Continue to Site

# Do holes in a semiconductor only move when there is a current going through the semiconductor?

#### s55

##### Junior Member level 2
I'm studing the PN junction process with the forward bias. The free electrons are recombined with empty holes, and they arrive at the positive polary of an external voltage source. In this process, the current flows in opposite direction with respect to the flow direction of the aforementioned free elections, that is to say, the current implies the movement of the emtpy holes.

Seems like from the above process, current can only flow by the movement of the empty holes. Here is my question. is it impossible that current can flow without the movement of the empy holes? Stochastically, some free electons may flow from n-type to p-type (or the positive polarity of an external voltage source) BYPASSING HOLES, and the flow of the free electrons may generate the current flow in thier oppositie directions WITHOUT RECOMBINATION WITH HOLES? Am I wrong?

In P type material - nominal current flow is by "holes" - of course really this is electrons jumping the other way

the "stickyness" of holes or rather their " in well " nature means that hole conduction is more resistive and slower than for N type material.

Thanks, let me ask one more question.
This link shows one page of the transistor circuit book, and this page explains the diffusion currents and the drift current when no external bias is applied to PN junction. Considering the equations (2.61) to (2.63) shown in the following link, there are I_drift.p and I_drift.n, where p indicates a hole, and n represents a electron.

Here is my question. In general, the current in the PN junction should flow by the flow of holes recombined with free electrons, and so I think even the drift current also needs to flow by the flow of the holes, which corresponds to I_drift.p only, which mentioned above. What makes I_drift.n? I'm not sure the reason why we have two different drift current? If I_drift.n is introduced by the flow of free electron, then I don't know the difference from I_drift.p and I_drift.n.

Holes don't move.
It's the electrons going from one hole location to the other, that occurs when there is current.

While "holes" are theoretical, they do have physical properties. In many semiconductors, the effective mass of holes is typically larger than that of electrons. This causes differences in the mobility of holes with many variables, { impurity density, carrier type, etc).
Thanks for your comment. Here is a simple question for the PN-junction diode. If free electrons move without the recombination with holes from the right end to the left end in the n-type, then dose current flow in the opposite direction of those electrons?

The mobility differences in flow exists due to the apparent differences in mass.

Recombination is not essential for semiconductor flow.
In fact, minimizing recomb. allows many of the best unique products to emerge. e.g. Ultra-high purity materials, permit high voltage IGBT's, high-speed microwave components, quantum dots and nanowires by design, special materials ( GaAs, GaN, InP) all benefit from low recombination rates.

Generally, lower voltage breakdown is always caused by material impurities which may have other benefits. .e.g. Base-emitter junctions and LEDs are typically -5V max before breakdown occurs.

The mobility differences in flow exists due to the apparent differences in mass.

Recombination is not essential for semiconductor flow.
In fact, minimizing recomb. allows many of the best unique products to emerge. e.g. Ultra-high purity materials, permit high voltage IGBT's, high-speed microwave components, quantum dots and nanowires by design, special materials ( GaAs, GaN, InP) all benefit from low recombination rates.

Generally, lower voltage breakdown is always caused by material impurities which may have other benefits. .e.g. Base-emitter junctions and LEDs are typically -5V max before breakdown occurs.
This is what I'm trying to find for last 3 days. Thanks a lot!

Impurities are a property of insulators.

Low ESR {or Rs} is a property of conductors within insulators due to the insulator/conductor boundary.

This low ESR is found in capacitors, batteries and semiconductors and may not be directly related to mobility, combined give unique properties. The larger width/gap ratios in junctions with large structures may conduct more current but at reduced mobility due to the inherent increase in capacitance.

I use ESR*C(@ 0V) = Tau is a figure of merit (FoM) for all diodes, BJT's & FET's, yet are never mentioned in datasheets.

Companies like Diodes Inc. have hundreds of patents in the fabrication of ultra low Rs in diodes and often specify this as low Rce in BJT's while retaining high mobility and higher speed or PIV parameters than conventional components but naturally at higher fabrication costs.

BJT's with Rce < 10 mohm are possible in BJTs and IGBT but with the advantage of much lower Cout with very high Vceo (kV) than MOSFETs yet both IGBTs & FETs have HV applications both due to the mobility effects.

Rce = Vce(sat)/Ic
RdsOn = Vds/Id = Vol/I in CMOS & (Vdd-Voh)/I (when Vgs > 200% to 250% of Vt = Vgs(th)
random page https://patents.justia.com/assignee/diodes-incorporated?page=3

Although much of this company's technology was bought from Zetex https://patents.justia.com/search?q=+Zetex in the UK.

Last edited: