To get the best response out of the people who monitor this board, you really need to provide as much information about your specific problem as you can. Many times, someone has already solved it and they are more than willing to tell you how to do it.
You might have a prayer with that one
http://www.sigmaaldrich.com/catalog/pro ... ®ion=US
using a SPME headspace procedure. You'll probably have to heat the sample to a high temperature. I'm afraid it's probably too heavy and too high of a boiler for static headspace. What is your matrix?