Chromatography Forum

hello everybody,

I have some problems (since couple of weeks huh ) with separation of mixture consisting mainly of naproxen (plus standard contaminations of naproxen), p-hydroxybenzoic acid and p-hydroxybenzoate (ethyl and isopropyl); I have tried method according to pharmacopeia, but it failed,

I have tried usual C18 (Varian), ACE column, Zorbax SB (25 cm, 5 u; 4,6)

I have tried different combinations of solutions (mobile phase was mainly A: 1% acetic acid, B: acetonitrile or methanol or acetonitrile plus methanol or phosphate buffer or even combination ACN:MeOH:THF); I have tried isocratic and gradient methods, but they failed;

if peaks were all separated, some of them were tailing or leading;
if peaks were all right, there was no separation or separation was not complete;

maybe somebody can help me?


Krysia

Leading peaks are typically caused by the fact that the sample is dissolved in a stronger eluting solvent than the mobile phase. I would go back to the condition where all peaks were separated and fix the injection. BTW, what were the mobile phases that worked? It also could be that your buffer was not strong enough, and that this was the reason for the distorted peaks.

the best mobilie phase was Acetic acid 1% (A) and ACN:MeOH (1:1 or 2:1) (B); there was gradient of course;

and I had good separation, but tailing (leading) mainly hydroxybenzoic acid and one of the benzoeate ester was observed in that case;
for sample preparation I used methanol or mobile phase and there were tailing problems too, unfortunately; (20 ul injection)

phosphate buffer as part of mobilie phase was completely disaster

moreover I had some problems with that buffer and my HPLC pump (VARIAN forever ) some time ago (other product) so I am trying to avoid it

My bet is that the solvent in which you dissolved the sample had a higher elution strength than the mobile phase at the point of elution. This was definitely true for the case when you dissolved the sample in the organic solvent. It is not clear, where your gradient starts. When I said that you need to dissolve the sample in mobile phase, I meant in the composition at the beginning of the gradient or a still weaker composition.

Another reason for fronting can be the attempt to work at elevated temperature without preheating the solvent.

A proper RP column is most likely sufficient for this application. Uwe's suggestions should fix most problems if not all. In case the result is still not satisfactory, I would recommend you consider one of mixed-mode columns out there. Attached is some information on Acclaim Mixed-Mode WAX-1 column that might interest you. I found this column is often quite useful for separating a mixture of neutral and charged analyets especially when a RP column fails to do so.
http://www1.dionex.com/en-us/columns_ac ... 48519.html