I'm a chemical engineer, and part of my job is to put together physical property packages for our simulator. I have no idea if you can legally do this by adapting something from another database, but I would contribute some of the information I have if you can get through the legal stuff.
Mainly, I rely on NIST's JANAF publication for thermodynamic data (inorganics only), the DIPPR database, and the Yaw's Physical Property Handbook. For binary VLE data DECHEMA is the best references Of these, JANAF is freely available (in horrible PDF format) while the other two would require some sort of membership or purchase. These are mainly in the form of equations or data tables. From there, we simply bring it into excel or sometimes some curve refitting routines and then on to the simulator, which takes the equations directly. We're members of AIChE which gives us access to the online version of these databases.
DIPPR was the result of a major effort of a consortium of companies who needed this data. I believe most of the serious big petrochemical companies are members of DIPPR. You will be trying to recreate this work, which is doable, but I think it will be harder than you think. I agree that this data shouldn't be locked away as tightly as it is.