...

Introduction

The changes in a computer technology rapidly increased through few decades which is heavily influenced by the introduction of CPU transistor. Nowadays beside computers everything has a CPU such as mobile phone, smart TVs, plenty of standalone electronic devices with a cheapest price. After the replacement of an older vacuum tubes technology by a transistor, the microprocessor or CPU technology grows rapidly. Since the vacuum tubes were unreliable, bulky and generated a lot of heat, too. Especially for computer technology Vacuum tubes were unpractical.
Scientist were brainstorming the idea of semiconductor technology for a long time until the invention of the transistor announced by the Bell Telephone Laboratories in 1948.
Since then many types have been designed. Transistors are very cheap, durable, and small and have a high resistance to physical shock. The vast majority of transistors now are built as parts of IC. Transistors are used in virtually all electronic devices, including radio and television receivers, computers, and space vehicles and guided missiles.
Today's computer technology is the modest achievement in human History. The reason for the radical grows in computer technology is the invention of semiconductor transistor from germanium and later from silicon. Imagine the most abundant element from the earth's soil, silicon dioxide, the silicon transistor have made today's modernization real.
In this report the history of semiconductor, the invention and improvement of semiconductor transistor throughout the last seven decades are presented. Additionally the main type of transistor and application areas will be discussed. Beside these theoretical background the main goal of this report is to present the CPU technology with respect to transistor technology.

...

History of Transistor

...

Overview of Semiconductor and Transistor History

During World War II most scientists are occupied by war related technology such as radar. When a war ended most military related laboratories disbanded and scientist returned to researches other than military efforts. Right after the war, In January 1946, Marvin Kelly put together a group of engineers and physicist at Bell Labs to create a solid state electronics. The team included Walter Brattain, John Bardeen, John Pearson, Bert Moore, and Robert Gibney headed by Bill Shockley and Stanley Morgan [1].
Right at the beginning the team made important decision and effort directing to the two simplest semiconductor silicon and germanium.  Additionally Shockley independently revived the idea of a field-effect device. They started investigating the nature of surface states and how to eliminate the effects. If the context of the knowledge of technology and science at the time of discovery considered, transistor is the greatest discovery in the history of human modernization. [2]
There are few invention that contributes to the discovery of transistor at Bell Labs

...

While those team of scientist tried to invent a better solid effect rectifier and amplifier electronics to replace vacuum tube transistor is discovered for the first time. Finally by late 1947, Bardeen and Brattain managed to make the first working point-contact transistor. Figure 1 shows the first transistor: [3]


Figure. 1 shows the first transistor (reprinted from [1].)
Figure 1 shows the first complicated transistor with germanium crystal base and two leads formed on the tip of the germanium crystal. The tip a metal coat, wax and another metal coat layer on it. The inside metal was the collector and outside metal was the emitter. The wax in the middle is a layer of insulation [1].

Figure 2. Schematic diagram of the first

...

The invention of the first Junction Transistor

Shockley, great creative burst, proceeded to write down the theory of the bipolar junction transistor, by injecting minority carrier into a semiconductor. John Shive improved the first design by putting the emitter and collector on the opposite sides of the crystal to eliminate surface paths between collector and emitter. John Shive experiment verified shockley's junction transistor theory. [4]
By April 1950, a couple of years later, Brattain, Shockley, Teal, and Sparks actually succeeded in growing the first junction npn device. In fact, the device behaved essentially as predicted by Shockley's theory. Figure. 4 is a picture of this device. The big problem, of course, in making a true bipolar semiconductor device was the need for a very thin base region. As we all know, today the base has to be on the scale of micrometers.

Figure 3. The first Junction transistor (reprinted from [1].)
By 1952, a Bell Labs team had developed a means of making high purity silicon and germanium crystals. This was done by a process called zone refining that was invented by Bill Pfann.

...

