There was a good Pensado's Place IN THE LAIR segment where he took a single vocal and faked a double. You end up taking a single part and automating the time and pitch to simulate a different take. He was using this to stack vocals, but it might be a little useful for making a mono source into stereo.
I'm actually rendering a live recording right now and slapped a reverb with no time or predelay onto a mono ambient track, which effectively made it a stereo track. Sounds good. I'm kind of rushed right now, but if I find time, I'll post a quick clip of before and after. Also, I don't see this working that great for a guitar track that you need super intelligible, but it also might work.