Well.... here's something I just thought of - sorta. So I'm in a heat-mass transfer class (I'm a mech engineering student, yay classes!), and I'm studying for finals and one tidbit is heat capacity.
That got me thinking. We want to store a great deal of heat - not necessarily at a high temperature, just a lot of it. The thermal mass of the coolant, head, oil etc. is huge.
One great way to store heat is to take advantage of latent heat | heat of fusion -- which is heat/energy in solid-liquid phase changes. I need to do some research/calculations - but I suspect 3L of wax will hold more heat than 3L of coolant. I think paraffin wax (yes, I know - fire hazard if exposed) does the solid-liquid phase change around 50C. And I'm sure there's other suitable (albeit less accessible/more expensive) materials out there.
What's great about latent heat storage is that there's A LOT of it. And once you begin the phase change, the average temperature stays constant until the change completes. Now I just need to think of a way to implement - probably VIA some sort of "bucket" heat sink/ heat exchanger.
If anyone here subscribes to Autospeed (a great Aussie motor mag online), they actually talked about a theoretical inter cooler that used wax ("Performal") to basically make your IC a bigger heat sink (because after all, that's what it is for the most part). This idea is backwards to theirs :P
------
Re-inventing the wheel? Perhaps. Or, perhaps, Toyota's design is an hexagon... and this is a half baked octagon cake