Nature of conventional current flow in a p-n junction

Question

The OP has the following confusion:

(1) I'm confused about the concept of hole diffusion, ...

(2) Wouldn't it be more accurate to say that the electrons "fill" the holes in the p-type material?

(3) Would it be accurate to say that positive current in a p-n junction involves hole movement in the p-side, but when it comes to the n-side, involves "movement" of positive charge by positive ions, ...

---

Answer

I think the root cause of confusion is the inaccurate use of the word "move".

Let me use the musical chair game as a analogy, to explain to my 3 year old niece Jenny, how an empty space of a chair can "move", though the chairs do not "move" themselves.

---

---

Part 1 - Accurate use of the word "move"

(a) Suppose in the beginning, the 8 chairs are occupied by 6 children, leaving 2 chairs say, blue and green empty.

(b) Now the teacher Simon says, "Everybody moves one chair to the right".

(c) After much chaos, everybody indeed has moved to the right.

Now if I ask Jenny if she agrees that everybody has moved to the right, she would say yes.

---

Part 2 - Inaccurate use of the word "move"

Now if I then ask Jenny if the original empty chairs, blue and green, have moved to the left, she would say No, because the chairs cannot "move".

Here comes the trick question for Jenny: "But have any empty spaces moved?" She would say, yes, two empty spaces have moved.

Now if I challenge Jenny, "I know a real thing, such as a child, can move, how come an "empty space" can also move?"

She would be annoyed an replied "Why not, of course a non real thing, such as an empty space, can also move. How stupid you!"

---

Part 3 - How come the semiconductor physicists use the word "move" inaccurately?

Well, the semiconductor physicist Simon Sze uses the "hole" in his book on Semiconductor Physics and Technology.

By the way, the other physicist Stephen Hawkins is also very inaccurate to say that Black Hole is a "hole".

/ to continue, comments welcome.

---

Discussion, Conclusion, Recommendation, and Jokes

Lost

Two atoms are walking down the street, and one says to the other, "Wait, wait, we have to go back. I've lost an electron somewhere."

The second atom says, "Really? Are you sure?"

To which the first atom replies, "Yes. I'm positive."

Heard on NPR and contributed by Peter MacLean Kunhardt

...

/ to continue, ...

---

References

(1) Meaning is use: Wittgenstein on the limits of language - Timrayne, Philosophy For Change, 2014mar11

(2) p-n junction - Wikipedia

(3) PN Junction (Solar Cell and LED) YouTube Video - RED Inc Communication, 2014nov06, 135,696 views

(4) Semiconductor Basics - Electronics Tutorials

(5) PN Junction Theory - Electronics Tutorials

(6) PN Junction Diode - Electronics Tutorials

(7) Semiconductor Devices Physics Technology，Simon Sze 2nd Ed Wiley 2002 (free ebook)

(8) Modern Semiconductor Devices for Integrated Circuits (Ch 1 - Electronics and Holes) - Chenming Hu

(9) Modern Semiconductor Devices for Integrated Circuits (Ch 1~8) - Chenming Hu

(10) p–n junction - Wikipedia

(11) Input characteristics of NPN transistor - Khan Academy Video

(12) Parts of a transistor - Khan Academy YouTube Video

(13) Transistor Current and Parameters - Khan Academy YouTube Video

(14) PN Junction Q&A and Chat - EE SE 2020jul26

---

End of answer

but when it comes to the n-side, involves "movement" of positive charge by positive ions?

Nope, for the most part, the positive ions in the n-side are the pos-charged dopant atoms, and those are locked into the crystal lattice.

It's confusing because there are actually four ions involved, not just two.

First, in the p-doped silicon, the neutral dopant atoms don't remain neutral. Instead, a "hole" is created by each dopant atom, and it moves away. This leaves the dopant atom with net negative charge. Yet the wandering "hole" is actually a positively-ionized silicon atom! At the same time, on average the entire hunk of p-type silicon has zero net-charge. After all, every positive-charged hole has a negative-charged dopant ion somewhere nearby. In other words, p-type silicon is actually made up of equal quantities of:

1. fixed negative-charged dopant ions
2. movable positive-charged silicon ions (the wandering "holes.")

---

The n-type silicon is the opposite. The dopant atoms in n-type silicon will contribute wandering electrons. But when each electron initially leaves its dopant atom, that atom becomes a positive-charged ion. And, when the wandering electron is sitting upon some distant silicon atom, that atom temporarily becomes a negative silicon ion. So, n-type silicon is overall neutral, but is is composed of:

3. fixed positive-charged dopant ions
4. movable negative charged silicon ions (the mobile electrons.)

---

It gets worse!

Suppose that some holes wandered out of the p-side and invaded the n-type silicon? Thermal motion causes them to jump around randomly, and the random jumping can take them over into the n-side. They won't last long over there, but while they're briefly existing in the n-side, the holes are producing a region of positive net-charge! (After all, they no longer are near their negative-charged dopant atoms, which were all left behind in the p-type side.)

but when it comes to the n-side, involves "movement" of positive charge by positive ions?

YES! Because actually a "hole" is a positive-charged silicon ion ...so if holes invade the n-side, electrically it's just as if some pos-charged silicon atoms were invading. Yet the atoms themselves don't have to move. Just their "ionization" is wandering around through the crystal. (Heh, but at the same time, the n-side is full of positive-charged dopant ions which cannot move. So, whenever n-type silicon is full of wandering holes, it actually contains two kinds of positive ions, but only one of them can move around.)

PS

Important question: does p-type silicon have a positive charge? Nope, since p-type silicon is full of non-movable negative dopant ions. Their quantity is exactly the same as the quantity of wandering, positive-charged holes. P-type silicon is a conductor of course, and that means we can give it a positive net-charge by hooking it to the positive terminal of a power supply.

And to make things even more interesting, if we connect p-type to n-type, some electrons will diffuse from n-type to p-type and become trapped in the depletion zone (because they encountered holes, and "fell in.") This causes the p-type side to become negative charged, and the n-type side to become equally positive. This is the "built-in potential" of semiconductor junctions, which is caused by the "built-in" trapping of mobile charges in the depletion-zone. Or in other words, a diode junction is also a spontaneously self-charged capacitor.