You're looking for differences in the part-per-million range (PPM), which is the realm of ICP-OES and (with better detection limits) ICP-MS. You might find a local lab that will do it for $20-40 per element per test. Can't think of anything that you can homebrew that'll give you the precision and accuracy you'll need for something like that.
As for a control, you could use anything coated in PTFE. Aluminum may have some nominal amount of iron in it, and although the amount that would transfer over would be orders of magnitude less than what is likely already there (given that both plants and animals that would be cooked intrinsically contain iron), it would be difficult to account for it.