In silica gel column chromatography, the order of elution of compounds depends on their relative affinities to the stationary phase (silica gel) and mobile phase (solvent). The order of elution is usually determined by the polarity of the compounds and the polarity of the solvent system used.

The picture in TLC and the real situation on the column re often different because the quality of SiO2 on plates and on column different. Try to start to elution with the very non-polar solvent on which all the products will be on the start in TLC. And then increase the polarity of the solvent gradually

It is likely that the selectivity of the TLC plate and the column silica were different. Things that affect selectivity include moisture and the pH of the silica.

See: teledyneisco.com/chromatography/blog/improve-flash-chromatography-method-development-with-matching-tlc-plates

Also, make sure you are running your TLC plates properly:

teledyneisco.com/chromatography/blog/back-to-basics-how-to-run-tlc-plates-properly

As I work for Teledyne, I avoid the issue by using the flash focus gradient generator because the method development is done with a column: Flash Method Development in a Flash (teledyneisco.com)

I expect that the primary mistake is that in a mixed solvent TLC development that the liquid wicking up the plate is the same composition as the liquid in the bottom of the tank. The composition of the solvent in the silica gel will be impacted by the rate of evaporation of the solvent from the plate, which will be different for the different solvents, leaving the solvent in the silica gel different than the liquid in the bottom of the tank.

John Canham you are correct, although I would rephrase the mistake as the user not allowing the TLC chamber to saturate with vapor before running the TLC plate.

See: Article Let Us Teach Proper Thin Layer Chromatography Technique!

@Jack Silver

@John Canhm

@Kirill I. Petko

@Hayder T Qaddoori

Thank you so much for your answers