Well, we know phosphorous pentoxide is a dehydrating agent. The silica commonly used for chromatography contains about 10-15% water by weight (as I recall). I would think that if one wanted to perform some kind of dehydration reaction, one would first drive the water from the silica before mixing with the P2O5. I doubt the fluorescing agent or binder in the F254 gel had a negative effect.