Ok, so the first one is ring flipping of the amine leading to rapid interconversion of the two enantiomers of that amine. So chral TLC shows one spot for the amine, because the molecules are interconverting rapidly compared to the timescale of your TLC, so you see one spot which is effectively an average of the two enantiomers.
The phosphine doesn't racemise rapidly, so you can resolve them as this interconversion is slow relative to the timescale of TLC - 2 spots.
The reaction with MeI produces the ammonium/phosphonium salts which do not interconvert readily.
The next one is about flipping again. The amine will have scrambled stereochemistry at the N centre, and so the cyclisation gives two possible diastereoisomers which can be separated on standard TLC plates.
The stereochemistry at the P centre in the phosphine will not scramble quickly, so only one product is formed.
Last question. Identify all the possible structures of the products. I count 2 for the first, 4 for the second. Remember that you will not resolve enantiomers on silica, only (in the context of this question)diastereoisomers.
For relative intensities think about the Felkin-Ahn model.
This one's alot of work for 4 marks...