I'll disagree with Klav slightly. It may be perfectly legal to use "may" in claims, but I recommend against it. I don't even like "can". I prefer to recite that the thing responds to stimulus/i to do a certain thing -- the thing that others like to recite a thing "may" or "can" do. I prefer to really nail down in affirmative language that the thing performs the function when so triggered and, if appropriate, what triggers the performance of the function by the thing.
Good point Jim, though I think our differing opinions comes from the fact that you and I typically practice in different technological areas. In the mechanical/CS art area, you are of course correct that it is better to use positive language that recites how a particular component functions in response to a particular event or stimulus. In the chemical arts, however, optional limitations are frequently used in order to ensure the broadest possible coverage for a particular genus.
For example, a client may have discovered a class of a new chemical entitities comprising a phenyl ring. In such cases, it is common for a client to request language that covers not only the unsubstituted phenyl, but all possible substituents on the phenyl. Ergo, an optional limitation in the base claim is almost necessary, i.e., "A composition comprising a compound of formula 1, wherein the phenyl ring in formula 1 is optionally substituted with a [insert substitutents here]. . ." Of course, if the client doesn't mind havign an enormoous number of claims, you could claim compounds of formula 1 more specifically in a number of distinct claims, but that will just set the stage for a restriction/species election.
Best,
Klav