The term protecting group is not absolutely strictly defined. In general, anything can be a protecting group if it fulfills the following conditions:
- it is attached somewhat easily and adequatly selectively to certain specific sites
- it is inert towards a sequence of following reactions which would affect the unprotected site; and
- it is somewhat easily and adequately selectively removed from those sites to reliberate the original functional groups.
However, the term is sometimes used in an even broader sense meaning when ignoring the first bullet point.
Which functional groups you want to protect and which protecting groups you wish to use for the job depends entirely on your desired sequence. In natural product synthesis, it is very common to transform alcohols into silyl ethers — I would want to argue that this is the most common usage of the term. Alcohols have the disadvantages that they are nucleophilic, weakly acidic, can coordinate metal centres and can even act as bases. If you transform the alcohol into a silyl ether, it is no longer acidic, its ligand properties are substantially impeded and it can no longer attack nucleophilicly. Thus, it is possible to perform reactions, e.g. titanium(IV) catalysed aldol additions, which would not be possible with free alcohols.
Sometimes it is also desireable to introduce different protecting groups to similar sites or to selectively cleave only some protecting groups. For example, you may have two primary alcohols and want to turn only one of them into an aldehyde; by clever choice of your reaction sequence and protecting group strategy you will arrive at a point where only one protecting group is easily removed to give only one of the two free alcohols selectively.
While silyl groups — among them the relatively bulky tert-butyldimethylsilyl (TBS), tert-butyldiphenylsilyl (TBDPS) and tetraisopropylsilyl (TIPS) groups are exceedingly common, not all protecting groups need be that large. Methoxymethyl (MOM) is a rather small protecting group that enables the alcohol (then acetalic oxygen) to keep its ligand properties. Even smaller is the methyl group used to protect carboxyl functions or, arguably the smallest, the chlorine atom, used by Woodward in his synthesis of 6-demethyl-6-deoxytetracycline.
: J. J. Korst, J. D. Johnston, K. Butler, E. J. Bianco, L. H. Conover, R. B. Woodward, J. Am. Chem. Soc. 1968, 90, 439. DOI: 10.1021/ja01004a041.