The Transistor - History
& Developments
|
Anyone who works in electronics, as I have done, knows what an enormous & constantly changing field it is. Keeping up to date with all that's going on is impossible. As this section of my site concentrates on computers & CPUSs I thought it appropriate to include some pages on a little device that made it all possible: THE TRANSISTOR. This in itself is a large subject area so I have tried to include what I think were the main advancements & inventions in this field. I have used the following articles as the explanations are such that they make it easier to understand how these devices work, as well as describing the history of developments.
A Brief History |
How The First
Transistor was Invented,
|
This picture shows the workbench of John Bardeen and Walter Brattain at Bell Laboratories. They were supposed to be doing fundamental research about crystal surfaces. The experimental results hadn't been very good, though, and there's a rumor that their boss, William Shockley, came near to cancelling the project. But in 1947, they switched to using tremendously pure materials. And it dawned on them that they could build the circuit in the picture. It was an amplifier!
The three shared a Nobel Prize. Bardeen and Brattain continued in research (and Bardeen later won another Nobel). Shockley quit to start a semiconductor company in Palo Alto. It folded, but its staff went on to invent the integrated circuit (the "chip") and to found Intel Corporation.
By 1960, all important computers used transistors for logic, and ferrite cores for memory. Memory chips replaced core in the 1970's.
Getting
Wet
On November 17, 1947, Walter
Brattain dumped his whole experiment into a thermos of water. The
silicon contraption he'd built was supposed to help him study how
electrons acted on the surface of a semiconductor -- and why
whatever they were doing made it impossible to build an
amplifier. But condensation kept forming on the silicon and
messing up the experiment. To get rid of that condensation,
Brattain probably should have put the silicon in a vacuum, but he
decided that would take too long. Instead he just dumped the
whole experiment under water -- it certainly got rid of the
condensation!
Out of the blue, the wet device created the largest amplification he'd seen so far. He and another scientist, Robert Gibney, stared at the experiment, stunned. They began fiddling with different knobs and buttons: by turning on a positive voltage they increased the effect even more; turning it to negative could get rid of it completely. It seemed that whatever those electrons had been doing on the surface to block amplification had somehow been canceled out by the water -- the greatest obstacle to building an amplifier had been overcome.
Putting
the Idea
to Use
When John Bardeen was told what
had happened he thought of a new way to make an amplifier. On
November 21, Bardeen suggested pushing a metal point into the
silicon surrounded by distilled water. The water would eliminate
that exasperating electron problem just under the point as it had
in the thermos. The tough part was that the contact point
couldn't touch the water, it must only touch the silicon. But as
always, Brattain was a genius in the lab. He could build
anything. And when this amplifier was built, it worked. Of
course, there was only a tiny bit of amplification -- but it
worked.
Big
Amplification
Once they'd gotten slight
amplification with that tiny drop of water, Bardeen and Brattain
figured they were on the road to something worthwhile. Using
different materials and different setups and different
electrolytes in place of the water, the two men tried to get an
even bigger increase in current. Then on December 8, Bardeen
suggested they replace the silicon with germanium. They got a
current jump, all right -- an amplification of some 330 times --
but in the exact opposite direction they'd expected. Instead of
moving the electrons along, the electrolyte was getting the holes
moving. But amplification is amplification -- it was a start.
Brattain
Makes a
Mistake
Unfortunately this giant jump in
amplification only worked for certain types of current -- ones
with very low frequencies. That wouldn't work for a phone line,
which has to handle all the complex frequencies of a person's
voice. So the next step was to get it to work at all kinds of
frequencies.
Bardeen and Brattain thought it might be the liquid which was the problem. So they replaced it with germanium dioxide -- which is essentially a little bit of germanium rust. Gibney prepared a special slab of germanium with a shimmering green oxide layer on one side. On December 12, Brattain began to insert the point contacts.
Nothing
happened.
In fact the device worked as if
there was no oxide layer at all. And as Brattain poked the gold
contact in again and again, he realized that's because there
wasn't an oxide layer. He had washed it off by accident. Brattain
was furious with himself, but decided to fiddle with the point
contact anyway. To his surprise, he actually got some voltage
amplification -- and more importantly he could get it at all
frequencies! The gold contact was putting holes into the
germanium and these holes canceled out the effect of the
electrons at the surface, the same way the water had. But this
was much better than the version that used water, because now,
the device was increasing the current at all frequencies.
Bringing
it All
Together
In the past month, Bardeen and
Brattain had managed to get a large amplification at some
frequencies and they'd gotten a small amplification for all
frequencies -- now they just had to combine the two. They knew
that the key components were a slab of germanium and two gold
point contacts just fractions of a millimeter apart. Walter
Brattain put a ribbon of gold foil around a plastic triangle, and
sliced it through at one of the points. By putting the point of
the triangle gently down on the germanium, they saw a fantastic
effect -- signal came in through one gold contact and increased
as as it raced out the other. The
first point-contact transistor had
been made.
