Collect the peak, separate it on either TLC and collect two spots separately in solvent. If you want to separate it on HPLC slow down the speed of HPLC detection and slow rate of separation differ in retention time and you will get two separate peaks on the same column. 

Hi

There are many solutions. The first one concern isolation of each product. You are doing reverse phase Chromatography. Probably with the same apparatus you can try normal phase chromatography with an other column. You can also try column chromatography. If you want to use silica for this separation, try first TLC with various combination of solvent. You can also use various kind of column (with HPLC or column chromatography with different polarity of stationary phase, exclusion chromatography, etc…)

The second option could be to use an other identification method. You can identify by NMR (mainly carbon NMR) each compound of the mixture, even if they are stereoisomers. If they are known compound it would be much easier.

I hope it well help.

I note you are using C18. Things you can try include:

In all cases, we are changing conditions to cause the compounds to possibly interact differently with the column or solvent and hopefully resolve them. 

I agree with jack Silver and with Sylvain Surtour: if you can do NMR spectra of your unknown compounds the selection of a different HPLC separation method will be easier: there is a plenty of stationary phases that you can use. If  an apolar mobile phase is not good for your compounds you can also try a reverse phase separation with a silica-CN, which can be use both in normal and reverse phase. If your compounds are isomers you can try to form inclusion compounds with a cyclodextrin stationary phase. 

Thanks to all of you for valuable suggestions. We are recently using TLC with different solvent systems. Let see if it works.

why not we use DOSY technique (Diffusion Order Spectroscopy) to identify the compounds of mixture. Our samples are not very complex?

As suggested by Chiara, silica tips of C4 or C8 can also be tried for quick result. 

Hi, 

an other solution is using centrifugal partition chromatography, which can help to separate co-eluting peaks in HPLC. Just for an example, see this paper: 

Preparative isolation of closely-related xanthones from Gentianella amarella ssp acuta by high-speed countercurrent chromatography

Urbain, A.;Marston, A.;Batsuren, D.;Purev, O. and Hostettmann, K.

Phytochem Anal 2008, 19, 514-519

Hope it helps, 

Veronika

Hi,

If the above suggestions don't work, you can also try to use a chiral phase column with HPLC.

Hope it helps.

Jean-Pierre

try gradient flow

I think you can get a quick results using a dimensional TLC of course after you collect your peak.

If you can, use UHPLC. It works at higher pressures and is much better at separating compounds than HPLC.

I agree with Jack Silver idea. The reversed-phase C18 can go so far.  You use it to separate your anlayte from very polar and very non-polar compounds (before and after your peak). It is possible that your interference may have the same polarity index as your anbalyte. Now you have to use different mode of separation e.g. Normal phase (bonded one such as diol or CN) , if you think your analyte has ionic property, you may need ion-exchange or use HILIC. Since you have no idea about your compound structure, you need to use trial-error-method.

Dear,

Thanks to all of your for valuable suggestions. Using reverse phase HPLC first peak of unknown compounds elutes after five minutes and second peak elutes after 28 minutes using gradient flow of ACN:water with 0.1% trifluoroacetic acid.

We will try to use normal phase HPLC and TLC methods.

I agree with Olga: try silica gel  mono and two directional silica gel TLC or HPTLC after collecting the HPLC peak, evaporate/freeze dry the sample and re-dissolve in methanol    

The reversed-phase C18 can go so far.but we  need establish standard curves