FIVE YEARS AGO, in June of 1948, the Bell Telephone Laboratories announced and demonstrated publicly a semiconductor amplifier invented by John Bardeen and Walter H. Brattain, and coined for it the name TRANSISTOR. Demonstrations were given of transistor amplifiers and oscillators functioning in telephone and television repeaters and in radio receivers.

It is interesting at this fifth anniversary to recall the words used at that time by Laboratories Vice-President Ralph Bown to tell about the invention. He said in part:

"The genesis of this device is an interesting story of fundamental research and I think I ought to begin at the beginning and give it to you in orderly sequence..."

"Scientific research is coming more and more to be recognized as a group or teamwork job. This is true not only in industrial research but to a rapidly increasing degree in academic research. In spite of this fact, there continues to be plenty of opportunity for individual work. What we have to show you today represents a fine example of brilliant individual contributions growing out of basic research in an industrial group framework..."

"A considerable number of people have been working hard on this matter to bring it to the stage you will see today. Physicists, chemists, metallurgists, engineers, laboratory and shop technicians, auxiliary and office personnel - yes, even executives have played a part."

"In our laboratory the semiconductor research work is carried on by a group under the immediate guidance of William Shockley, a well-known solid-state physicist. Shockley, while examining critically the prevailing theory of electrical conduction in semiconductors, predicted that it should be possible to control the meager supply of movable electrons inside a semiconductor by influencing them with an electric field imposed from the outside without actually contacting the material. Realizing the practical implications of such a possibility, he devised some experiments to test his hypothesis, but was unable to secure positive results. The electrons seemed to get tangled up in the surface of the material and did not behave just as anticipated. This part of the problem was tackled on a theoretical basis by John Bardeen. Bardeen developed a theory of what happened at the surface which was able to explain satisfactorily many of the observed facts and which led to further experiments carried out in collaboration with Walter Brattain."

"In the course of these experiments, Bardeen and Brattain invented the device we shall show you today. We have called it the Transistor because it is a resistor or semiconductor device which can amplify electrical signals as they are transferred through it from input to output terminals."

The announcement received little public notice - one of the most restrained send-offs in recent memory, according to Fortune magazine but its importance was quickly recognized by the electronics industry. It has taken nearly five years for the transistor to reach a point of development and of general recognition commensurate with its initial promise. In that time the family of transistors has grown. Other types than that first demonstrated have been invented and developed for manufacture, and the word transistor has become a generic term for semiconductor amplifiers. The transistor invented by Bardeen and Brattain we now call a point-contact transistor. It was the only transistor to reach practical application outside the Laboratories for nearly two years.

The point-contact transistor is by now quite well known. It consists of a very small piece of germanium with two fine wires contacting its surface, separated from each other but a few thousandths of an inch. The flow of current in one of the fine points controls the flow of current in the other, somewhat as the grid in a conventional vacuum (radio) tube controls the flow of current through the tube. But here the analogy ends. The transistor is solid, there is no vacuum to maintain. It is cold, there is no need for a heater to supply electrons. Because of this it requires very little power to be ready to operate and very little power to operate, for it is efficient. It is small and rugged. It will almost certainly have long life in service. And we are finding many ways for it to be of service.

Early Uses of the Transistor

A GREAT MAJORITY of conventional vacuum tubes are used to control or amplify small currents; in communication systems, in radio and television sets, in complicated military equipments such as radar bomb sights and computers. In such applications the transistor can be used. It does not replace the vacuum tube in the same socket and the same circuit. But with its own circuits it does the same jobs.

The point-contact transistor has been used in oscillators from very low frequencies to the very high frequencies of short-wave radio. It has been used over this wide frequency range to amplify and to control electrical signals. And it has been used, together with other very small components, to make a variety of compact electronic packages which perform a great many of the functions required in the modern electronic art.

With its virtues there are also some limitations: the point-contact transistor is more noisy in a circuit than one would like, and its ability to handle power is limited. The noise tends to restrict its use to the control of electrical signals rather than their generation or amplification and its power capacity limits it to relatively small currents.

