Ammonium adducts in ACN vs MeOH

[Camisotro]

I'm working in a lab where we're trying to get detection methods set up for a list of pesticides by LC-MS/MS. One particular class of compounds, synthetic pyrethroids (e.g. cyhalothrin, deltamethrin, permethrin, flucythrinate) was really leaving me stumped - using a gradient method recommended by Agilent with ACN/water and ammonium formate, I just couldn't see those compounds clearly by MRM or in full scans. I tried using some isocratic methods so that the full scans would be less noisy than with a gradient, and same deal.

On a whim I changed the system over to methanol/water with ammonium acetate. Using similar isocratic methods I found all of the above compounds "loud and clear" so to speak, with the strongest peak being the [M + NH4]+ adduct. I changed the aqueous phase back to ammonium formate additive while still using methanol, and got essentially the same chromatogram and spectrum.

Finally I switched back to acetonitrile/water with formate, and this time I did see the adduct, but the intensity was a good two orders of magnitude lower.

Could there perhaps be an impurity in our acetonitrile that is selectively preventing ammonium adduct formation? I can't really think of any other explanation at this point. I've never seen such a stark difference in signal between ACN and MeOH. Any thoughts from experienced folk would be welcome.

[Kenn]

I can't answer your question as to why but would say MeOH is usually more conducive to adduct formation. I usually infuse in MeOH:H2O:formic and have on a number of occasions switched to ACN:H2O:formic and the adduct disappears or is greatly reduced.

I just finished a rather extensive study with one of the pyrethroids you mention and monitored the NH4 adduct. Used 10 mM ammonium acetate and methanol gradient with low std of 20 ppt. Never could have gone there without monitoring the adduct. Is there a reason you can't make your method using a methanol MP?

[Camisotro]

I can't answer your question as to why but would say MeOH is usually more conducive to adduct formation. I usually infuse in MeOH:H2O:formic and have on a number of occasions switched to ACN:H2O:formic and the adduct disappears or is greatly reduced.

I just finished a rather extensive study with one of the pyrethroids you mention and monitored the NH4 adduct. Used 10 mM ammonium acetate and methanol gradient with low std of 20 ppt. Never could have gone there without monitoring the adduct. Is there a reason you can't make your method using a methanol MP?

It's mainly for convenience: We have the Agilent pesticide MRM database which includes retention times with every compound for a specified combination of column, mobile phase and method. This allows us to quickly put together dynamic MRM methods that look for specified transitions only in particular time windows, so that we can screen a large number of compounds at once without wasting the duty cycle.

We could certainly switch to methanol if we had to, but it would just mean finalizing a new method, and then sweeping through our compound library to store new retention times for everything. It's also unclear to me as to whether any other key pesticides in the multi-residue method would do any worse with MeOH.

I also found today that a few of the JetStream source parameters that Agilent selected for their multi-pesticide method make a big difference in whether we can see the NH4 adducts at all. In particular: turning the sheath flow from 10 to 12 L/min, the gas flow from 6 up to 10, the nebulizer gas from 35 up to 50 psi - taken together, all of these changes from the Agilent method bring up the NH4 adduct signal in ACN by more than an order of magnitude, whereas using Agilent's setting will bury the full scan signal so deeply in noise that I can't find it at all. Nonetheless MeOH is still giving better signal by a massive factor.

I can also report that I observed today more than a twofold difference in signal just by using one ACN bottle versus another - and both were HPLC Chromasolv grade but not LC-MS grade. So now we need to try again when our order of LC-MS grade ACN arrives.

[lmh]

I have had, in the past, some chemicals which simply do not ionise by electrospray in acetonitrile, but which work very well in methanol. I have no idea why!

[Klaus I.]

Finally I switched back to acetonitrile/water with formate, and this time I did see the adduct, but the intensity was a good two orders of magnitude lower.

And what are the ratios for other adducts like M+H, M+Na or M+K?

[Camisotro]

Finally I switched back to acetonitrile/water with formate, and this time I did see the adduct, but the intensity was a good two orders of magnitude lower.

And what are the ratios for other adducts like M+H, M+Na or M+K?

I don't have the data in front of me, but they were all present. M + NH4 was the most prominent ion whenever the compounds were detectable at all. It's just that under the conditions where I couldn't see it, none of the adducts showed up appreciably.

Someone at Agilent suggested we might also have some old ACN that has a bit of polymerization, which would result in some ion suppression.

[Alp]

Just a thought... have you checked for M+ACN+NH4+??
(or M+ACN+ H+??)

[Camisotro]

Just a thought... have you checked for M+ACN+NH4+??
(or M+ACN+ H+??)

I would have noticed it if it were there. Anyway I'm now using methanol. We got in some proper LCMS grade ACN, and when I get a chance I'll check again if it looks any different with better solvent.

[hatlevis]

Finally I switched back to acetonitrile/water with formate, and this time I did see the adduct, but the intensity was a good two orders of magnitude lower.

And what are the ratios for other adducts like M+H, M+Na or M+K?

I don't have the data in front of me, but they were all present. M + NH4 was the most prominent ion whenever the compounds were detectable at all. It's just that under the conditions where I couldn't see it, none of the adducts showed up appreciably.

Someone at Agilent suggested we might also have some old ACN that has a bit of polymerization, which would result in some ion suppression.

Can you tell me what was meant by "polymerization."