[Jaz]

Hi,

I am trying to derivatize 5-hydroxypyrazinoic acid and pyrazinoic acid. It only works for pyrazinoic acid though. The derivatization method I am using is simply methanol and acid. The only structural differance between the two molecules is the hydroxy group and I can not figure out why the hydroxy group would antagonize the derivatization process. Is there anyone with knowledge in this matter?

Regards

[Arkcon]

Well, what's the acid? Just mineral acid?  What actually is the derivitization reaction?

Note: I've moved the question here from the Analytical Chemistry sub-forum, so this question can get better help. But in practical terms, it might also help to know what the analytical method is, and how it detects the derivitized molecule.

[Jaz]

The acid used is HCl. The reaction is an ester formation of the carboxylic acid with applied heat. There is nothing in the mixture but the acid, the methanol and the compound.

The analytical method used is LC-MS/MS, detection by mass that is.

[kriggy]

Try using stronger acid catalyst like H2SO4. Also, remember that those nitrogens can get protonated so maybe the reaction is not working because the catalytic amount of acid is consumed by the pyrazine

[Dan]

The acid used is HCl. The reaction is an ester formation of the carboxylic acid with applied heat. There is nothing in the mixture but the acid, the methanol and the compound.

The analytical method used is LC-MS/MS, detection by mass that is.

What is your source of HCl? How much are you using? Are you doing this water-free?

It should work, see example 62a in this patent where they use HCl(dioxane) with methanol.

You can generate dry methanolic HCl by adding SOCl₂, AcCl or TMSCl to methanol if you don't have a dry HCl source.

Conc. Sulfuric acid is another anhydrous option.

[Jaz]

The acid used is HCl. The reaction is an ester formation of the carboxylic acid with applied heat. There is nothing in the mixture but the acid, the methanol and the compound.

The analytical method used is LC-MS/MS, detection by mass that is.

What is your source of HCl? How much are you using? Are you doing this water-free?

It should work, see example 62a in this patent where they use HCl(dioxane) with methanol.

You can generate dry methanolic HCl by adding SOCl₂, AcCl or TMSCl to methanol if you don't have a dry HCl source.

Conc. Sulfuric acid is another anhydrous option.

Yes it should work. I have seen similar publications where they use this method in order to derivatize small carboxylic acids. I use 12 M HCl in the mixture with a final concentration of 0.1 M according to references. So it is not fully dry. I have tried both longer incubation time and higher concentration (up to 1.0 M) of HCl.

The question is though why it works with the pyrazinoic acid and not with the 5-hydroxy pyrazinoic acid?

[Jaz]

Try using stronger acid catalyst like H2SO4. Also, remember that those nitrogens can get protonated so maybe the reaction is not working because the catalytic amount of acid is consumed by the pyrazine

It could be part of the reason. I will see if I can try it with H2SO4. But how come it works for one of the compounds?

[kriggy]

Coluld be that you are getting some kind of dimer. What exactly is not working? No conversion at all, by products?..

[Dan]

I have seen similar publications where they use this method in order to derivatize small carboxylic acids. I use 12 M HCl in the mixture with a final concentration of 0.1 M according to references. So it is not fully dry. I have tried both longer incubation time and higher concentration (up to 1.0 M) of HCl.

Try it water free and see if that works (try a method reported for exactly the same molecule).

The question is though why it works with the pyrazinoic acid and not with the 5-hydroxy pyrazinoic acid?

Well, the addition of a hydroxyl group substantially changes the electronics of the system. It will increase N basicity and decrease carbonyl electrophilicity, for example. It's not surprising to see a difference in reactivity, but I still would have expected it to work under the conditions you describe. To speculate any further, you need to know what did happen (Kriggy's question), not just what didn't happen.

[Jaz]

The samples does look like blanks both underivatized and as an derivative. It would be expected not to be seen as an underivatized product.

It should be mentioned that this has not been done before. There is no method to measure these compounds with LC-MS.

I have been looking for dimers but I could not find anything like it in the samples. As mentioned, the samples look like blanks.

[kriggy]

Blanks you mean that you have no signal? Then try higher concentration of the sample. You dont need specific method for HPLC-MS unless you are looking for specific compound in very difficult matrix (blood samples for example). Just use classic gradient ACN:Water 20:80  80:20

It might be possible that your compounds are not visible in UV detector in HPLC so maybe try TLC with UV detection (that works fine for even simple aromatics)

[Dan]

It should be mentioned that this has not been done before. There is no method to measure these compounds with LC-MS.

Ah ha, so the problem may be the detection method, not the reaction itself.

TLC should give you an idea (bear in mind you will have to neutralise an aliquot before running the TLC otherwise it will probably not migrate), but NMR will give you a definitive answer. If you can isolate the compound you can then work on a detection method by LC-MS if necessary.

MS of the crude should also give you an indication of whether the methyl ester is present.

[Jaz]

Blanks you mean that you have no signal? Then try higher concentration of the sample. You dont need specific method for HPLC-MS unless you are looking for specific compound in very difficult matrix (blood samples for example). Just use classic gradient ACN:Water 20:80  80:20

It might be possible that your compounds are not visible in UV detector in HPLC so maybe try TLC with UV detection (that works fine for even simple aromatics)

Blanks mean samples without the compounds for background check. A control so to speak.

Right now I am running a standard gradient similar to the one that you described but it will be optimized later on to refine the method.

I do not use any UV detection at all. It is just the MS. I have seen publications on methods that uses UV detection (HPLC) for these compounds but this method will be run with several other compounds analysed by LC-MS.

[Jaz]

It should be mentioned that this has not been done before. There is no method to measure these compounds with LC-MS.

Ah ha, so the problem may be the detection method, not the reaction itself.

TLC should give you an idea (bear in mind you will have to neutralise an aliquot before running the TLC otherwise it will probably not migrate), but NMR will give you a definitive answer. If you can isolate the compound you can then work on a detection method by LC-MS if necessary.

MS of the crude should also give you an indication of whether the methyl ester is present.

What do you mean by crude? The underivatized compound?

We do not have any NMR or TLC equipment in our lab so unfortunatly that is not an option.

[kriggy]

Im sorry to ask but you dont have TLC plates in your lab? That seems very weird.
By using "just MS" you mean direct injection into MS or that you have HPLC with mass detection instead of UV?