The first thing to do is to write down exactly what you want to match. From your examples, it looks to me like this is a good start:
a non-letter
f
a
c
e
optional apostrophe
optional s
a non-letter
Translated into a regex, this turns into
[^a-zA-Z]face\'?s?[^a-zA-Z]
You then need to think about refinements. What about
FACE Face faCE
and so on? What if
face
is at the very beginning or end of the string you're scanning?
Then, what do you want to do with the results? Just count matches, list them in context, substitute for them, ...? For instance, my example includes (captures) the "before and after" non-letters, which may not be what you want if you're substituting. (It's easy to use non-capturing groups for these - just replace
[^a-zA-Z]
by
(?:[^a-zA-Z])
in two places.)
If you're going to do anything serious with regexes, I strongly recommend downloading a copy of Expresso. It's free and brilliant.
Cheers,
Peter
[edit] fixed typos [/edit]