I have used dichloromethane : methanol = 90 : 10 but the caffeine got eluted with the mobile phase. Both the chemicals used were of HPLC grade (Merck).
You didn't mention which type of stationary phase You are using in TLC plates.
If You are using non modified silica the dcm:MeOH is not polar enough and the caffeine should stay on the start point, so I guess that You are using C18 modified silica HPTLC and for that You need to use polar solvent for this separation like water:MeOH 99:1. Dcm:MeOH on C18 TLC have much more elution strength than pure ACN.
Adam Zmysłowski Thank you for the answer. I have used this HPTLC plate - 20 cm × 10 cm aluminium-backed HPTLC plates coated with 200 µm thick layers of silica gel 60 F254 (E. Merck, Darmstadt, Germany). Can you please give more information regarding to this?
So the TLC plates are standards silica plates and for that You need to go to more polar solvents like ethyl acetate as Soleya has written or even more polar like isopropanol, ethanol or methanol. However I would be careful using modifier like formic acid because it could cause caffeine molecule to be protonated and for that more polar and could cause tailing effect.
Method development should first retain all of the compounds of interest, resolve them apart, then elute them (that last step is not needed in TLC). When developing any type of chromatography method it is necessary to answer one key question.
Adam Zmysłowski William Letter Thank you for your precious answers. I will try those for my method. And William Letter I am seperating caffeine from green tea and normal tea.