Telling
the Brass
For a week, the scientists kept
their success a secret. Shockley asked Bardeen and Brattain to
show off their little plastic triangle at a group meeting to the
lab and the higher-ups on December 23. After the rest of the lab
had a chance to look it over and conduct a few tests, it was
official -- this tiny bit of germanium, plastic and gold was the
first working solid state amplifier.
The transistor and subsequently the integrated circuit must certainly qualify as two of the greatest inventions of the twentieth century. These devices are formed from materials known as semiconductors, whose properties were not well-understood until the 1950s. However, as far back as 1926, Dr. Julius Edgar Lilienfield from New York filed for a patent on what we would now recognize as an NPN junction transistor being used in the role of an amplifier (the patent title was "Method and apparatus for controlling electric currents").
Unfortunately, serious research on semiconductors didn't really commence until World War II.
At that time it was recognized that devices formed from semiconductors had potential as amplifiers and switches, and could therefore be used to replace the prevailing technology of vacuum tubes, but that they would be much smaller, lighter, and would require less power.
All of these factors were of interest to the designers of the radar systems which were to play a large role in the war.
Bell Laboratories in the United States began research into semiconductors in 1945, and physicists William Shockley, Walter Brattain and John Bardeen succeeded in creating the first point- contact germanium transistor on the 23rd December, 1947 (they took a break for the Christmas holidays before publishing their achievement, which is why some reference books state that the first transistor was created in 1948).
In 1950, Shockley invented a new device called a bipolar junction transistor, which was more reliable, easier and cheaper to build, and gave more consistent results than point-contact devices. (Apropos of nothing at all, the first TV dinner was marketed by the C.A. Swanson company three years later.)
By the late 1950s, bipolar transistors were being manufactured out of silicon rather than germanium (although germanium had certain electrical advantages, silicon was cheaper and easier to work with). Bipolar junction transistors are formed from the junction of three pieces of doped silicon called the collector, base, and emitter. The original bipolar transistors were manufactured using the mesa process, in which a doped piece of silicon called the mesa (or base) was mounted on top of a larger piece of silicon forming the collector, while the emitter was created from a smaller piece of silicon embedded in the base.
n 1959, the Swiss physicist Jean Hoerni invented the planar process, in which optical lithographic techniques were used to diffuse the base into the collector and then diffuse the emitter into the base. One of Hoerni's colleagues, Robert Noyce, invented a technique for growing an insulating layer of silicon dioxide over the transistor, leaving small areas over the base and emitter exposed and diffusing thin layers of aluminum into these areas to create wires. The processes developed by Hoerni and Noyce led directly to modern integrated circuits.
n 1962, Steven Hofstein and Fredric Heiman at the RCA research laboratory in Princeton, New Jersey, invented a new family of devices called metal-oxide semiconductor field-effect transistors (MOS FETs for short).
Although these transistors were somewhat slower than bipolar transistors, they were cheaper, smaller and used less power. Also of interest was the fact that modified metal-oxide semiconductor structures could be made to act as capacitors or resistors.
Individually packaged transistors were much smaller than their vacuum tube predecessors, but designers desired still smaller electronic switches. To a large extent the demand for miniaturization was driven by the demands of the American space program. For some time people had been thinking that it would be a good idea to be able to fabricate entire circuits on a single piece of semiconductor.
The first public discussion of this idea is credited to a British radar expert, G.W.A. Dummer, in a paper presented in 1952. However, it was not until the summer of 1958, that Jack Kilby, working for Texas Instruments, succeeded in fabricating multiple components on a single piece of semiconductor. Kilby's first prototype was a phase shift oscillator and, although manufacturing techniques subsequently took different paths to those used by Kilby, he is still credited with the creation of the first true integrated circuit.
By 1961, Fairchild and Texas Instruments had announced the availability of the first commercial planar integrated circuits comprising simple logic functions. This announcement marked the beginning of the mass production of integrated circuits. In 1963, Fairchild produced a device called the 907 containing two logic gates, each of which consisted of four bipolar transistors and four resistors. The 907 also made use of isolation layers and buried layers, both of which were to become common features in modern integrated circuits.
In 1967, Fairchild introduced a device called the Micromosaic, which contained a few hundred transistors. The key feature of the Micromosaic was that the transistors were not initially connected to each other. A designer used a computer program to specify the function the device was required to perform, and the program determined the necessary transistor interconnections and constructed the photo-masks required to complete the device. The Micromosaic is credited as the forerunner of the modern application-specific integrated circuit (ASIC), and also as the first real application of computer aided design.
In 1970, Fairchild introduced the first 256-bit static RAM called the 4100, while Intel announced the first 1024-bit dynamic RAM, called the 1103, in the same year.
A vacuum tube is just that: a glass tube surrounding a vacuum (an area from which all gases have been removed). What makes it interesting is that when electrical contacts are put on the ends, you can get a current to flow though that vacuum. Thomas Edison noticed this first in 1883. While fiddling with lightbulbs he saw that he could get current to jump from the hot filament to a metal plate at the bottom. What Edison discovered (and it was promptly dubbed the "Edison effect") was that electrical current doesn't need wire to move. It can travel right through a gas or through a vacuum. (The Edison effect is, incidentally, the only piece of real science Edison contributed. He was an inventor, a tinkerer, not a scientist).
