Transistors
What is a transistor?
Transistors are one of the basic and most important components of an electronic circuit. It is a small device made up of semiconductor materials and those semiconductor materials like silicon or germanium are doped with impurities like boron or phosphorus to enhance their conductivity. And it has two main uses, amplifying small input signals into large electrical output signals. And the other important use of a transistor is as a switch in an electrical circuit and this switch like behavior of a transistor makes it the most important component of a digital circuit. Today’s digital circuit’s like CPU’s or even a small chip in your cell phone or any electronic contains thousands on these tiny transistors packed on them and it is said that the number of transistors on chip approximately keep doubling every two years. And the most fascinating thing about these transistors is that it was invented at a telephone research lab for use as an audio amplifier and now it has become the primary ingredient of every electronic device that exists today.
A
Transistor
Source:
http://nobelprize.org/educational_games/physics/transistor/history/images/transistor.gif
What is it made of?
A transistor is made up of three layers of silicon or germanium, which are two most commonly used semiconductor materials. And as mentioned before, these materials are not pure, instead they are made impure and based on the type of impurity introduced; it can be either a p-type or an n-type semiconductor. For e.g. adding boron as impurity to silicon creates a scarcity of electrons making it a p-type semiconductor, while on the other hand, adding phosphorus (an electron rich element) as an impurity makes it an n-type semiconductor. Transistors can be made in two configuration of these layers PNP or NPN where ‘P’ stands for positive and ‘N’ stand for negative. These types of transistors are also called bipolar junction transistors because of the PN or NP junctions present in the transistor. The three layers of the transistors are named the emitter, the base and the collector. The emitter and the collector are the corner layers. And as it can be seen in the diagram below the base is sandwiched between the emitter and the collector.
How does it work?
A transistor can work as an amplifier as well as a switch. Let’s look at an example where an NPN transistor is used in as a switch. We connect the emitter to the negative terminal and the collector to the positive terminal. And no current flows from the collector to the emitter until the base is connected to the positive terminal. In this way the transistor works as switch, which can be turned on by passing current to the base and can be turned off by cutting the current supply to the base. This way of using the transistor is known as common emitter mode and there are many configurations in which a transistor can be used.
Source: http://plasticdog.cheme.columbia.edu/undergraduate_research/projects/diana_nguyen_project_files/image002.jpg
Why are transistors used in computers and
what did they replace?
As you might know, modern computers understand only binary language, which is a language having only two symbols 0's and 1's. Everything that is stored on a computer is represented in binary language. And as we have seen before, when used as a switch a transistor has two states on or off which is equivalent to 1 and 0 in binary. And this binary behavior makes it so well suited to be used to represent and store data on a computer.
Vacuum tubes worked in the same way but the performance was comparatively much slower than transistors, also transistors are much smaller than vacuum tubes. Another major advantage was transistors had much longer life than the short-lived, which solved the problem of replacing vacuum tubes again and again. And transistors did not produce heat like vacuum tubes as well as needed a lot less power to operate. All these advantages led to the huge success of transistor and vacuum tubes becoming a predecessor of transistors.
Who made it and When?
Transistor, also called a “Triode” since it evolved from single PN junction device called diode. Diodes are also semiconductor devices with two layers of doped semiconductor, one p-type and one n-type. Diodes were first made in early 1920’s by making point contact between a metal and semiconductor crystal. They were used to change oscillating signals to steady signals and vice versa, a simpler example of this is converting AC current (the type of current we get from wall outlets) to DC current (the type of current we get from batteries) and vice versa. But it wasn’t until 1940’s that the first transistor was made.
William
Shockley
Source: http://nobelprize.org/nobel_prizes/physics/laureates/1956/shockley.jpg
Transistor was invented by a team of three gentlemen at Bell Labs in Murray Hill, New Jersey. Among these three gentlemen were John Bardeen and Walter Brattain who worked for William Shockley. And the invention of transistor evolved from their research on producing extremely pure germanium crystals mixer diodes used in radars. While experimenting for that, John Bardeen and Walter Brattain noticed the amplifying effect of these semi-conductors, they observed that the output voltage was higher than the input voltage and then they went on to create a three layered device (triode or transistor). William Shockley was not there to witness this first observation, but he realized the importance of this effect and on December 16th 1947 they created the very first point contact transistor. And then they made couple of these transistors and put together an audio amplifier using these transistors and on 23rd December they showed it to their colleagues and their managers at Bell Labs who were greatly impressed by the performance of this device that was going to replace vacuum tube technology. Compared to vacuum tubes transistors were very small, used less power, and most importantly it worked instantly, didn’t require any time to warm up like the vacuum tubes.
