You should be able to separate the materials using fractional crystallisation. Which solvent were the crystals obtained from? This solvent would likely be a good choice for recrystallisation, perhaps with a co/anti solvent. Are you able to tell us anything about the structure of these compounds? Such information is useful in choosing a solvent for recrystallisation.

Pasteur's manual method of separation (1849) ;-)

If crystals are different componds by TLC, you can use NMR for structure determination.

Please note that ordinary TLC can not differentiate  some mixes. For instance : mix of racemate (binary) compound crystals and single stereoisomer crystals. So it better to use chiral phase - sorbent or solvent.

Huq,

your question is

"now from here how can i separate them?"

You use TLC to adjust the eluent that separate both compounds from each other as Gollapudi mentioned then use column chromatography or plate chromatography to separate mgs of your mixture using the eluent that separate them most.

Good luck

 Thanks to all for your valuable suggestions. By the way, the crystals i have are not soluble in small volume of MeOH even though i collected them from MeOH elution. but if higher volume is added then soluble. but in that case i can not run TLC as it is very diluted and can not see any spot on TLC.

There are several separation techniques out there. For an insoluble crystal, you may try Gas Chromatography (GC). However, you will get to know the constituents of your crystals but initial crystals used are destroyed in the process. HPLC is a better method if your crystal is soluble in any liquid medium. I hope this helps.

Have you tried recrystallization already?

Also, as Clifford mentioned earlier, if HPLC works out well for you, then you could determine suitable solvent conditions for P-HPLC. However, it is always useful to consider processes that result in less material loss. You mentioned that the crystals are soluble in MeOH only after adding copious volumes of solvent. How about considering dissolving them in a mixture of solvents (such as MeOH and DCM or MeoH and H2O)? Once that can dissolve your material, then you could try separating the sample on Sephadex LH-20. At least you are guaranteed that you would have little to no material loss.

If that doesn't work, the sample (depending on weight of material) could be put on a small ODS column and eluted with H20/MeOH (gradient). Consider these steps first before moving on to HPLC and subsequently to P-HPLC.