The invention of the First FET transistor

By 1952, Ian Ross and George Dacey succeeded in making a unipolar device. This first unipolar device was a precursor to today's FET. This configuration was made using junctions as gates rather than having the metal oxide gate structure that we have today. This junction FET worked in a pinch-off mode rather than enhancement or depletion mode as in a planar insulated gate device. [5]
Discovering that the natural oxide of silicon that occurs when you put it in a high oxygen environment has a tremendously good interface between it and pure silicon. The silicon dioxide-silicon interface is sufficiently free of surface states that you can actually make an FET. In 1955, Duane Kahng joined Bell Labs at about that time, and he fabricated the first field-effect transistor using Atalla's oxidation process. But, this turned out to be a pretty poor device.
It took until the early 1970's, 15 years before planar FET's came into common use. The delay was due to the difficulties encountered controlling impurities. This was a materials problem, and for a long time people did not realize that sodium was the killer. Specifically, any sodium at the interface between silicon and silicon dioxide had devastating effects. It was not understood how to isolate the devices from the sodium. It was a major problem to understand the purification of this interface. As you know, we now have the very highest quality silicon-silicon dioxide interface. Today we have impurities that are less than one part in one billion, which is a tremendous accomplishment. [6]

...

The Invention of the First Integrated Circuit

In 1957, Texas Instruments developed the mesa transistor. Then came some very important events in the late 1950's. Jack Kilby of Texas Instruments developed the first IC using these mesa techniques. Kilby used discrete wire interconnection. See Fig. 4 for a picture of that device. You can see it really was a simple device by our standards today. It has one transistor, a capacitor, and resistor all together on a piece of silicon. This is the first integrated circuit, not really large scale integration. The next thing that happened was at Fairchild, where Jean Hoerni developed the planar process for transistors. In particular, the planar process offered the capability for doing thin-film metal interconnection. Bob Noyce, using this process, made an IC using vapor deposited metal connections, which became public in 1959. [7]


Figure 4. Jack Kilby's first integrated circuit (reprinted from [10].)

...

Types of Transistor

Transistor can be categorized based on semiconductor material used, structure, and power rating operation frequency, amplification factor, electrical polarity or application. Although one of the above basis categorizes transistor, the major types falls to junction transistor and field effect transistor (FET) categorized mainly based on the structure.
The transistor is an arrangement of semiconductor materials that share common physical boundaries. Materials most commonly used are silicon, gallium-arsenide, and germanium, into which impurities have been introduced by a process called "doping." In n -type semiconductors the impurities or dopants result in an excess of electrons, or negative charges; in p -type semiconductors the dopants lead to a deficiency of electrons and therefore an excess of positive charge carriers or "holes" [8]. 

Junction Transistor

...

The transferred resistance or transistor is a multi-junction device that is capable of Current gain, Voltage gain, and Signal power gain Invented in 1948 by Bardeen, Brattain and Shockley. Contains three adjoining, alternately doped semiconductor regions: Emitter (E), Base (B), and Collector (C) The middle region, base, is very thin compared to the diffusion length of minority carriers Two kinds: npn and pnp.

The Bipolar junction transistor is an active device that works as a voltage controlled current source and whose basic action is control of current at one terminal by controlling voltage applied at other two terminals. Emitter is heavily doped compared to collector. So, emitter and collector are not interchangeable. The base width is small compared to the minority carrier diffusion length. If the base is much larger, then this will behave like back-to-back diodes.[9]

Figure 4. NPN and PNP Junction transistor 

FET transistor

...

FET transistor commonly called as unipolar transistor since it contains one type of carrier electrons or holes (unipolar). The conventional bipolar transistor has two type of current carriers of both polarities (majority and minority) and FET has only one type of current carriers, p or n (holes or electrons). The BJT is current controlled and FET is voltage controlled current between two other terminals.

Field effect transistor is a unipolar transistor, which acts as a voltage controlled current device and is a device in which current at two electrodes is controlled by the action of an electric field at another electrode. Field effect transistor is a device in which the current is controlled and transported by carriers of one polarity (majority) only and an electric field near the one terminal controls the current between other two.
Family of FET 

Junction FET (JFET)

...

JFET is a unipolar transistor, which acts as a voltage controlled current device and is a device in which current at two electrodes is controlled by the action of an electric field at a pn junction. In addition to the channel, a JFET contains two ohmic contacts: the source and the drain. The JFET will conduct current equally well in either direction and the source and drain leads are usually interchangeable.