That didn't turn out to be very useful information until 1904. That's when a British scientist named John A. Fleming made a vacuum tube known today as a diode. Then the diode was known as a "valve," because it forced current in the tube to travel exclusively in one direction. Getting that single directional flow was critical for radio sets which needed to turn alternating current into direct current.
The vacuum tube didn't reach its full maturity until Lee De Forest came along a decade later. De Forest invented something he called the "audion." Not only did it force current to move in a single direction, but it could be used to increase the current along the way. De Forest put a metal grid in the middle of the vacuum tube. By using a small input current to change the voltage on the grid, De Forest could control the flow of a second, more powerful current, through the tube. The strength of two currents was not necessarily related -- a weak current might be applied to the tube's grid, but a much stronger current could come out the main electrodes of the tube.
Turning weak currents into strong currents was crucial for a number of new technologies at that time. Bell Labs made use of it for its coast to coast phone system and vacuum tubes soon found their way into everything from hearing aids to radios to televisions.
Metal conducts electricity; rubber doesn't. Gold conducts electricity; Styrofoam doesn't. Most materials fall easily into one category or the other. Everyone knows, for example, that if you want a good wire you're going to make it out of copper, not plastic. But there's a whole group of materials that fall in between. Their conductivity is in between metals and insulators. And their conductivity can be modified transiently, by shining a light on them or injecting charges. They're known as semiconductors, and they first became interesting to physicists in the late 1920s.
At first no one could figure out how they worked. Scientists once thought that certain atoms simply held onto their electrons more strongly than others. But as physicists got a better understanding of what an atom looked like, they understood what was really going on.
Different kinds of atoms have different numbers of electrons swarming around them. These electrons can only sit in specific places around the atom. It's sort of like rows of seats in a theater-in-the-round: a few electrons get to sit in the first row around the stage, and when that's filled the next electrons sit in the next row and so on. Electrons in a filled row stay put -- just as in a theater it's harder to get out when you've got people sitting on each side of you. In an insulator, every row is completely filled. Consequently the electrons rarely move. No moving electrons means no electricity can pass through.
But if you're sitting in the back row of a movie theater and the seats aren't full, you could easily get up, switch seats, maybe even decide to check out a different movie in the next theater. In a metal, the last row isn't filled with electrons. The outer electrons have little loyalty to the atom they're with and readily wander off in search of other atoms. This translates to many moving electrons, which means metals can easily conduct electricity.
So what happens with semiconductors? They reside somewhere in the middle. They are mostly made of atoms that don't conduct electricity, but they have a handful of atoms with loose electrons. Under certain circumstances -- by changing things like temperature or how much energy is injected -- these loose electrons will start a flowing current.
That means that depending on what you do, semiconductors can transiently conduct more or less electricity. It's just that property that transistors exploit.
| "He called these fellows
up and said to drop their work and come on up here and take a look at
something. Then all these men gathered in Kelly's office. He wouldn't
take no for an answer. He said, 'Drop it, and come on up here!' And so
they all came up here." -- Russell Ohl, 20 August 1976, about a day in 1940 when he demonstrated the P-N junction to his boss |
In 1939, vacuum tubes were state of the art in radio equipment. People had previously used crystals for radios, but the crystals were so maddeningly inconsistent and mysterious it was a wonder they worked at all. Vacuum tubes were simple, and they worked. Most scientists agreed tubes were the future for radio and telephones everywhere.
Russell Ohl didn't agree. He kept right on studying crystals, occasionally having to fight Bell Labs administration to let him do it. Ohl thought silicon crystals' erratic behavior was due to impurities in the crystal, not any problem in the silicon itself. He thought that if he could purify silicon enough, the crystals just might provide the improved radio broadcasting capabilities for which everyone was looking.
A Quirky Crystal
Much of his research in 1939 was devoted to producing ultra-pure crystals. As he expected, his purified silicon crystals-- now 99.8 percent pure -- were much more consistent. They worked the way a rectifier should, allowing current to flow in one direction and not the other. At least, most of them worked. On February 23, Ohl sat down to examine a particularly curious crystal that was as quirky as the cat's whisker crystals of old.
The crystal had a crack down the middle. Ohl was examining how much current flowed through one side of the crack versus the other, when he noticed something peculiar. The amount of current changed when the crystal was held over a bowl of water. And a hot soldering iron. And an incandescent lamp on the desk in the room.
By early afternoon, Ohl realized that it was in fact light shining on the crystal that caused this small current to begin trickling through it. On March 6, he showed his prize silicon rod to Mervin Kelly. Kelly quickly called Walter Brattain and Joseph Becker to the scene.
