False positive results

MopedGirl

<p><span>Hi All,</span></p><p><span>I analyze water sample extracts for 74 pesticides at ppt levels using the Acquity/Xevo and 1 instrument run. Previously I used an Acquity/Premier and have not experienced this issue. I suspect false positive results for 3 pesticides (chlorpyrifos, triallate and ethion- 2 precursor/product ion transitions per compound, exact retention time as standards and ion ratio). The sample injection sequence was:</span></p><p><span>Standard</span></p><p><span>Sample 1 (no positive results for chlorpyrifos, triallate or ethion)</span></p><p><span>Sample 2 (positive results above the reporting limit for chlorpyrifos, triallate and ethion)</span></p><p><span>Standard</span></p><p style="height: 8pt"/><p><span>I reinjected Sample 2 as follows:</span></p><p><span>Standard</span></p><p><span>Sample 2</span></p><p><span>Sample 2</span></p><p><span>Sample 2</span></p><p><span>Standard</span></p><p style="height: 8pt"/><p><span>When the sample was reinjected the reponse for these compounds was reduced by a factor of 20-40X. The response did not change for other positive results for compounds present in sample 2. We are suspicious of false positives because we could not reproduce the response from the original injection. I have noticed this in a few other samples (~3 out of 100) but did not reinject them so I don't know for sure if they were false positives. Ethion solubility in water is 2 ppm, triallate is 4 ppm and chlorpyrifos is 1.4 ppm. All three compounds are the latest eluters (between 10 and 10.5 minutes-and I do see trace levels of these compounds in almost all sample injections varying from 3 to 100 X below the reporting limit).</span></p><p style="height: 8pt"/><p><span>The Acquity conditions are as follows:</span></p><p><span>mobile phase A and weak needle wash: Water with 0.2% formic acid and 5 mM ammomium acetate</span></p><p><span>mobile phase B and strong needle wash: Methanol with 0.2% formic acid</span></p><p style="height: 8pt"/><p><span>Linear gradient from 30 % A to 10% A in 10 minutes, 0.5 minute hold and return to initial conditions for 2.5 minutes</span></p><p><span>0.3 mL/minute flow rate</span></p><p><span>PLNO 10 uL injection with 20 uL sample loop</span></p><p><span>800 uL weak needle wash and 400 uL strong</span></p><p><span>BEH C-18 1.7 uM 2.1 X 100 mm column</span></p><p><span>30 C column temp</span></p><p><span>Sample needle is what ever was shipped with the instrument.</span></p><p style="height: 8pt"/><p><span>Any ideas on how to correct this issue?</span></p><p><span>Thanks,</span></p>

lizh

Hello

This is interesting, so for the same vial (sample 2) only the first injection showed a high response? After this they were reproducible? The standards that bracketed them were reproducible?

Could we see a chromatogram?

What is the sample format - vials, plate etc?

Could you re-inject with a blank in between?

Many TX

Liz

MopedGirl

Liz,

Below are the answers to your questions.

This is interesting, so for the same vial (sample 2) only the first injection showed a high response? yes

After this they were reproducible? yes

The standards that bracketed them were reproducible? yes

Could we see a chromatogram? It is attached. It is sample 75 and not sample 2.

What is the sample format - vials, plate etc? vials

Could you re-inject with a blank in between? I will be injecting water sample extracts tonight and can include blanks.

Thanks,

MopedGirl

lizh

Many thanks

Liz

Newman768

Is it right that you use only water as sample solution?

I experienced (but not very often) problems if a compound is drawn throught a peek capillary (like the uplc standard needle) and no organic solvent is used in sample solution. Even the solubility of the compounds are higher than the concentration there is always the possibility that matrix or metall ions or whatever else leads to adsorbtion effects on the surface of the peek capillary for the compounds AND affected later injections when the adsorption effect comes to an end - most likely it affects only one injection.

We get this problem solved by replacing 5% to 20% of the water used for sample preparation with an (or adding 10% of) organic solvent.

One other possibility is you experienced is a hot spot within your sample. And if you run into the hot spot problem, you have to check your whole sample preparation if there is a step where hot spots could be build or contaminate your sample. Because hot spots can have a wide range of reasons there is no overall solution.