UPLC/TQD deviation

DM

<p><span>Hello, we have the UPLC/TQD instrument. Just recently we found the deviation between two injections from the same solution is greater than 5%. Anyone knows what should do to make the deviation less than 5%? </span></p>

lizh

Hello:

We will need a little more data, we need chromatogram, method, gradient, injection mode, injection volume, sample and sample diluent column, MS set up conditions etc. Is this a method that you run routinely? Is it just this method or all methods? Anything change in the system lately, routine maintenance event etc.

Let us know and we can decide where to start troubleshooting.

Liz

DM

Than you Liz for your reply. WE use: full Loop, acetonitrile/ water gradient (both contain 0.2% formic acid). Injection volume 5ul, urine sample, using BEH-C18(50mm) column. MS is using MRM for the quantification. We use this method for the routine analysis. It had been running very well until two months ago. Equipment PM is done at end of May, 2009. We did the calibration a month ago using waters API calibration solution but the deviation happened before that. We found this because we prepared a new standards solution, which was needed to compare with the previous solution. The deviation of two injections for the same solution is always larger than 5%. We did the same standards comparison in May and June, 2009. It seemed very good at that time. WE have not tested the other methods. But I am thinking all the methods might have such a problem. We have cleaned the cone and changed a new column, but those do not help reducing the deviation. Another thing we got the new 96 wells cover two months ago, I am not sure this is a reason (I will test it).

Dong

RD

What was the deviation in May and June, 2009 ?

5 % deviation for MS detection seems reasonable to me but if you previously achieved better results using the same conditions, then I am sure you are puzzled as to why the performance has deteriorated.

It would help to determine if the variation is due to the MS or to the inlet (UPLC). Do you have a UV detector you can use to determine the health of the inlet ?

Rich

DM

Our previous deviation was only two or three percents (or less) from the same solution (methanol or acetonitrile solution). As we need the deviation of two standards prepared at same time or different time is less than 5% (compared by area, each injected three times), then we can use them for the standard curve. (Sorry I did not describe clearly, it is not the testing sample with IS, that is ok over 5%). We have found a reason in recent days; it might cause by the 96 well cap mats. We put the solution in separate wells and each well just injects once, the deviation is reduced a lot and meets our requirement, if two injections from the same well, the deviation is larger, weird? We have a UV detector. How to determine the health of the inlet? Thanks

D

lizh

Hello:

I think that it is your cap mat, the puncture/pre-pirecing needle is becoming contaminated. Take a look next time you inject. It could be time for a sharper puncture needle, or just clean up any remaining cap mat bits from the surface of the pre-piercing needle. If you are OK with vials then the system is not contaminated.

You may need to change the cap mats to a style that does not leave bits. What type are the mats and what needle (injection needle) are you using?

many TX

Liz

RD

Because MS measurements are much more variable than UV measurements, I test the precision of the inlet by using a UV detector alone whenever I experience poor precision with an LCMS system. This will tell me if the problem is with the injector. You can use any UV absorbing compound and make repeated injections from the same vial.

DM

we use steel needle and waters cap mats 186002484