Ohl had his coal-black crystal attached to a voltmeter in front of him. He turned on a flashlight, aimed it at the silicon, and the voltage instantly jumped up to half a volt. This was ten times anything Brattain had ever seen before. He was stunned, but not too stunned to produce an off-the-cuff explanation. The electrical current must be due to some barrier being formed right at the crack in the crystal.
The Quirks Explained
With more research, what was going on became clear: the crystal had different levels of purity on either side of the crack. Due to the subtle traces of extra elements, one side had an excess of electrons, and the other side a deficit. Since opposites attract, the electrons from one side had rushed over to the other -- but they went only so far, creating a thin barrier of excess charges right at the central crack. That barrier created a one way street -- electrons could now only travel in one direction across it.
When Ohl shined light on the rod, energy from the light kicked sluggish electrons out of their resting places and gave them the boost they needed to travel around the crystal. But due to the barrier, there was only one way they could travel. All those electrons moving in a single direction became an electric current. Ohl's crystal was the ancestor of modern day solar cells, which take energy from the sun and convert it into electricity. But for Bell Labs on that day, it opened up the idea that crystals might be just the thing needed to replace vacuum tubes.
A diode, or "rectifier," is any device through which electricity can flow in only one direction. The first diodes were crystals used as rectifiers in home radio kits. A weak radio signal was fed into the crystal through a very fine wire called a cat's whisker. The crystal removed the high frequency radio carrier signal, allowing the part of the signal with the audio information to come through loud and clear. The crystal was filled with impurities, making some sections more resistant to electrical flow than others. Using the radio required positioning the cat's whiskers over the right kind of impurity to get electricity to flow through the crystal to the output below it.
At the time, though, no one really understood about the impurities -- then in 1939 Russell Ohl accidentally discovered that it was the boundary between sections of different purity that made the crystal work. Now that the way they work is understood, manufacturers make crystal diodes that work much more consistently than the ones in those original radio kits.
A crystal diode is made of two different types of semiconductors right next to each other. One side is easy for electrons to travel through; one side is much tougher. It's something like trying to swim through a pool filled with water and then a pool filled with mud: swimming through water is easy; swimming through mud is next to impossible. To an electron some semiconductors seem like water, some like mud. (For more information, read about semiconductors in Everything You Ever Wanted to Know about Conduction.)
One side of the semiconductor boundary is like mud, one like water. If you try to get electricity to move from the mud side to the water side, there's no problem. The electrons just jump across the boundary, forming a current. But try to make electricity go the other way and nothing will happen. Electrons that didn't have to work hard to travel around the water side just don't have enough energy to make it into the mud side. (In real life, there are always a few electrons that can trickle in the wrong direction, but not enough to make a big difference.)
This boundary has turned out to be crucial for our daily lives. Diodes change the alternating current that comes from your wall outlet into the direct current that most appliances need. And transistors need two such boundaries to work.
The first transistor was about half an inch high. That's mammoth by today's standards, when 7 million transistors can fit on a single computer chip. It was nevertheless an amazing piece of technology. It was built by Walter Brattain.
Before Brattain started, John Bardeen
told him that they would need two metal contacts within .002
inches of each other -- about the thickness of a sheet of paper.
But the finest wires then were almost three times that width and
couldn't provide the kind of precision they needed.
Instead of bothering with tiny wires, Brattain attached a single
strip of gold foil over the point of a plastic triangle. With a
razor blade, he sliced through the gold right at the tip of the
triangle. Voila: two gold contacts just a hair-width apart.
The whole triangle was then held over a crystal of germanium on a spring, so that the contacts lightly touched the surface. The germanium itself sat on a metal plate attached to a voltage source. This contraption was the very first semiconductor amplifier, because when a bit of current came through one of the gold contacts, another even stronger current came out the other contact.
Here's why it worked: Germanium is a semiconductor and, if properly treated, can either let lots of current through or let none through. This germanium had an excess of electrons, but when an electric signal traveled in through the gold foil, it injected holes (the opposite of electrons) into the surface. This created a thin layer along the top of the germanium with too few electrons.
Semiconductors with too many electrons are known as N-type and semiconductors with too few electrons are known as P-type. The boundary between these two kinds of semiconductors is known as a P-N junction, and it's a crucial part of a transistor. In the presence of this junction, current can start to flow from one side to the other. In the case of Brattain's transistor, current flowed towards the second gold contact.
Think about what that means. A small current in through one contact changes the nature of the semiconductor so that a larger, separate current starts flowing across the germanium and out the second contact. A little current can alter the flow of a much bigger one, effectively amplifying it.
Of course, a transistor in a telephone or in a radio has to handle complex signals. The output contact can't just amplify a steady hum of current, it has to dutifully replicate a person's voice, or an entire symphony. Luckily, a semiconductor is perfectly suited to this job. It is exquisitely sensitive to how many extra or missing electrons are inside. Each time the input signal shoves more holes into the germanium, it changes the way current flows across the crystal -- the output current instantly gets larger and smaller, perfectly mimicking the input.
