Chromatography Forum

I´m trying to analyze Clindamycin palmitate hydrochloride in an oral powder by rp-hplc. My problem is that it has a lot of tailing. I´ve tried lots of methods and always the same. Tried with C18, C8, C6, TMS and CN, different pHs but couldn´t find a solution. It retains a lot. Does anyone know a good method to try? Perhaps Normal Phase or HILIC?

Most of the Google links to Clindamycin palmitate HPLC are over 30 years old and use ion-pair with RI detection. What kind of detector were you using and is there a chance that you are overloading the column to try to get a detectable peak?

Thanks for your reply Tom.
I´m injecting 10-20 µL. I suspect column is not overloaded. I´m using DAD.

Recently I was trying to react metabisulfite with benzaldehyde to form an aldehyde-bisulfite adduct. I can see benzaldehyde, but not the adduct. And I think reaction is happening because benzaldehyde peak gets lower. Perhaps I cannot see adduct by UV. Fluorescence? Which lambda emm and exc?

I´m injecting 10-20 µL. I suspect column is not overloaded. I´m using DAD.

As far as overload is concerned, what matters is not the injection volume so much as the mass injected.

Saying you used a DAD doesn't tell much. What wavelength. If you look at the structure of clindamycin, there is no much in the way of a UV chromophore, which means that you will probably be limited to the "end absorbance" of carbonyls down around 210 nm -- and probably won't have a very high molar absorbance even there.

All of which makes me suspect that if you are injecting enough material to see a peak, you may well be overloading your column.

If this were my problem, the first thing I would do is to obtain a UV spectrum of clindamycin in something approximating your mobile phase. That will let you establish the absorbance maximum and at least estimate the molar absorbance. Then go and do a loading study -- inject your standard, then dilute by a factor of 5 or so and reinject. If the peak shape improves, then you were overloading. If it doesn't improve, that will confirm that overload is not an issue.