pn = ni^2 exp(Va/Vt)

First, let me warn you that I'm not an engineer nor a student -- all I've got is this library book called "Microelectronic Devices and Circuits" by Clifton G. Fonstad, so I could be very very wrong, but are you sure about this line:

For a p-n junction wouldn't there be two different expressions for n_0 and p_0, one for each side of the junction? In other words, if you had a p-n junction (fixed-width font like Courier required):

Quasineutral Depletion Quasineutral vvvvvvvvvvvv vvvvvvvvv vvvvvvvvvvvv +---------------------------------------------+ | | | | | | | | A o--| p, N_Ap | | | n, N_Dn |--o B | | | | | | | | +---------------------------------------------+ -w_p -x_p x_n w_n

^^^^^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^^^^^ P-Side N-Side

then wouldn't the equations be something like:

p_0(-x_p) = N_Ap p_0(x_n) = ni^2/N_Dn

and

n_0(-x_p) = ni^2/N_Ap n_0(x_n) = N_Dn

or something like that?

Then I think if you multiply them together you get ni^2 in both cases.

But I dunno... I'm just trying to learn this stuff myself, and a lot of it is barely making sense. Maybe somebody who actually knows what he's talking about can give you a better (correct) answer.