1948-1951
There was no doubt about it, point-contact transistors were fidgety. The transistors being made by Bell just didn't work the same way twice, and on top of that, they were noisy. While one lab at Bell was trying to improve those first type-A transistors, William Shockley was working on a whole different design that would eventually get rid of these problems.
Early in 1948, Shockley conceived of a transistor that looked like a sandwich, with two layers of one type of semiconductor surrounding a second kind. This was a completely different setup which didn't have the shaky wires that made the point-contact transistors so hard to control.
Not Just on the Surface
A working sandwich transistor would require that electricity travel straight across a crystal instead of around the surface. But Bardeen's theory about how the point-contact transistor worked said that electricity could only travel around the outside of a semiconductor crystal. In February of 1948, some tentative results in the Shockley lab suggested this might not be true. So the first thing Shockley had to do was determine just what was going on.
Careful experiments led by a physicist in the group, Richard Haynes, helped. Haynes put electrodes on both sides of a thin germanium crystal and took very sensitive measurements of the size and speed of the current. Electricity definitely flowed straight through the crystal. That meant Shockley's vision of a new kind of transistor was theoretically possible.
Growing Crystals
But Haynes also discovered that the layer in the middle of the sandwich had to be very thin and very pure.
The man who paved the way for growing the best crystals was Gordon Teal. He didn't work in Shockley's group, but he kept tabs on what was going on. He'd even been asked to provide crystals for the Solid State team upon occasion. Teal thought transistors should be built from a single crystal-as opposed to cutting a sliver from a larger ingot of many crystals. The boundaries between all the little crystals caused ruts that scattered the current, and Teal had heard of a way to build a large single crystal which wouldn't have all those crags. The method was to take a tiny seed crystal and dip it into the melted germanium. This was then pulled out ever so slowly, as a crystal formed like an icicle below the seed.
Teal knew how to do it, but no one was interested. A number of institutions at the time, Bell included, had a bad habit of not trusting techniques that hadn't been devised at home. Shockley didn't think these single crystals were necessary at all. Jack Morton, head of the transistor-production group, said Teal should go ahead with the research, but didn't throw much support his way.
Luckily, Teal did continue the research, working with engineer John Little. Three months later, in March of 1949, Shockley had to admit he'd been wrong. Current flowing across Teal's semiconductors could last up to one hundred times longer than it had in the old cut crystals.
Growing Even Better Crystals
Nice crystals are all well and good, but a sandwich transistor needed a sandwich crystal. The outer layers had to be a semiconductor with either too many electrons (known as N-type) or too few (known as P-type), while the inner layer was the opposite. Under Shockley's prodding, Teal and Morgan Sparks began adding impurities to the melt while they pulled the crystal out of the melt. Adding impurities is known as "doping," and it's how one turns a semiconductor into N- or P-type.
As they pulled the seed crystal out of an N-type germanium melt, they quickly added some gallium to turn the melt into P-type. As a layer of P-type formed on the ever-lengthening crystal, they added antimony, which compensated for the gallium and turned the melt back into N-type. Once the process was done, there was a single, thin crystal formed into a perfect sandwich.
By etching away the surface of the outside layers, Sparks and Teal left a tiny bit of P-type crystal protruding. To this they attached a fine electrode-creating a circuit the way Shockley had envisioned. On April 12, 1950, they tested what they had built. Without a doubt, more current came out of the sandwich than went in. It was a working amplifier.
 |
| The First Junction Transistor |
The first junction transistor had been born.
But It Wasn't a Very Good One . . . Yet
This transistor could amplify electrical signals, but not particularly complicated ones. If the signal changed rapidly, as a voice coming over a phone line does, the transistor couldn't keep up and would garble the output. The problem lay in the middle of the sandwich: it was too easy for electric current to spread out and become unfocused as it crossed the P-type layer. To solve the problem, the layer had to be even thinner.
In January of 1951, Morgan Sparks figured out a way to accomplish that. By pulling the crystal out more slowly than ever, while constantly stirring the melt, he managed to get the middle layer of the sandwich thinner than a sheet of paper.
This new, improved sandwich did all that the researchers hoped. They still weren't up to the point-contact transistor's ability to handle signals that fluctuated extremely rapidly, but in every other way they were superior. They were much more efficient, used very little power to work, and they were so much quieter that they could handle weaker signals than the type-A transistors ever could.
In July of 1951, Bell held another press conference -- this time announcing the invention of a working and efficient junction transistor.
The sandwich transistor was William Shockley's brainchild. It's also called the junction transistor. While the rest of the lab was busy researching Bardeen and Brattain's point-contact transistor, Shockley secretly worked ahead on his own project.
The sandwich transistor does in fact look like a little sandwich: two "bread" layers surrounding a piece of "meat." If you attach a wire with current to one side of the bread and measure what comes out the other side, not much happens. But trickle a little current into the meat and suddenly the contraption springs to life. Electricity can race through.
The transistor base (the "meat" of the sandwich) acts like the handle on a faucet. Turn it one way and current will gush through like water through a hose. Turn it the other and the current will stop altogether.
