consider a p-type semiconductor slab with a concentration gradient highest towards extreme left and gradually reducing towards right.looking at the atomic level we have more more holes at the left end and when we apply a bias with positve terminal to the highest concentration side and negative terminal to the lowest concentration side ,more holes on positive side facilitate movement of electrons from negative terminal to positive side resulting in a net movement of holes from +ve to -ve terminal as movement of holes is nothing but movement of bonded electrons.now reverse the bias. positive terminal is towards the low concentration side.but since there are few holes electron movement is restricted.thus there is a very negligeable hole current.this phenomena is as if a water pipe with crosssection reduced.thus we have current in one direction and a very less current in the reverse direction.is it not approaching the characteristics of an ideal diode in the sense that it is not having a cut in voltage.
note:there is no experimental proof to show above one and it is only a thought.anyone who has the facility may kindly do the above and send the result to the above mail address.

 

Anurup,

Are you talking about a slab that is first produced by conventional crystal growth methodology, at which time impurities are added, and then, in effect, performing some sort of machining "operation" to produce the "shape" you describe?

Regards,
Phil Corso, PE
Boca Raton, FL
[[email protected]] ([email protected]) {[email protected]}