Oh right... so this is the situation, now it all seems more clear (to me at least
)
Anyway I don't think Lee's explanation of the phenomenon is very clear as we should consider just one small thing before talking of hybridization: it's a mathematical way to explain atomic phenomena and moreover it's not even the clearest way or the one which accords better to reality. Today chemists work mainly using MO theory that, while being less accessible that VB for the students it describes better the extremely difficult behaviour of the electrons in the atom, especially in the form of a strong ionic bond like the one you talk about.
In fact PCl
5 is one of the molecules that show the limits of this sort of description: as you pointed out you can't understand why these orbitals contract more than the others. This is an explanation found out to justify the fact that, with their original energies the orbitals wouldn't hybridize so the largest orbital needs to decrease its size (and energy) to get in touch with the others. I know i may sound a bit boring repeating again this but the fact that P is bonded to 5 Cl atoms and that the whole molecule has a bipyramidal structure makes us think that the only way (according to VB) to realize that geometry is to have a sp
3d hybridization. We could say (obviously I'm only trying to explain this fact better) that this is a sort of "stratagem" to interpretate reality (which is as usual our touchstone of theoretical approach). Don't be disappointed but you seem to stumble always upon this sort of matters (
) and this is quite interesting because remembers me my way to study chemistry, trying to understand deeply the phenomenon, independently from what the book wants to prove, trying to understand "why that happens" and not only to convince me "that happens".
If you need any more help feel free to ask again... but let me tell you just another example of how VB fails to describe even the most common properties of molecules. Take O
2: to explain its paramagnetic properties scientists hypothesized a 3-electron bond (and this is really strange for O) which was only a way to fit a property which didn't accord to the theory introducting a new kind of bond (a stratagem, as I said). MO theory doesn't need to do this: its Aufbau of the O
2 molecule shows that there are two unpaired electrons (disposed according to Hund principle which is still applied) in the external orbitals which give O
2 its characteristics.
My only advice is that you try to read something (and if you've already studied it try to deepen your knowledge of this field) about MO, I found it just great and even it is quite different in its approach you'll find it's just great the way how the molecules are described: organic and comprehensive (obviously it isn't perfect, new kinds of bonds need to be found out to justify some structures such as the diborane of the HF
2- which I already cited but it's the best we have for now
)