The home is in SW Ohio. The house is an 125+ year old solid brick house.

We are converting an attic/storage area to living space. The area has a cathedral ceiling as open to peak inside. I am looking for comments on insulating the roof underside; possibly using closed cell foam and a radiant barrier.

If foam sprayed on bottom of roof deck and then foil covered foam board fastened to bottom of rafters which would then be followed by drywall or other ceiling covering, would this work? The questions I have are will there need to be an air space between the foam and the foil barrier, or does the foam board with foil on both sides not need air space above? What are the downsides of allowing the mass of foam to be heated by the radiant heat through the roof deck?

What other suggestions or considerations?

Putting foil anywhere in the stackup would eliminate the ability of the roof deck to dry toward the interior, putting it at higher risk for roof-rot.

The roof deck is fairly well protected from taking on wintertime moisture if just 1-2" of cc foam is used, followed by a full rafter-fill of cellulose or dense-packed fiberglass. You can then use a semi-permeable foam such as unfaced EPS or up to 1" of XPS between the gypsum and rafters.

Mind you, this would not meet the IRC code, even though there's some pretty- good science indicating that it will work, and work well:

http://www.buildingscience.com/documents/reports/rr-1001-moisture-safe-unvented-wood-roof-systems

The IRC requires at least R15-R20 (depending on whether you're in US climate zone 4 or 5) of closed cell foam under the roof deck, which would also inhibit the drying capacity toward the interior. If you go the full code-complaint route using 3-4" of Icycnene MD-R-200 water-blown 2lb foam would still offer sufficient drying capacity without risk of winter moisture load-up.

It's unlikely that the roof deck will exceed the operating temp of any closed cell polyurethane. BTW: There is no radiated heat transfer between the roof deck and the adhered foam- it's all conducted heat. Heat transfer via radiation requires a air (or other gas, or vaccuum) gap between the radiating surface and heated surface. Any time they are in contact the vast majority of the heat transfer is conducted. There is absolutely no rationale for using radiant barriers anywhere in a cathedral ceiling stackup.

See also:

http://www.greenbuildingadvisor.com/blogs/dept/musings/how-build-insulated-cathedral-ceiling