JFET consists of a piece of high resistivity semiconductor material (usually Si) which constitutes a channel for the majority carrier flow and a gate. Conducting semiconductor channel between two ohmic contacts source & drain. The magnitude of this current is controlled by a voltage applied to a gate, which is a reverse biased. (Ohmic contacts means following Ohm's law current proportional to V under constant physical condition. [9] 

MOSFET

...

Field effect transistor is a unipolar transistor, which acts as a voltage controlled current device and is a device in which current at two electrodes drain and source is controlled by the action of an electric field at another electrode gate having in between semiconductor and metal very a thin metal oxide layer.[9] 

CPU Transistor Technology

...

CPU architecture design, implementation and form have changed through time. However fundamental operation principle remain almost unchanged. CPU has the following most principal components:

...

All these principal components are made from different types of transistor in order to perform their objectives.
Nowadays modern CPU are Microprocessor. A microprocessor is a single integrated circuit or IC chip that contains CPU, memory, peripheral interfaces and other components. The most interesting change in CPU and computer technology over the course of history are abundancy and the changes in transistor count.

...

Abundancy

The term abundancy especially in transistor technology it is impressing in two ways. Firstly transistor is made from the most abundant compound called silica (silicon dioxide) which is 59% of earth's crust and the main constituent of more than 95 percent of the known rocks. Who would imagine that the most abundant component of the earth crust is a tool to create transistor and the most interesting reason for today's modernization? Secondly when the transistor count per inch increased, transistor abundancy per IC chip made it possible to change the CPU and computer technology architecture design and implementation to grow rapidly.

...

...

Transistor Count

Transistor count is the most common tool to measure the size of IC chip. It approximately follows the most known Moore's law. According to Moore's Law, the transistor count of the integrated circuits doubles approximately every two years. The table in appendix one shows the change in microprocessor technology, which directly related to CPU technology, with respect to transistor count through history.

...

Conclusion

While writing this report I have learned the technological changes in transistor history. Beside the change in physical characteristics of transistor, the most important changes are toward the count of transistor. Changes in transistor technology has a direct effect on CPU technology, therefore it can be concluded that the reason for modern computer system improvement is also the change in transistor technology.

...

...

References

...

1. William F. Brinkman, Douglas E. Haggan, and William W. Troutman, "A History of the Invention of the Transistor and Where It Will Lead Us" IEEE journal of solid-state circuit, vol. 32, NO. 12, DEC 1997.
2. M. Riordan and L. Hoddeson Crystal Fire the Birth of the Information Age. New York: Norton, p. 102.
3. "A History of Engineering and Science in the Bell System—Electronics Technology (1925–1975)," p. 2, F. M. Smits, Ed
4. W. Shockley, M. Sparks, and G. K. Teal, "P-N junction transistors," Phys., Rev. 83, July 1, 1951, pp. 151–162.
5. G. C. Dacey and I. M. Ross, "Unipolar field-effect transistor," Proc. IRE 41, Aug. 1953, pp. 970–979.
6. D. Kahng and M. M. Atalla, "Silicon-silicon dioxide field induced surface device," presented at Solid State Device Research Conf., Pittsburgh, PA, June 1960.
7. J. A. Hoerni, "Planar silicon diodes and transistors," IRE Trans. Electron Devices, Mar. 8, 1961, p. 178; also presented at Professional Group on Electron Devices Meeting, Washington, D.C., Oct. 1960.
8. Types of Transistor
   URL: http://www.infoplease.com/encyclopedia/science/transistor-types-transistors.html
   Accessed date: 20 April 2015
9. Transistor and FET Characteristics
   URL: http://www.dauniv.ac.in/downloads/Electronic%20Devices/09EDCJFETLesson09.pdf
   Accessed date: 20 April 2015
10. Kilby's Solid Circuit
    URL:http://en.wikipedia.org/wiki/Integrated_circuit#/media/File:Kilby_solid_circuit.jpg
    Accessed date:21 May 2015