Injection modes Acquity

[Mattias]

Hi,

In the UPLC method (Acquity) for one of our peptides I have always injected 10 µl using the full loop function. Since I wanted to injected more, I had the 50 µl loop / 250 µl syringe installed and qualified by the service engineer.

Now when I inject 20 µl using the "partial loop needle overfill" function I only get 140% of the area of the 10 µl full loop injection. I should get 200%? The injection repetability is very good, so I don't suspect air or leakage.

The sample, column, mobile phase is the same as before the shift. Any ideas of what is wrong?

[tom jupille]

In my past experience, I've found that loop volumes were "nominal", which makes sense because they are not made from precision-bore tubing. Because we calibrate and analyze on the same instrument, loop volume errors either cancel out (filled loop injection) or don't matter (partial loop injection; volume is determined by the syringe).

I'm not sure what "partial loop needle overfill" means, but my guess is that your original "10 microliter" loop may only have actually held 7 microliters. If the present setup was properly qualified, then you have confidence that it is working correctly. If this were my problem, I wouldn't worry about it.

[Mattias]

Hi Tom,

The Acquity system measures the actual loop size when a new loop is installed. Looking at IQ/OQ paper the previous loop was 11.7 µl in size. The new loop is 47.8 µl (but I guess the loop size in partial loop injection is not important).

So I should be comparing 23.4 µl (11.7*2) and 20 µl. But the difference I see is much larger than that!

[Chromatographer2010]

The full loop injection uses a programmed overfill up to 5x I think on Acquity SM. The partial loop injections do not. Basically the sample is diluted when it is moved through the fluidics and some of the sample is left in the needle. With the full loop injection the part that is injected is the heart cut that has minimal dilution.

If you are worried about recovery use full loop. If you want to conserve sample use partial loop.

The new Acquity sample manager (FTN) doesn't move the sample to the valve so the recovery should be close to 100%.

[Mattias]

It sounds reasonable, but then I wonder what is the function of the pre and post airgap in the loop (that the console is showing during injection)? I had a feeling that this was done to prevent diffusion of the sample.

I have plenty of sample, so I can just inject more. And the injection repetability in partial loop mode is excellent (RSD 0.15%). I am more worried what will happen when this method is transferred to another Acquity instrument.

[danko]

Mattias,

Try and validate the actual volume drawn from the vial.
F. ex. weigh a vial filled with water, then inject 5 – 6 times from that vial and weigh it again. Does the weight loss correspond to 5 – 6 times the injected volume?

Best Regards

[tom jupille]

I have plenty of sample, so I can just inject more. And the injection repeatability in partial loop mode is excellent (RSD 0.15%). I am more worried what will happen when this method is transferred to another Acquity instrument.

Unless you are doing your calibration by injecting different volumes of the same standard, then injection volume accuracy is not very important; it's precision that matters.

I much prefer to calibrate by keeping injection volume constant and injecting standards at different concentration for just that reason. The two instruments can inject slightly different volumes, but so long as each is consistent, the differences cancel.

[Mattias]

Thank you for your answers,

Danko> That is certainly possible! I have to take into account that the needle overfills with 15 µl, so I should expect 35 µl of sample used per 20 µl injection (if I have understood the manual)

Tom> I do not use variable injection volume for calibration, so everything should be OK! I am mostly curious why 20 µl is not 20 µl on the Acquity... But also slightly worried if the next system injects 25 instead of than 16 µl, when it is set to inject 20 µl. That could affect the assymetry of the main peak, and the resolution to the following peak (I am on the borderline to overloading since I want to detect impurities and do assay in the same injection).

[danko]

Danko> That is certainly possible! I have to take into account that the needle overfills with 15 µl, so I should expect 35 µl of sample used per 20 µl injection (if I have understood the manual)

The 15 µl assumption may need to be verified. I’d perform the following test:

1. Inject 10 times 10 µl which should result in 250 mg weight loss
2. Inject 10 times 20 µl which should result in 350 mg weight loss
3. Inject 10 times 30 µl which should result in 450 mg weight loss

Besides the needle overfill verification, you’ll also acquire awareness of whether or not you can generate standard curves by injecting various volumes (constant concentration) as opposed to same volume different concentrations procedure.

Best Regards