Random response of some compounds in the chromatography

[martinabelli]

Hi all,
We have in the lab an LC-DAD, and we run cannabinoids on it.
Like every beginning of the week we prepare a fresh calibration from stock to run our samples.
It's the second time that our middle standard plays up and gives us lower response for some of the compounds (I have to admit there is a pattern, they are all the acidic compounds).
But please bare in mind this:
- all the reagent are exactly the same we use every week.
- the middle standard is the only one affected
- I have twice exchange the "faulty" vial with a new one (we prepare the calibration in series, from the same stock, using the same solvent for dilution) and the response is now ok.
This tells me it cannot be the preparation (the stuff that is in the faulty vial is the same that is in the vial that is now working), it cannot be the mobile phase (for what I have just mentioned, plus the other standards are fine).
We used also the same consumables. I cannot find an explanation of what could cause this. Any idea? Thank you in advance

[TylerSmith123]

Hi Martina,

What is the exact standard you are using in the middle standard (I think it should be a few compounds by your comment)? When you say that this middle standard contains acidic compounds, are they all acidic and acting in a similar manner? Do you notice anything peculiar like peak-shoulders or the peak-width increasing compared to your ideal runs? If these compounds are acidic, are you calculating the amount you should see on DAD based on the mass of the free acid or in conjunction with the acids' conjugate ion pair (ie, sodium, calcium, etc.), or was the original standard a different formulation of these acid standards?

Sorry I'm not being very specific, at this point it could be a variety of things but it does not look like it's the fault of the standard itself due to the fact you are able to get the correct response at least partially.

Here's the thing about the faulty vial, there is nothing wrong with the vial-- it's a faulty response you're receiving from a faulty sample that you made. I would certainly not ignore sample preparation simply because you are able to make it once and have it work. A difference in wait time (from preparation) can be enough to change the concentration of a sample if it is particularly volatile or sensitive. Additionally, you cannot excuse the mobile phase because your other standards are okay! The other standards may be completely different molecules altogether and have completely different chemical properties that are each affected separately in your mobile phase. Especially when it comes to acids, your mobile phase may be strong enough to protonate a free acid with a higher pka, but if your standard contains a range of acids then your mobile phase may not be strong enough to completely de-ionize a few of them. I would consider taking a closer look at some of these areas that you ignored, but it would also help if your explained them to us in more detail. You could always order a new standard and check to see if that's your issue, although you have probably already thought of that. The fact that they are all acids suggests sample prep or mobile-phase issues to me, but I do not know the acids or anything specific about your mobile phase or sample prep solutions so I'm just assuming they are okay.

Let us know some more about your sample prep and mobile phase conditions and hopefully we can find something that could be affecting your samples.

TS

[martinabelli]

Hi Tyler,
Thanks for getting back to me.

The standard comes from a stock 10mg/ml in MeOH with 16 cannabinoids compounds. Stock that we already used several weeks and never gave problem, following the same procedure every time.
It seems to me an intermittent problem, because I ran the exact same standard I have prepared (but vialed in a different vial), using the same mobile phase I used yesterday, and it looks perfect.

The compounds are not all acidic, there are CBC, CBD, CBDV, CBG, CBN, THCV, THC (delta9 and and their acidic form, CBCA, CBDA, CBDVA, CBGA, CBNA, THCVA and THCA. Plus CBL.

I cannot see any difference in the peak shape, only the peak size (area).

It's not a new standard, it's a fresh standard. So we start from the 1000ppm for each compound to obtain our stock 10ppm, with all the 16 compounds. The stock lasts 4 weeks, and we use it every week to prepare fresh calibration. We prepare 1ml of each level and we vialed it in multiple insert vials. So literally the same stuff. The stock worked in the past few weeks, and the calibration as well. The calibration I have prepared yesterday had only that one vial of middle standard not working.

With other standards I meant the other levels of the calibration, so same thing, but more or less concentrated.

Mobile phase is 20mM in Ammonium Formate and pH 3 in Water.

I'm sorry, it's really hard to explain everything without having the person in the lab

Thanks for your help!

[TylerSmith123]

Hi Martina,

All of that was extremely helpful with my understanding of things so thank you!

So all of the standard solutions that are injected are essentially the same like you said so there should be no issues with the repeatability (assuming everything is measured precisely and what not, which it probably is) on the solution side-- especially after you made a fresh one. By any chance, is the inconsistency occurring by the end of your standards' shelf-life? Like per-se a week 1 standard versus a week 3 or 4 standard?

Even then, a stock you prepared yesterday was working as expected except for a single concentration dead in the middle. I'm not sure if these are made as serial dilutions (if they are then this point stands) or are just made to the concentration from the main standard solution. But if they were serially diluted, then the middle standard must have something wrong with it other than the solution since the following standards all passed. However, if they are made from the stock solution, then it could be a simple manual error in the preparation of that one solution, although very unlikely.

Back to the vials then, so why are different types of vials used for your standards? Is it simply that they are used on different sized columns or is it that they are simply made up in differently branded vials available to you? If the latter is the case you may want to take a look at your auto-sampler or injection ports-- is the sample being completely injected or is there any noticeable difference between the two vial types volume wise post injection?

When the samples are prepared in the standards are they acidified as well or are they injected within the methanol? Some of your acids may not have time to protonate (and basically gain retention) which could result in some peak area differences in exclusively the acidic compounds. Mayhaps their PKA is higher than your mobile-phase, or similar enough to allow for disruptions? (although that may be a stretch) Additionally, is there any chance for evaporation of the methanol or are they all stored within a cold environment to prevent evaporation and concentration? (Scratch that, you're getting lower responses not stronger)

The volume still may be an issue-- when you inject standards how large is the stock solution that the autosampler has to pull from? Per se, if you're injecting 20 ul onto the column, how large is the sample/standard reservoir? Sorry I'm kind of reaching on a few of these suggestions-- I am hoping it is not some sort of mechanical issue.

Thanks for getting back with all the info!
Hopefully we can solve this soon,
TS

[martinabelli]

Hi Tyler,
Sorry for the delay!

There should be indeed no issues with repeatability, the solutions are always prepared in the same way, and there was no pattern with the "age" of the stock.

The calibration is not prepared as a serial dilution, but still using the same methanol and the same stock, same tips and same pipette (electronic).

When preparing the standard, we do a dilution of the stock basically. This same dilution we distribute in multiple insert vial, to save reagents.
But it seems like only one of the vial had the problem, not the rest, although containing the same thing.

And the same vials are used for all the calibration points.

I am lost with this part, could you explain it to me again?
"When the samples are prepared in the standards are they acidified as well or are they injected within the methanol? Some of your acids may not have time to protonate (and basically gain retention) which could result in some peak area differences in exclusively the acidic compounds. Mayhaps their PKA is higher than your mobile-phase, or similar enough to allow for disruptions? (although that may be a stretch)"

I can tell you the once the standards are prepared we close them with crimp caps and keep them at -20C until use.

Finally, we do inject 3ul and we use a full 250ul insert, so I can't think of that being a problem, otherwise all the peaks would look smaller.

I appreciate your help. I have to admit that since my last message, the problem disappear. Mystery.

I might be back if it will show up again. In the meantime thanks again!