The point-contact transistor, first member of the family, has been joined by a number of other types which have been announced during the five years under review. A photo-transistor invented by J. N. Shive has already gone into Bell System service in the card translator used in telephone exchanges for automatic routing in toll dialing. The photo-transistor is a device in which the flow of current through a point-contact on a small germanium wafer is controlled by a fine beam of light shining on a sensitive area of the germanium surface. In its telephone exchange use it conserves both space and power.

The Junction Transistor

EARLY IN THE STUDY of transistors, it was predicted by Shockley, on the basis of theory, that a junction transistor, different in structure, would have useful properties. The idea was to have in a germanium crystal a very thin region of one electrical type separating the two adjoining end regions of different electrical type. The two boundaries or junctions between the thin region and the two end regions were to serve the same functions as the two points in the point-contact device. The thing itself was simple. Shockley’s theory could predict the performance to be expected. The problem was to find ways to make it.

With metallurgists and chemists and physicists working together, such units were made. A thin layer with the necessary electrical property was produced in a germanium crystal without destroying the otherwise essentially perfect regularity of the crystal. Junction transistors produced in this way not only behave as the theory says they should; they have other remarkable properties as well. They are not noisy. In fact, they compete favorably in respect to noise with the very best vacuum tubes. They are remarkably efficient, closely approaching the maximum possible. But perhaps most interesting, they are ready to operate with as little as one-millionth the power necessary to keep an ordinary vacuum tube, with its hot cathode, in the ready condition.

For telephone uses the significance of this extremely low power drain is not hard to see. The sound power in a telephone receiver for normal conversation is comparable to the power needed by the transistor to keep it in working readiness. Conventional vacuum tubes, requiring many, many times this standby power, are used of course for special needs and in all long distance circuits where many conversations may be amplified at the same time. But they are economically unjustified in the local telephone plant or in the telephone instrument itself. With the junction transistor this is no longer the case: its minute power requirements remove that economic limitation and make it available for extensive telephone use.

"Single Crystals" of a Very Pure Material

WE HAVE passed over lightly a matter of major importance in the history of all transistors: the single crystal and the extreme purity required of transistor materials. The earliest transistors were made of germanium purified and allowed to solidify by conventional metallurgical and chemical techniques. The germanium usually solidified from the molten condition as a random collection of smaller crystals. Transistors from this poly-crystalline germanium showed erratic performance from one unit to the next. The problem was to get the germanium into one big perfect crystal. Our chemists and metallurgists did this. In fact, they have found several ways to make single crystals. This source of difficulty is now well past - all transistors are now made of single crystal germanium in which the successive layers of atoms are carefully ordered, one upon the other.

The question of purity has also been vital. Transistor action depends on the presence in the germanium crystal of a very few foreign atoms - perhaps one foreign atom for each 100,000,000 germanium atoms - and their arrangement in means that we must purify the germanium to an even higher degree of purity - then introduce the known and carefully controlled impurity where we want it. The germanium so prepared may very well be the purest material in existence.

Some Telephone Uses

IN THE LABORATORIES many possibilities are now being explored. Will telephone instruments of the future use transistors directly? Very likely, and there is active work in progress. Will transistors be used in the complicated exchanges where the telephone user is automatically connected to the party he is calling? Very likely, and very promising. Promising in terms of what can be done, speed of operation, power and space needed, as well as probable cost and reliability in service. In our automatic telephone exchanges today there are substantially no operators. But even with mechanization, electronics has had only a small part to play in the evolution of complex switching systems such as No. 4 and No. 5 cross-bar. Transistors, as well as other solid-state devices, with their capability of high speed. operation, their small bulk and low power requirements, will find natural application in switching apparatus.

Will the transistor have a role to play in distributing television pictures in communities or improving long distance communication? It will have a role to play in both. It may well be that broadband circuits for long distance uses will take new form through lower circuit mile costs that the transistor will make possible. New opportunities for extensive distribution of television in urban areas are possible. The transistor may allow the use of our carrier methods over shorter distances and lightly loaded lines. Its use in exchange areas and on rural lines cannot be excluded. A recently announced Laboratories development, the tetrode transistor of R. L. Wallace, Jr., is an important step.

