Do you really need to get it all out or do you merely have to know how much is there? In your original post, it's not clear. On the one hand, you mention that you need to know how much limonene is in the plastic but on the other hand you mention that you're alarmed that subsequent injections of the same sample shows a lot of limonene in the vapor.
Static headspace relies on an equilibrium (dictated by the partition coefficient, a thermodynamic parameter that is constant for a given set of sampling conditions):
Kd = Cg/Cs
of the analyte between the condensed phase and the vapor phase. Cg = conc analyte in the vapor phase, Cs = conc analyte in the condensed phase. It doesn't mean that all of the material leaves the condensed phase and moves into the vapor phase.
You don't need to have it all out of there to know how much is in there.