Point
Contact Transistor
Source:
http://nobelprize.org/educational_games/physics/transistor/function/images/bild11.gif
Shockley was not the one to make the first observation but he was the one who promoted the idea to the company officials and later he designed the “junction” transistor, which was much easier produce than the point contact transistor. In 1956 he was awarded Nobel Prize for Physics for his invention of the transistor, and this award was shared with his two colleagues, John Bardeen and Walter Brattain. Soon Shockley became the founder of Shockley Semiconductor Laboratory with financial help from Beckman Instruments. He founded this organization in February 1956, where his primary goal was to make silicon transistors. Gordon Moore even said “it was Shockley who brought silicon to Silicon Valley”. Before this all transistors used germanium, but silicon had two huge advantages over germanium, one was that it could operate at higher temperature as it had a higher melting point and the other is that it is commonly available in large quantities. But the first advantage also became somewhat of a disadvantage since purification and refinement of silicon was difficult due to the high melting points. But eventually with help of research and development team, he was able to make silicon transistor and that is what gave birth to the silicon chips used in almost every single electronic device today.
Transistors to IC’s
Now that the switching device had become much smaller from the bulky vacuum tubes to the tiny transistors, it was possible to design much more complicated and larger circuits, which were not feasible with the use of vacuum tubes. More and more complicated circuits were designed to make use of transistors, and they soon reached their limitations since such complex circuits became slow due to too many connections with wires. So the components of the circuit needed to be packed closer and closer to make effective use of the circuits. And this was not possible with use of wires for connecting the components of the circuit. An employee of Texas Instruments by the name of Jack Kilby found a solution to this problem in 1958. He was working on miniaturization of his circuit and to achieve that he decided to make all the components of his circuit with the same block of semiconductor, in a way carve the full circuit on one piece of semiconductor. And this resulted in very first integrated circuit. Later on in the same year, Robert Noyce (co-founder of Intel) came up with his own integrated circuit, which solved one of the major problems that Jack Kilby’s design had. And that problem was about connecting the components together and Noyce solved it by having a layer of metal one the circuit and then get rid of most of the metal leaving only enough to make wire connections between the components. And this made the production of Integrated Circuits much more practical and feasible in large quantities.
First
chip made by Kilby
Source: http://nobelprize.org/educational_games/physics/integrated_circuit/history/images/chiphand.jpg
The Future
Since the first IC was made, the sizes of these IC’s or chips have decreased dramatically. Typical computer chip contains millions of these components. According to Gordon Moore(Intel co-founder), the number of components on a chip double every 18 months, and this is known was the Moore’s Law. And this increase in number of components increases the performance too. But it is predicted that it would be eventually be impossible to squeeze any more components into the same small chip and that there would be need for a new technology to replace silicon transistors. But this will not be happening till about 2025, which gives ample amount of time for researchers to come up with a replacement technology for silicon transistors. There are already prototypes of new technologies like optical switches, which are much smaller and faster than silicon transistors. In the near future there will be a successor to transistor technology and transistors will become an obsolete technology like vacuum tubes. But as of now, transistors are the dominant components of computer technology.
References
http://nobelprize.org/nobel_prizes/physics/laureates/1956/shockley-bio.html
http://www.time.com/time/time100/scientist/profile/shockley.html
http://nobelprize.org/educational_games/physics/integrated_circuit/history/index.html
http://computer.howstuffworks.com/diode1.htm
http://www.intel.com/education/transworks/index.htm
http://nobelprize.org/educational_games/physics/transistor/history/index.html
Gaurav Singhal – Stony Brook University – CSE 301 History of Computing
– Spring 2008