Determination of unstable impurity

[Mattias]

I am seeing an impurity that is increasing for every injection, i.e. it is a degradation product of the main peak (it is not in the blank).

I have tried to inject it directly after I have dissolved the substance (within 1 min), and I can see the impurity clearly. Therefore I suspect that the impurity is present also in the solid material.

I have tried different pH values (2, 4 and 7) and the impurity is there anyhow. Has anyone a good idea how the measure how much impurity there is in the substance?

I have heard that you can do MS directly on the solid material, but I have no idea how. The molecular weight of the main peak is about 200.

[Rob Burgess]

If this is a pharma product you are looking and you are analysing by UV, then you can quote % area of the impurity relative to the main peak. You make the assumption that the response factor for your impurity is equal to the response factor for your main peak (not very elegant or accurate but is ok for a 1st approximation).

I'm not sure how MS would help in this case... it may give you an idea of a mol. wt. to help dtermine what the unknown impurity is, but again you are left with a unknown reponse factor scenario.

[Serpico]

Not sure I fully understand the issues but I'll have a stab...

It seems there is a link between the degradation impurity quantity and increasing length of time in the sample diluent. If, as it seems, the impurity is directly related to the degradation caused by the sample diluent then why do you suspect it to be in your solid material?

Have you tried a wide range of solvents as your sample diluent? Does the quantity alter with different solvents? Do different impurities form with different diluent? Does it stop forming after a certain period of time?

As an aside, its certainly possible to analyse solids by mass spec, but you need the kit. You can get Direct Insertion Probes (DIP) or Direct Exposure Probes (DEP) for GCMSs or go down the LCMS route with Desorption Electrospray Ionization (DESI). Might tell you if and what impurity is present but quantification wouldn't really be practical.

Hope this helps.

[Ary]

Have you tried dissolving your sample in different solvents as well as changing the pH? Your compound may be reacting with the solvent.

If this is not a possibility repeat injections over 1-2 hours measuring the amount of the impurity and parent should give you sufficient data to extrapolate back to how much is present at time zero.

Direct probe MS may be able to tell you if there is any of the impurity present in the solid material but wont be quantitative

Apologies for giving much the same info as Serpico his/her answer appeared while I was typing my response

[Mattias]

Thanks for your suggestions.

I have now tried to dissolve the substance in organic solvents, and the impurity seems to be stable in waterfree MeOH. I was able to dissolve a lot of substance so I could get away with only 1 µl injection (so the chromatography was not affected).

I think this must be the evidence that the impuity is not only a degradation product in the sample solution, but it is also present in the substance it self.

I have also fraction collected the impurity now, and it is unstable in the mobile phase (t½ = 3 hours). It will surely be a challenge to prepare an NMR sample...

[Ary]

NMR will depend on what else your compound is soluble in. You can buy ampoules of DMSO which are water free so you could use this to make up and run your sample. Or better still if your sample is soluble in chloroform you could use that.

[yogesh#14]

Hi,

You can try the following options.

1) Try a LCMS of the sample to determine th Mol. wt of the Impurity.

2) If you have LC -NMR inject it to get the NMR pattern.

3) Is the PDA spectrum of the impurity similar to the main peak or different?

4) you can report the % of the impurity by area normalization wrt main peak , as of now...

5) Have a look at the structure of the main compound and if their is a possibility to degrade the compound to get the same impurity try it...

6) Try a different diluting solvent for your sample like acetonitrile....

Hope this will help you...

regards,

Yogesh

[Mattias]

Dear Yogesh and Ary,

I will do LC/MS on this peak, but I have to wait until it is free (very occupied instrument)

We do not have an LC-NMR equipment, but I could ask the University to help me.

The UV-spectrum is very different from the main peak, with many different UV-max (highest 490 nm). Aromatic? I see no color of the solution, although the peak is around 5-7% (at 220 nm).

If I isolate the peak with fraction collection, it will inevitably be solved in mobile phase and degrade quickly.

[Ary]

Hi Mattias

If you have plenty of sample you could try and isolate some material by injecting as much as possible onto your column. I understand the impurity appears to be unstable in the mobile phase so you could then evaporate off the solvent either under a stream of gas, or in a freeze drier or you could attempt a liquid/liquid extraction into another solvent. It sounds as though your impurity may be significantly different in nature(though this is a big assumption based on UV profile data alone) to the parent so you may even be able to try extracting the impurity dirctly from your solid sample. There would then be no need for LC-MS or LC-NMR. A simple NMR may give you the answer you need.