The two slabs of "bread" (one is called the emitter and one is called the collector) have excess electrons. These electrons can scoot around jumping from atom to atom, even into the center if circumstances are right. The meat, on the other hand, is missing electrons. (The sandwich could just as easily have bread with few electrons and meat with an excess, but Shockley's first version is described here.)
The sandwich transistor works because of the curious things that happen at the border between the bread and the meat. A boundary between semiconductors with different amounts of electrons like this is called a P-N junction. And P-N junctions can do interesting things -- they create a one-way road in a crystal. The crystal can conduct in one direction, but not in the other.
A positive voltage in the base pulls the electrons across in the correct direction. It's like squeezing a tube of toothpaste -- the electrons rush out. Changing the voltage to one that makes the electrons travel the wrong way is the equivalent of putting the cap back on. No matter how much you squeeze, the toothpaste won't come out. The current has been stopped completely.
These days, nobody makes "sandwich transistors" as Shockley designed them. But the concepts he developed with his design created a new class of transistors called junction transistors, some of which are still used today in specialized applications.
1954
It was late afternoon at a conference for the Institute of Radio Engineers. Many people giving talks had complained about the current germanium transistors -- they had a bad habit of not working at high temperatures. Silicon, since it's right above germanium on the periodic table and has similar properties, might make a better gadget. But, they said, no one should expect a silicon transistor for years.
Then Gordon Teal of Texas Instruments stood up to give his talk. He pulled three small objects out of his pocket and announced: "Contrary to what my colleagues have told you about the bleak prospects for silicon transistors, I happen to have a few of them here in my pocket."
That moment catapulted TI from a small start-up electronics company into a major player. They were the first company to produce silicon transistors -- and consequently the first company to produce a truly consistent mass-produced transistor.
Scientists knew about the problems with germanium transistors. Germanium worked, but it had its mood swings. When the germanium heated up -- a natural outcome of being part of an electrical circuit -- the transistor would have too many free electrons. Since a transistor only works because it has a specific, limited amount of electrons running around, high heat could stop a transistor from working altogether.
While still working at Bell Labs in 1950, Teal began growing silicon crystals to see if they might work better. But just as it had taken years to produce pure enough germanium, it took several years to produce pure enough silicon. By the time he succeeded, Teal was working at Texas Instruments. Luring someone as knowledgeable about crystals as Teal away from Bell proved to be one of the most important things TI ever did.
On April 14, 1954, Gordon Teal showed TI's Vice President, Pat Haggerty, a working silicon transistor. Haggerty knew if they could be the first to sell these new transistors, they'd have it made. The company jumped into action -- four weeks later when Teal told his colleagues about the silicon transistors in his pocket, TI had already started production.
| "It was to be used in
telephone exchanges and Shockley felt it would be a revolutionary
product that would replace all of the mechanical relays in these
gigantic electronic switching stations and it would revolutionize the
industry. And the idea was true. The trouble with the
four-layer diode was it was years ahead of its time in processing
complexity and we just couldn't make it." -- Harry Sello, interview for "Transistorized!" |
The four-layer diode was the key to William Shockley's plan to revolutionize AT&T's phone system. It was a great device in theory, but not in practice -- at least not at the time when Shockley wanted to build it. The four-layer diode, also known as a Shockley Diode, is a crystal made of alternating layers of N- and P- type semiconductors. By putting in four layers, instead of the three used in transistors, the Shockley Diode could do more than a transistor. For one, it acted like a rectifier, able to turn alternating current into direct current. Two, it switched on and off when a specific amount of voltage -- known as the breakover voltage -- was applied. The four-layer diode, therefore, could be used to replace both the rectifiers and transistors necessary to connect long distance phone calls.
In essence, the four-layer diode was the first integrated circuit since it did the work of two transistors, two resistors, and a diode -- all in a single crystal. Unfortunately, they were so tricky to make that Shockley's company, Shockley Semiconductor, never managed to build any that were truly commercially viable. When the integrated circuit was invented in 1958, it eclipsed the four-layer diode's capabilities and any market for the diode quickly dried up.
Four-layered semiconductors are, however, used today. They're known as "thyristors" and a variety of types, including Shockley Diodes, exist. Thyristors are chiefly used as switches to control power supplies -- often in electrical utility systems.
In 1945, Shockley had an idea
for making a solid state device out of semiconductors. He
reasoned that a strong electrical field could cause the flow of
electricity within a nearby semiconductor. He tried to build one,
then had Walter Brattain try to build it, but it didn't work.
Three years later, Brattain and Bardeen built the first working transistor, the germanium point-contact transistor, which was manufactured as the "A" series. Shockley then designed the junction (sandwich) transistor, which was manufactured for several years afterwards. But in 1960 Bell scientist John Atalla developed a new design based on Shockley's original field-effect theories. By the late 1960s, manufacturers converted from junction type integrated circuits to field effect devices. Today, most transistors are field-effect transistors. You are using millions of them now.