The Tetrode Transistor

AN INITIAL LIMITATION of the junction transistor lay in its inability to handle the higher frequencies. This limitation for many uses is now largely removed by the invention of the tetrode transistor, which, however, retains the other virtues of its predecessor. We cannot foresee all of its ultimate uses, but already it has been used in a high-performance amplifier of minute dimensions. The entire amplifier is housed in a tiny "coaxial cable" about one-eighth inch in diameter and one and a half inches long. It requires no more power than its small size suggests, and it can readily satisfy the severe requirements of television transmission.

Research, invention, development, and application have grown at a hearty pace. But the field is still very new, and as the studies progress and understanding increases so also does the prospect. Before the first announcement, five years ago, when the new physical phenomena were observed and radically new electronic devices invented, it was evident to us that we were seeing the beginnings of an era in electronics technology of significance not only to the telephone system but also to the military services and to industry and civilian use generally.

How did we open the door to this new era?

Organized Creative Technology

I HAVE HAD OCCASION before to stress the value of continuity in the procedures from the initial steps of forward-looking basic research to the terminal operations of manufacture and use. I speak of them as the single entity: organized creative technology. No example illustrates its meaning more dramatically than the story of the transistor.

I have already emphasized the comprehensive work of the research group in studying the mechanisms of conduction of electricity in solid semi-conductors. Physicists, both experimenters and theorists, metallurgists, chemists-were joined together in a common effort to gain understanding in a field of relevance and potential importance to communication technology. The scientists published the results of their findings and frankly discussed the new phenomena with their peers.

In accord with our policy of concentrating the efforts of our research scientists on research, we immediately formed a fundamental development group under J. A. Morton. This group was closely associated with research to acquire that body of technological knowledge essential to the development and design of transistors for the many specific communications applications that would certainly follow. The fundamental development group was composed of men of electronics, skilled in the development of new electronic devices for communication and military uses. In the beginning, their interests and activities were not distinguishable from those of the research group. Their first task was to learn, and this was quickly done. However, their ultimate role was never in question. They were to apply the new knowledge from research and to supplement it in a more specific way to create new devices for our systems development groups to utilize in creating new systems both for the Bell System and for the military services that would find application because of economic or fundamental advantages.

Contributions to new knowledge were made and, as applications of transistors became more and more clear, the group turned its attentions to the development of specific designs which would be required on a large scale. They have also interested themselves in such problems as the factors controlling the band width of amplification, the noise figure, the amount of amplification possible per stage, basic materials, processing, and structure studies essential for controlled development and design of transistors for a variety of specific functions.

The link with research was strong and stimulating. A linkage with the systems development people, the likely users, was also encouraged. Jointly with the user, operating characteristics of transistors were established, units were constructed for trial, and pre-production methods were worked out.

Following the pattern used for electron tubes, the final stages of design for manufacture are carried out by the Laboratories group located at the Western Electric Company factory at Allentown, Pennsylvania. Because of the novel processes and technology used in making transistors, Western Electric sent its engineers at an early stage into laboratories work both at Murray Hill and at Allentown. Twelve different transistor types have so far been through this development sequence and have been put into pilot plant production by Western at Allentown. Meanwhile, Western has made ready at Laureldale, Pennsylvania, a production plant for military transistor manufacture. Tens of thousands of transistors of all types have been produced by Western and supplied not only for Bell System and military uses but also to various other developers and manufacturers of military equipment.

The course of development is one we well know: severe requirements to meet high standards, unsuspected difficulties, failures and success. The pattern continues, and in taking stock we can say that while there are probably as many problems ahead as have already been solved, transistors are built on an unusually firm scientific foundation; and they have been developed to a status of flexible utility, reproducibility, and reliability. They are now made of material of crystalline perfection and remarkable purity, produced through the ingenious processes of our chemists and metallurgists. Their service life in proper use we expect to be very long. They are reproducible in the same sense that vacuum tubes are reproducible and as the art of their manufacture matures, they should be inexpensive.

Apparatus Components for Use with Transistors

A NEW DEVICE or a new invention stimulates and frequently demands other new devices and inventions for its proper use. This is particularly true of the transistor. Its circuit technology, its efficiency and low power requirements we have already mentioned. Operating voltages in most applications are remarkably low, perhaps 1/20^th the voltage required by ordinary vacuum tubes - only a fraction of the voltage readily available in telephone central offices.

