p-n junction question

When various textbooks explain the operation of a p-n junction or a diode, they usually do not talk about metal contacts. That is usually a separate topic. But let's consider the whole thing together since this the only way the diode can operate and let's assume that both anode and cathode contacts are made from the same metal and therefore have the same energy level of free electrons.

The energy of an electron flowing through the diode under a forward bias condition would have to first be elevated from the metal energy level to the energy level of the conduction band of the N-type semiconductor, then it would drop when the electron crosses the junction an falls into a hole in the P-type semiconductor (and releases thermal or light energy) and then elevated again to the original metal energy level.

The voltage drop on a diode will be equal to the band gap and the energy spent to get an electron through the diode will be eaual to the energy released during the recombination. This energy is needed to elevate the energy of an electron before it plunges to a hole and, again, to get it out of a hole back to the metal, but any textbook would tell you that the voltage drop occurs on the junction itself to overcome a barrier created there though carrier diffusion before an external voltage was applied.

As far as I can see there is no barrier to overcome at the junction in the forward direction - it's a fall, like in a waterfall. But you do need to overcome a barrier between the metal terminal and the N-type semiconductor of the cathode to get to that high energy level and that's where an external energy is spent.

Am I missing something basic here? Does everyone else understand the story about the depletion region and how a dynamic balance (diffusion

  • drift) existing before an external voltage is applied to a diode somehow affects the operation of a diode ever after?
Reply to
vic
Loading thread data ...

In practice all you need know is its 0v7 forward volts drop and it wont conduct reverse polarity.

I can walk but dont understand how my legs work !

Reply to
Marra

Yep, this is a weird and fascinating topic. Search for info on "nonrectifying junction", also "ohmic contact"

If manufacturers just stuck some metal contacts on the semiconductor, they'd form Schottky diodes in series with the main diode (metal/ silicon junctions.) The diode as a whole would always turn off regardless of polarity.

So instead they apply heavy doping to the semiconductor surface before adding the metal contacts. This converts the semiconductor surface into a "metal-like" conductor. In diode diagrams you'll often see a layer of p+ next to the metal contact on the p-doped layer, and a layer of n- next to the metal touching the n-doped layer. In that case there still is an energy shift, but it doesn't present a barrier. Don't forget that energy shifts are NOT BARRIERS, they are more like ideal batteries. Only a depletion layer can act as a high- resistance barrier. ALso don't forget the thermocouple effect: that whenever you use copper and aluminum or iron wires in the same circuit, there is an energy shift at the metal junctions. These mismatched "work functions" don't create a high resistance because the energy shift around the circuit as a whole is zero. Any energy gain at one metal-metal junction will be cancelled by an energy loss at other, opposite metal-metal junctions. It's similar to hooking two batteries back to back in series: the voltages cancel out, but the batteries still form a low-resistance conductive path.

But why can we prevent the formation of a metal/semiconductor diode by adding extra-heavy doping to the semiconductor? It's because of quantum mechanics: the heavier the doping, the thinner the insulating depletion layer is formed at any particular reverse voltage. To form a diode, light doping is required. With heavy enough doping, the depletion layer becomes so thin that electrons can "tunnel" quantum-mechanically through this insulating region. The metal contact becomes like a tunnel diode, but a tunnel diode that turns fully on at all values of applied voltage.

(((((((((((((((((( ( ( ( ( (O) ) ) ) ) ))))))))))))))))))) William J. Beaty

formatting link
beaty chem.washington.edu Research Engineer billb eskimo.com UW Chem Dept, Bagley Hall RM74

206-543-6195 Box 351700, Seattle, WA 98195-1700
Reply to
billb

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.