MOS-FETs
Most of today's transistors are "MOS-FETs", or Metal Oxide Semiconductor Field Effect Transistors. They were developed mainly by Bell Labs, Fairchild Semiconductor, and hundreds of Silicon Valley, Japanese and other electronics companies.
Field-effect transistors are so named because a weak electrical signal coming in through one electrode creates an electrical field through the rest of the transistor. This field flips from positive to negative when the incoming signal does, and controls a second current traveling through the rest of the transistor. The field modulates the second current to mimic the first one -- but it can be substantially larger.
On the bottom of the transistor is a U-shaped section (though it's flatter than a true "U") of N-type semiconductor with an excess of electrons. In the center of the U is a section known as the "base" made of P-type (positively charged) semiconductor with too few electrons. (Actually, the N- and P-types can be reversed and the device will work in exactly the same way, except that holes, not electrons, would cause the current.)
Three electrodes are attached to the top of this semiconductor crystal: one to the middle positive section and one to each arm of the U. By applying a voltage to the electrodes on the U, current will flow through it. The side where the electrons come in is known as the source, and the side where the electrons come out is called the drain.
If nothing else happens, current will flow from one side to the other. Due to the way electrons behave at the junction between N- and P-type semiconductors, however, the current won't flow particularly close to the base. It travels only through a thin channel down the middle of the U.
There's also an electrode attached to the base, a wedge of P-type semiconductor in the middle, separated from the rest of the transistor by a thin layer of metal-oxide such as silicon dioxide (which plays the role of an insulator). This electrode is called the "gate." The weak electrical signal we'd like to amplify is fed through the gate. If the charge coming through the gate is negative, it adds more electrons to the base. Since electrons repel each other, the electrons in the U move as far away from the base as possible. This creates a depletion zone around the base – a whole area where electrons cannot travel. The channel down the middle of the U through which current can flow becomes even thinner. Add enough negative charge to the base and the channel will pinch off completely, stopping all current. It's like stepping on a garden hose to stop the flow of water. (Earlier transistors controlled this depletion zone by making use of how electrons move when two semiconductor slabs are put next to each other, creating what is known as a P-N junction. In a MOS-FET, the P-N junction is replaced with metal-oxide, which turned out to be easier to mass produce in microchips.)
Now imagine if the charge coming through the gate is positive. The positive base attracts many electrons – suddenly the area around the base which used to be a no-man's-land opens up. The channel for current through the U becomes larger than it was originally and much more electricity can flow through.
Alternating charge on the base, therefore, changes how much current goes through the U. The incoming current can be used as a faucet to turn current on or off as it moves through the rest of the transistor.
On the other hand, the transistor can be used in a more complex manner as well -- as an amplifier. Current traveling through the U gets larger or smaller in perfect synch with the charge coming into the base, meaning it has the identical pattern as that original weak signal. And, since the second current is connected to a different voltage supply, it can be made to be larger. The current coming through the U is a perfect replica of the original, only amplified. The transistor is used this way for stereo amplification in speakers and microphones, as well as to boost telephone signals as they travel around the world.
Footnote on Shockley
Shockley watched as Silicon Valley grew but could not seem to enter The Promised Land he had envisioned. He never was able to make field effect transistors, while other companies designed, grew, and prospered. Fred Seitz called Shockley "The Moses of Silicon Valley."
The concept behind an
integrated chip is relatively simple: an entire electrical
circuit with numerous transistors, wires, and other electrical
devices all built into a single square of silicon. These
chips are smaller than a centimeter-by-centimeter square, yet
they can hold millions of transistors. If one person
sat down to build all those miniscule parts and then connect
them, it would take a whole year. But companies turn out several
million integrated chips every few seconds -- that's about the
time it took you to read this sentence.
The reason integrated chips are possible at all is because engineers learned ways to build layers, making mllions of transistors across the chip all at the same time. The first ideas on how to build the chips were developed by Jack Kilby and Robert Noyce in 1958, and they've been developed further over the years.
The chip is built upwards, layer by layer. Each layer is made by putting masks with particular patterns over the silicon and then altering the qualities of the silicon -- or perhaps putting down metal or insulators -- in the exposed parts. It's as if you could build a house by laying down a pattern which covered the entire foundation except for where the outside walls were supposed to go. Sprinkle some bricks all over and suddenly there are walls. Next you'd lay down another pattern which has holes only where the inside walls and the furniture are supposed to be. Sprinkle wood all over the house and now there are wooden walls and tables and chairs. Other patterns might allow you to lay down porcelain for the bathrooms, pipes for the heat, and, as a final step, shingles for the roof.
Of course, the chip isn't built of wood and bricks and porcelain, it's made out of a semiconductor crystal. The chip starts out as a thin wafer of P-type silicon. This is then coated with a layer of silicon dioxide -- kind of a silicon rust, which doesn't conduct electricity. On top of this is placed a chemical called photoresist. Flashing a pattern of light (like the grid of light and dark that's formed by a window screen) on the photoresist turns any parts exposed to the light hard. The bits left in shadow stay soft. When an etching chemical is applied those soft parts, and the silicon dioxide underneath them, are removed. The hard photoresist is then dissolved, leaving a pattern of raised silicon dioxide along the surface. Since the silicon dioxide doesn't conduct electricity, it keeps different parts of the final circuit separated from others.
