An Alternate Theory on the Way Things Work: The Truth About Semiconductors

In order to learn how to design circuits and systems using transistors and other solid state devices, students of electronics are told in their courses how semiconductors function. The atomic structure of crystalline silicon is examined in its intrinsic and doped states. Discussion of energy levels, conduction band electrons and hole production quickly follow. Soon, the student encounters the PN junction, a basic building block of modern electronics, and learns about majority and minority carriers, depletion regions, barrier potentials, leakage current and other exotica. All of this information is intended to explain just how solid state devices really work, and it all ends up sounding quite obtuse. It sounds so complicated in fact, that the student assumes that only a genius could design such devices and relegates him or herself to a lower paying engineering or technician job. This should not be the case!

Recent investigation by this author has uncovered some startling facts:

1. Semiconductors don't really work the way we've all been told. In fact, the fundamental theory is much simpler.

2. This lie has been fabricated and perpetuated by the scientific elite, a group of people with cushy, high paying jobs; jobs so easy in the light of the real theory that even business executives, politicians, lawyers, friendly TV psychics, and other folk of nil capability could do it in their sleep (well, OK, the executives would still need an assistant and the lawyers would double bill you, but you get the general idea). By making their jobs sound difficult these people get to sit around all day eating eclairs and reading Esquire for six figure salaries.

It is high time that the truth be told and this farcical academic sham be torn down! As an example, we shall see how a simple diode really works.

So, you think a diode is composed of semiconducting material? Think again! One of the chief researchers at Bell labs in the 1940's and 50's was a certain Doctor Schlockling. After several experiments involving solid state diodes, Dr. Schlockling wrote in his diary:

"This stuff don't work at all. Better go back to tubes before they can me. Ooops, must let the dog out."

Schlockling was often pestered by his dog Melvin, who reminded him of a small self-propelled dust mop, and who was approximately as clean. The diary continued:

"If only there was a way in which Melvin could let himself out. Even better, if he couldn't get back in!"

This ominous tone lend to Schlockling's invention of the now-famous one-way doggie door, which can be seen mounted to the bottom of normal doors all across the countryside. Schlockling knew how to milk an idea, and took the form to its extremes by developing the cat door, the mouse door, the grasshopper door, the flea door (an early attempt at a flea and tick collar), and even the amoeba door. This last unit when properly designed could force bacteria out of the human body and not allow them to re-enter. This was instrumental in the development of the polio vaccine, in spite of the fact that the vaccine was produced some years earlier. Schlockling's greatest achievement, however, came when he shrunk the doggie door still further to produce the electron door. This is the fundamental unit of modern electronics.

In the figure below, we see a cross section of a diode and close-ups.
i-e419e66dfe0a68c1b9a226b856e7b1b1-diode1.gif
Even at 10,000X magnification we can still see nothing of the PN junction. If we go a bit further, something interesting comes into focus (see the second figure).
i-5bfcf9535097314e1a24fc00eeb0253d-diode2.gif

Yes! A PN junction is nothing more than a huge array of real tiny one-way doggie doors! Here's how it works: Electrons are a lot like marbles. When one hits a doggie door from behind, the door flips open allowing the marble through (i.e., allowing current to flow). If the electron hits the doggie door from the front, the flap closes and the electron can't get through (i.e., no current flow). Now obviously, if we hang the diode vertically, gravity should open all the doors and we'll get lots of electrons (i.e., current flow) in either direction. In truth, a real diode doesn't do this. Its operation will not matter on how the diode is oriented in space. This feature is accomplished by simply adding a small coil spring to the doggie door's hinge, forcing it to stay shut in the face of gravity. This has the negative side effect of requiring somewhat higher energy levels from the electrons to force the door open. This force happens to be the barrier potential of the diode! It has nothing to do with so-called depletion regions. If you were an electron, would you want to go through a place called a depletion region? Of course not! Neither would electrons. They're not stupid, you know. In any case, the stronger the spring, the greater the barrier potential. Diodes have been made of either silicon or germanium with barrier potentials of approximately .7 volts or .3 volts, respectively. In fact, silicon and germanium are really code words meaning strong spring and weak spring! It took a while to develop small, strong springs, and this is why germanium diodes were the first ones built.

Note that spring strength also plays a role in how tightly the flap can shut, thus indicating the reverse leakage current. Here again, we see strong spring "silicon" units having lower leakage. Theory also indicates that leakage should increase with temperature. This effect can be seen clearly in the doggie door model. At present, it is impossible to create both the frame and the door out of precisely the same material, and thus two different expansion coefficients exist. Because the flap is smaller than the frame, it will tend to curl away at higher temperatures, allowing more electrons to sneak through the gaps. At very low temperatures the flap tends to stick to the frame in much the same fashion that your tongue or lips will stick to an aluminum flag pole in freezing weather. This is sometimes referred to as the Christmas Story effect.

At very high forward energy levels the flaps may be literally torn off their hinges. This high volume of electrons at high energy will yield the maximum forward current. Also, note that if the energy level is high enough in the reverse direction, either the flaps will be bent and pushed through the frames or they will start to bounce violently at resonance, allowing electrons through. These two modes are referred to as avalanche and zener conduction, respectively. The required energy level indicates the reverse breakdown voltage.

Other fine points can be explained equally by the doggie door model, as well as bipolar and field effect transistors, SCRs, triacs, UJTs, and just about everything else in the filed of solid state electronics with the exception of the original 7400 series TTL logic gates which utilized an array of small, edible fungi and miniature harvester ants.

More on that, later.

More like this

The blogosphere is full of April Fools jokes today.

By crimsongirl (not verified) on 01 Apr 2007 #permalink

It's a conspiracy, crimsongirl. They want us to believe it is the 1st of April. It's actually the 1st of March. Yes, that's right - this evil conspiracy is TRYING TO MAKE EVERYONE 31 DAYS OLDER!!!!

There are several transistor models, covering different modes of operation. In poisonous material, such as GaAs or InP, the chemical reactivity is overseen by Maxwell's demons. So they can be made to replace the springs, allowing faster action and higher frequencies. (Thus inspiring the saying that the road to hell is easy.)

Also, normally electrons flow complacently as a current, despite being their own men. (As all decent fermions should be.) But if the voltage goes up in narrow devices, the material can actually loose its marbles. The electrons will stop hitting the locals pubs (atoms) at every corner and start chasing down the street, as you say tearing the hinges. These devices are going ballistic.

It's actually the 1st of March.

Oh? Well, I've never quite got that stuff about the Tides of March.

By Torbjörn Larsson (not verified) on 01 Apr 2007 #permalink