Yes,
several spring to mind.
First, you have lost fine-grained control over what is and is not encrypted. It's volume based, instead of tablespace based.
Second, you have introduced an unwanted mode of failure. If you use oracle's built in facilities, Oracle naturally certifies and guarantees its software, and will respond to a call for assistance on metalink.
If you reveal that third-party software is in the mix, Oracle may well turn round and say "Sorry, we can't help, because as far as we can see, the problem is not within Oracle." You then place a call to the encryption provider, and they are equally certain that their software is in order, and have you raised a call on metalink? After three rounds of telephone table tennis, you will be the one bouncing up and down a lot.
Oracle uses ASSM and LMT by default. If you are using OMF and/or ASM to boot, how is that interfacing correctly to your encrypted volume? Do you know that all of this technology works because of a successful test, or are you just hoping?
If a tablespace uses many DBF's, and the encrypted volume reaches capacity, will you be able to just use a convenient unencrypted disk, whilst another volume is prepared, or will undesirable side effects occur?
If due to space requirements you have several volumes, how will the db manage all its keys, so that it can do its job properly, and open these several volumes?
How are you guaranteeing that your back-ups are encrypted. If you just use RMAN and rely on the encrypted volume approach, then you will run out of space fairly smartly. If you (quite sensibly) move old archive logs from disk to tape, how will you restore them later, when an encrypted volume is full, and you have nowhere to which the tapes can be loaded. This of course assumes that the tapes can fully interoperate with decrypted and/or encrypted volumes.
I may be wrong in some of my assertions, but really I can summarise what I'm trying to say in one golden oldie
"KISS - keep it stupidly simple".
Regards
T