Following the same method, a pattern of polysilicon (which does conduct electricity and is part of the transistor) is added. Then, again using projected photoresist masks, areas of the chip are doped to become N-type silicon, another crucial part of a transistor. Lastly, metal leads are added to connect the various components on the chip.
Since the chips are so small, hundreds are made on a single silicon wafer at once. After all the patterns have been faithfully reproduced on to the chips, the wafer is sliced up into individual chips.
In day to day life, we intuitively understand how the world works. Drop a glass and it will smash to the floor. Push a wagon and it will roll along. Walk to a wall and you can't walk through it. There are very basic laws of physics going on all around us that we instinctively grasp: gravity makes things fall to the ground, pushing something makes it move, two things can't occupy the same place at the same time.
At the turn of the century, scientists thought that all the basic rules like this should apply to everything in nature -- but then they began to study the world of the ultra-small. Atoms, electrons, light waves, none of these things followed the normal rules. As physicists like Niels Bohr and Albert Einstein began to study particles, they discovered new physics laws that were downright quirky. These were the laws of quantum mechanics, and they got their name from the work of Max Planck.
"An Act of Desperation"
In 1900, Max Planck was a physicist in Berlin studying something called the "ultraviolet catastrophe." The problem was the laws of physics predicted that if you heat up a box in such a way that no light can get out (known as a "black box"), it should produce an infinite amount of ultraviolet radiation. In real life no such thing happened: the box radiated different colors, red, blue, white, just as heated metal does, but there was no infinite amount of anything. It didn't make sense. These were laws of physics that perfectly described how light behaved outside of the box -- why didn't they accurately describe this black box scenario?
Planck tried a mathematical trick. He presumed that the light wasn't really a continuous wave as everyone assumed, but perhaps could exist with only specific amounts, or "quanta," of energy. Planck didn't really believe this was true about light, in fact he later referred to this math gimmick as "an act of desperation." But with this adjustment, the equations worked, accurately describing the box's radiation.
It took awhile for everyone to agree on what this meant, but eventually Albert Einstein interpreted Planck's equations to mean that light can be thought of as discrete particles, just like electrons or protons. In 1926, Berkeley physicist Gilbert Lewis named them photons.
Quanta, quanta everywhere
This idea that particles could only contain lumps of energy in certain sizes moved into other areas of physics as well. Over the next decade, Niels Bohr pulled it into his description of how an atom worked. He said that electrons traveling around a nucleus couldn't have arbitrarily small or arbitrarily large amounts of energy, they could only have multiples of a standard "quantum" of energy.
Eventually scientists realized this explained why some materials are conductors of electricity and some aren't -- since atoms with differing energy electron orbits conduct electricity differently. This understanding was crucial to building a transistor, since the crystal at its core is made by mixing materials with varying amounts of conductivity.
But They're Waves Too
Here's one of the quirky things about quantum mechanics: just because an electron or a photon can be thought of as a particle, doesn't mean they can't still be though of as a wave as well. In fact, in a lot of experiments light acts much more like a wave than like a particle.
This wave nature produces some interesting effects. For example, if an electron traveling around a nucleus behaves like a wave, then its position at any one time becomes fuzzy. Instead of being in a concrete point, the electron is smeared out in space. This smearing means that electrons don't always travel quite the way one would expect. Unlike water flowing along in one direction through a hose, electrons traveling along as electrical current can sometimes follow weird paths, especially if they're moving near the surface of a material. Moreover, electrons acting like a wave can sometimes burrow right through a barrier. Understanding this odd behavior of electrons was necessary as scientists tried to control how current flowed through the first transistors.
So which is it - a particle or a wave?
Scientists interpret quantum mechanics to mean that a tiny piece of material like a photon or electron is both a particle and a wave. It can be either, depending on how one looks at it or what kind of an experiment one is doing. In fact, it might be more accurate to say that photons and electrons are neither a particle or a wave -- they're undefined up until the very moment someone looks at them or performs an experiment, thus forcing them to be either a particle or a wave.
This comes with other side effects: namely that a number of qualities for particles aren't well-defined. For example, there is a theory by Werner Heisenberg called the Uncertainty Principle. It states that if a researcher wants to measure the speed and position of a particle, he can't do both very accurately. If he measures the speed carefully, then he can't measure the position nearly as well. This doesn't just mean he doesn't have good enough measurement tools -- it's more fundamental than that. If the speed is well-established then there simply does not exist a well-established position (the electron is smeared out like a wave) and vice versa.
Albert Einstein disliked this idea. When confronted with the notion that the laws of physics left room for such vagueness he announced: "God does not play dice with the universe." Nevertheless, most physicists today accept the laws of quantum mechanics as an accurate description of the subatomic world. And certainly it was a thorough understanding of these new laws which helped Bardeen, Brattain, and Shockley invent the transistor.
