Here's something I can't get my head around: Take a diode (can be schottky, p-n etc) where the predicted depletion region is larger than the actual thickness of that layer will ever be. What happens? I've seen papers where its assumed that this curtails the built in potential. But it sounds fishy, namely because the all the textbook solutions are based on the assumption that there's a quasi-neutral region beside the depletion region and this gives you the boundary conditions you need to solve for depletion widths. The laws that always hold true, as far as I understand, are that the built-in potential equals the difference in Fermi levels and the vacuum level is continuous....so you should always get the same built-in potential across two layers? Regardless of their thickness?