisa

  • Lisa:

    I still fear overload.

    The Rule of thumb is that injection volume should be limited to less than 5% of void volume -2.1 x 50 mm or a max injection of 7.5uL and here the sample diluent is stronger than the initial mobile phase. Could send chromatogram, sometimes pictures worth a 1,000 words.

    Liz

  • Hello Liz

    I am working in pesticides . I obtain good peaks and good RSD, but I am thinking of increasing my volume of injection to raise my LD. But I do not want to overload my column.

    That you would advise me to control ?

    Broad peaks shape, tailing, etc.

    Column: ACQUITY UPLC BEH C18 1.7 um; 2.1 X 50 mm.

    Mobile phase A: 0.01 % formic acid in water

    Mobile phase B: 0.01 % formic acid in methanol

    Flow rate: 0.250 ml/min.

    Column Temperature: 40.0 C

    Loop option: Full Loop

    Injection Volume (ul): 10.0 (water)

    MS Method

    MS System : Quattro Premier

    Ionization Mode: ES +

    Capillary voltaje: 3,5 kv

    Source Temperature (°C) 100

    Desolvation Temperature (°C) 350

    Cone Gas Flow (L/Hr) 50

    Desolvation Gas Flow (L/Hr) 700

    MS/MS: Operated in MRM mode.

    Software: Masslynx 4.1

    Thanks

    Manuel

  • You are already pretty high for injection volume under normal circumastances.

    However, perhaps we can try for some "at line column concentration" playing with injection mode and sample mode. Go to partial loop and a larger loop, but use the WW strength to focus and concentrate. But I will consult some better chemists than I and see what they suggest.

    Liz

  • Thanks Liz, the Forum is very useful for my.

    My question is because , I found papers where they inject a large volume.

    Peter Hancok and col. Application note Waters .
    Determination of pesticides in food using UPLC with polarity switching tandem quadrupole LC/MS/MS
    ACQUITY UPLC BEH C18 1.7 um; 2.1 X 50 mm
    Injection Volume (ul): 20

    Cristiana C. Leandro a,b, Peter Hancock c, Richard J. Fussell a,, Brendan J. Keely. Ultra-performance liquid chromatography for the determination
    of pesticide residues in foods by tandem quadrupole mass spectrometry with polarity switching. Journal of Chromatography A, 1144 (2007) 161-169
    ACQUITY UPLC BEH C18 1.7 um; 2.1 X 50 mm
    Injection Volume (ul): 20

    James Morphet and Peter Hancock.Determination and confirmation of priority pesticide residues in baby food. Application note Waters 2009
    ACQUITY UPLC BEH C18 1.7 um; 2.1 X 50 mm
    Injection Volume (ul): 50

    Manuel

  • Hello:

    That's easy then! I shall just ask Peter directly and get back to you.

    Best regards,

    Liz

  • Liz Sorry

    An information that seems to me to be important is my gradient

    gradient

    Time (min)

    Flow Rate (ml/min)

    %A

    %B

    Initial

    0.250

    90

    10

    1.0

    0.250

    90

    10

    5.0

    0.250

    0

    100

    6.0

    0.250

    0

    100

    6.1

    0.250

    90

    10

    9.0

    0.250

    90

    10

    Regards Manuel

  • Hello again:

    This came from James Morphet.So it was a larger loop.

    I think that the answer is a simple one here - try a larger injection on the short column and see what happens. If it is OK then run with it. If you check out the 400 pesticide apps note (attached) you will see that the work was done with a 20µL injection. If you are still worried about overloading the column then use a longer one - 100mm was used for my work too. I would also advise running a little faster, 0.25 mL / min is quite slow.

    If you send me an email to (elizabeth_hodgdon@waters.com) I can put you in touch with James directly so you can discuss the application directly.

    Liz