APCI on electronegative groups

[JA]

We're trying to follow the hydrolysis reaction of a nitrile to a carboxylic acid. So far we've been unable to confirm the presence of the carboxylated product by LC-ESI-MS (-ve). ESI- infusion does indicate the desired product, but we have no quantification despite a tidy LC run. The LCMS was (unfortunately) run with TFA, and then afterwards in it's absence, but to no avail.

The starter and product are trifluoromethyl substituted. Is this group particularly amenable to ionisation in APCI-?

[MG]

While TFA can be used sparingly in positive ESI (if you can live with some suppression), it will absolutely kill your negative ESI sensitivity. It may be that your LC system was still contaminated with TFA, and that is why you could see your product by infusion but not LC-MS. Run a blank and scan down to ~ 100 m/z (negative ESI). If you see an intense ion at 113 m/z during the run, your system is contaminated. In that case, flush your system some more and consider using a column that has not seen TFA. Incidentally, TFA will also suppress negative APCI, but not positive APCI, in my experience.

I don't have a good answer for your original question. The carboxylic acid should work by negative APCI simply because it's an acid. There was a paper published on electron capture APCI of pentafluorobenzyl compounds (Anal. Chem. 2000, 72, 3007-3013), and another lengthy discussion of negative APCI can be found in J. Am. Soc. Mass Spectrom. 2001, 12, 385-398. I have seen all manner of strange ions form in APCI other than protonation/deprotonation, to the point where I would not want to use it to identify unknowns. If you try APCI, I would be interested to hear how your nitrile compound works.

[JA]

Thanks for your reply MG. I was hoping to hear from you

The chemist and I were investigating precedents for the conversion and it turns out the process can be complicated by the hydrolysis of the trifluoromethyl group and also another halide functionality which I hadn't mentioned. Other conditions (more frequently used, it seems) are being looked at and we have fingers crossed for a more selective reaction.

Another paper, put online by the Japan Society for Analytical Chemistry, discuss the use of nitro-trifluoromethylphenyl derivatisation of oxosteriods (carbonyl containing) and detection by APCI-. As the product we want is basically a trifluoromethylphenyl species, electron capture APCI is probably something work putting some time into. Additionally, I'm trying to convince the Chemist that perhaps IR & C-13 NMR will compliment the MS.

[Broesen]

Yes, TFA definitely kill all the negative signal in APCI and APPI. So You have to find another buffer like a basic one.

Peter