keto-enol tautomerism

[bert]

Dear colleagues,

we are working on a component, which is subject to keto-enol tautomerism. Unfortunately, the LC-MS method we are using shows two peaks. We noticed on a UV system that at higher salt concentrations (>1 M) we obtain a single peak. I know that the pH can affect keto-enol tautomerism, but I am not familiar with the effect of salt concentration. Is anyone of you? Do you have any idea how to translate this to LC-MS.

Thanks for your help.

Regards, Bert

[HW Mueller]

Sure would like to know what kind of compound you have, as the equilibrium is normally so fast one can´t separate them. My guess would be that salts would influence the eqhilibrium position via ionic strength. Also, one can imagine that some cations might complex with the enol, stabilizing it...that´s a long shot. I am inclined to think that the two peaks are some sort of artefact, both peaks have the same m/e? Can you run this on UV with the MS mobile phase?

[james little]

We see this a lot by GCMS where acetoacetylated compounds show two peaks. Usually between the two peaks, the baseline does not return to zero indicating the two peaks are "talking" to each other as they travel down the column.

In this case, we usually react with BSTFA to form an O-trimethysilyated derivative. This usually gives one major well-defined peak.

You can also react the ketone in this compound with hydrolylamine to form a derivative and then silylate. In this case, get two peaks due to syn and anti forms.

I have also seen where people react compounds with two carbonyls separated by a methylene group (get keto-enol problems) with 1,2-diaminobenzene to form a bis imine derivative.

[bert]

Sailor,

thanks for your response. I still hope we can solve the problem without labour intensive derivatization.......

Hans,

the compound is a tetracycline antibiotic. Indeed, both peaks have the same m/z. For the uv experiments, we started with the same column and mobile phase as the MS analysis. The chromatographic pattern was exactly the same as with MS detection: one peak comes of and, like Sailor noticed, the baseline does not return to zero. Than te second peak is eluting. We call this a 'camel saddle'. Increasing the salt concentration (on the uv system) lets the first peak merge with the second one. The final peak at a buffer concentration of 1 M is quite nice. The strange thing is that a pH shift did not really improve chromatography, although we did those experiments without control of the ionic strength. In case of keto-enol tautomerism, I would expect effects of changing the pH.

Regards Bert

[HW Mueller]

One never stops learning, too bad you guys are not allowed to give structures... Nice question for a cumulative exam.
I would suspect that acid pH would speed the equilibrium (peaks would merge into one) and basic pH could possibly stabilize the enol?

[bert]

I'm glad I don't have to join that exam

In fact, at low pH, the peaks are best seperated. At higher pH values, peaks merge, but still their is a big shoulder in front.

For the molarity experiment, we used an acetate buffer. Any suggestions how to translate this to MS would be very welcome (complexing cations to try?)!

Regards and thanks,

Bert

[Uwe Neue]

How high did you go in pH? Sugars give two peaks with a shoulder at acidic and neutral pH, but just one broad peak at pH 9.

The other trick to get to one peak is to increase the temperature. This speeds things up, and you get one peak. Of course this does not work if your peaks are a mile apart at room temperature. You need to see some element of conversion.

Maybe a combination of both effects - pH and temperature - can get you to where you want to be.

[bert]

Uwe,

we did go up to pH 11.5 and tried temperatures up to 50 degrees C. We did not try the combination, so thats the next thing we will try.

Thanks.

Regards Bert

[HW Mueller]

What was that 1M salt? What were all the cations present?
Also, I wonder whether anybody has evidence (nmr, etc.) that the isomers which separate in HPLC are present in solution as well, at the same ratio?

[Ricardinio]

keto-enol tautomerism of tetracyclines has been observed for chlortetracycline and doxycycline, but to my knowledge not for other tetracyclines. From what would result this difference?

[HW Mueller]

Ricardinio,
not seeing the structure one can only surmise that the usual equilibrium energetics hold here also.

[bert]

Hans,

the MS mobile phase was a mixture in a ratio of 750 ml of water, 250 ml of acetonitrile and 1.00 ml of formic acid. Per litre of this mixture, 37 mg of Titriplex III is solved.
For the UV experiments to study the effect of salt concentration, we prepared a 1 M buffer by mixing a 1 M sodium acetate solution with a 1 M acetic acid solution until the desired pH was reached.

Ricardinio, do you have a reference, describing tautomerism for CTC and DC?

Thanks.

Regards Bert

[HW Mueller]

bert,
the acetate was used in place of the water? The formic acid was still there? Anyway, it seems that you either had a different pH with the acetate present or some artefacts are in effect. But to be more certain about an ion strength effect why don´t you use a neutral salt instead of HOAc/HOAc?

[bert]

Hans, indeed, we replaced the water by the acetate buffer. FA was still present. To study the ionic strength effect, we changed the molarity of the buffer. In my opinion a sound way to study this effect?

regards Bert

[HW Mueller]

Remember that buffering capacity goes up with the concentration, probably you had a different final "pH". Also note that if you change the concentration of a buffer you shift the species around a bit.