At these low voltages, capacitors will be made and are now under development which are as much smaller than ordinary capacitors as the transistor itself is smaller than a vacuum tube. In telephone communication uses, with low voltages and low currents, coils and transformers may be correspondingly miniaturized. A whole series of new miniature components is now being developed in the Laboratories for use with the transistor - to take full advantage of its modest requirements.

Military Uses

BELL LABORATORIES ACTIVITY in transistors is primarily in two directions: to serve the needs of the Bell System and to contribute to the military strength of the country. It was recognized at the outset that secrecy would severely hamper both of these objectives by restricting the free and rapid flow of ideas and delaying broad use of the new technology in military applications. A policy was adopted to make transistor information easily available to those with a legitimate need for it. This policy has been implemented both through our usual practice of early publication of scientific work and through patent licenses offered by the Western Electric Company.

To expedite the availability of information, there was held at the Murray Hill laboratory in September 1951, with the cooperation of the Military Services, a weeklong symposium on the characteristics and applications of the transistor. Some 300 engineers from our own and Western European countries attended. Thirty-five papers by members of our staff were presented, and later issued as an 800 page volume.

In April of 1952, a symposium on transistor technology - devoted to the dissemination of "know how" - for licensees of the Western Electric Company was held at Murray Hill and at the Western Electric Company plant in Allentown. Representatives from 26 domestic and 14 foreign organizations attended intensive sessions lasting eight days. The material presented, covering the entire range of technology, was subsequently published as a two-volume classified work. Later still, in the summer of 1952, a short course for university professors was held, attended by professors from over thirty institutions. The result has been widespread development both in university research and in the industry.

Specific military applications of the transistor must, of course, remain classified, but an appreciation of its promise in military use can still be had, as the following example shows. In World War II, radar played a very important part. Today, aircraft bomb sights are built around radar. Detection of targets, range, and bearing all come from radar. This information feeds into a "computer" which quickly determines the proper release point, taking into account other information as well: aircraft and wind velocity, altitude, and the flight characteristics of the missile. The radar and computer make a highly intricate electronic system, using hundreds of vacuum tubes and considerable power, and occupying much valuable space. Perhaps 80 per cent of the vacuum tube functions may equally well be performed by transistors, with savings in space, weight, and power alone of major significance to our air forces.

In many other parts of our complex military technology there are uses, large and small, for transistors. Reliability, small size and weight, low power consumption - these are goals of the designers of military equipment. The transistor takes us a large step towards these goals.

Future Developments

TODAY’S ELECTRONICS INDUSTRY is built largely on an invention of Lee deForest in 1907: the three-element vacuum tube. The first commercial use of the vacuum tube in transcontinental telephony was in 1915, eight years later. The vacuum tube had extensive development for military purposes in World War I, and was used in telephone carrier in 1918.

The transistor is very young and already it is making its mark. Its impact will be felt not only in replacement of vacuum tubes for reasons of economy. It will make completely new fields for itself as it begins to do work where vacuum tubes have so far been excluded. Ultimately we shall see transistors at work automatically routing and connecting telephone calls through exchanges, doing accounting and computing, and performing in other so-called "electronic brains." They are now appearing in hearing aids, and we shall soon see them in home radio and television sets.

The transistor, with all its promise, will come into large scale use in the Bell System only gradually. Other fields of application - military electronic systems, home entertainment, special services - may well have the larger initial uses. For in these fields, particularly the military, it will now be possible to do things, urgent things, with transistors which have heretofore been prohibited by the size, weight, power requirements, or lack of reliability of alternative methods.

This is not the case in most telephone applications. The Bell System plant is a carefully integrated complex of instruments, transmission paths, and interconnecting mechanisms which has been evolved over the years to provide service at low cost and with great reliability. The transistor will emerge as an important part of our telephone technology only as rapidly as we gain sound experience with its practical ability to meet the requirements of the present system. New services and new methods of providing present services are being examined continuously from a systems standpoint. The transistor gives us a powerful new tool in these studies.

A new technology is growing around the transistor; a new industry will grow with it. Bell Laboratories, continuing in the forefront, will realize for the Bell System and the military services the promise of these five years.