US2854588A - Current multiplication transistors - Google Patents

Description

p 1958 R. w. LANDAUER 2,854,588

CURRENT MULTIPLICATION TRANSISTORS Filed Dec. 23. 1953 INVENTOR. ROLF W. LAN DAUER ATTORNE United States Patent 9 2,854,588 CURRENT MULTIPLICATION TRANSISTORS Rolf W. Landauer, Poughkeepsie, N. Y., assignor to International Business Machines Corporation, New York, N. Y., a corporation of New York Application December 23, 1953, Serial No. 399,996 4 Claims. (Cl. 307-885) This invention relates to transistors, especially to transistors of the current multiplication type, and to methods of producing such transistors and controlling their current gain.

The current multiplication or point contact type of transistor comprises a small block of semi-conductive material, to which are applied at least three electrodes or contacts. Where only three electrodes are used, they are respectively termed the base, collector and emitter electrodes. The base contact is conventionally ohmic, i. e., its resistance is independent of the direction and magnitude of current flow. The base contact usually has a substantial contact area. The emitter and collector electrodes are point contacts and have rectifying or asymmetric impedance characteristics, i. e., their impedance is different for opposite directions of current flow.

The semi-conductive material in the body may be either n-type or p-type. The n-type semi-conductive material contains impurities providing an excess of electrons which are free to move for the carrying of electric currents, whereas p-type semi-conductive material contains impurities which result in a smaller number of electrons than are present in the pure semi-conductive material. The spaces in the lattice structure which in the pure material would be occupied by the missing electrons are called holes and can be considered to act like movable positively charged current carriers.

For most point contact transistors the semi-conductive material used is n-type germanium. When potentials are properly applied between the base and each of the other two electrodes, a translating device is produced wherein variations in current in the input circuit (usually the emitter-base circuit) cause variations in current in the output circuit (usually the collector-base circuit).

There is disclosed and claimed in the copending application of Richard F. Rutz, Serial No. 354,955, filed May 14, 1953, entitled Method for Improving the Current Amplification of a Transistor, a transistor of the point contact type having a third point contact, which functions as a second emitter electrode, providing an auxiliary reservoir of excess current carriers. The second emitter electrode as disclosed by Rutz provides certain substantial advantages in the transistor characteristics, particularly an increased current amplification factor. The transistor disclosed by Rutz is, however, subject to certain limitations. One such limitation is that the spacing between the second or auxiliary emitter and collector must be equal to or greater than the diffusion length for the average lifetime of the excess carriers in the semi-conductor material. In order to reduce this necessary spacing, it is therefore desirable to construct transistors of the Rutz type from short lifetime germanium.

Another difficulty with the Rutz transistor is an undesirable decrease in the back resistance (resistance to current flowing from base to collector) under certain conditions of operation. This decrease in the back resistance is noted particularly if any attempt is made to re- Patented Sept. 30, 1958 duce the spacing between the auxiliary emitter and collector electrode below the diflfusion length. It is highly desirable, especially in transistors intended for use in switching circuits, that the back resistance be maintained at a high level throughout the transistor operation.

An object of the present invention is to provide an improved transistor having a high current amplification factor.

Another object is to provide an improved transistor of the type described in which the high current amplification factor is secured by a mechanism similar to that disclosed by Rutz, namely an auxiliary reservoir of excess carriers.

Another object of the invention is to provide a transistor of the type described in which the reservoir of excess carriers is controlled at all times.

The foregoing objects are attained by provision of a transistor having two emitter electrodes and two collector electrodes. Only one of the two collector electrodes serves as the output electrode of the transistor. The first emitter is more or less conventional, being preferably located very near the output electrode. The second emitter is located at a point substantially equidistant from the two collectors. The distribution between the collectors of carriers flowing through the second emitter is controlled by the flow of carriers from the first emitter. The potentials of the several electrodes may be controlled as desired to regulate the current amplification factor.

Other objects and advantages of the invention will become apparent from a consideration of the following specification and claims taken together with the accompanyin g drawing.

In the drawing, the single figure represents, somewhat diagrammatically, a transistor embodying the invention and associated circuit elements.

Referring to the drawing, there is shown a transistor including a body 1 of semi-conductive material having a base electrode 2, two emitter electrodes 3e and 4e, and two collector electrodes 50 and 6c. The body 1 is illustrated as being of n-type semi-conductive material, and the directions of current flow and the various polarities correspond to the requirements of that material. Those skilled in the art will readily recognize that p-type material may alternatively be used, with consequent reversal of the current directions and polarities.

The emitter electrode 3e is located mid-way between the collector electrodes 50 and Go. In other words, the distance x between emitter 3e and collector 5c is made substantially equal to the distance y between emitter 3e and collector 6c. While these three electrodes are preferably located in one straight line, as indicated in the drawing, it is only necessary that the distance x be substantially equal to the distance y.

The emitter electrode 4e is located between emitter 3e and collector 6c, and preferably very close to collector 6c, the spacing there being substantially the same as between the emitter and collector in conventional transistors.

The base electrode 2 is connected to ground. The emitter electrode 3e is connected to ground through a resistor 17 and a biasing battery 7. The emitter electrode 4e is connected to ground through a signal generator 8 which is shown by way of example as including an A. C. signal generator 9 and a D. C. signal generator 10, indicated as a battery. The direction of current flow from emitter 4e into the body is the direction usually associated with an emitter. Therefore the emitter 4e must be positive with respect to the semiconductor material immediately adjacent, but it does not have to be positive with respect to the base 2. If collector 6c is negative and emitter 4e very close to it, it might be necessary to bias emitter 4e negatively with respect to the grounded base 2.

The collector electrode 50 is connected in series with a battery 11 and a resistor 12 to ground. The collector 6c is connected in series with a load resistor 13 and a battery 14, to ground. Output terminals 15 and 16 are respectively connected to the collector 6c and to ground.

Operation In the following discussion, the operation of the transistor will be discussed in terms of holes and electrons, in a manner consistent with the assumption noted above, that the boy 1 is of n-type material. That is to say, instead of using the generic terms majority carriers and minority carriers, reference will be made to electrons and holes, respectively. It is considered that the use of this more specific terminology will make the explanation more concise and more readily understandable. It should be understood, however, that it is not intended, by the use of this terminology, to limit the invention to n-type semi-conductive material.

In any transistor, the hole current flowing from the emitter to the collector produces a concomitant electron current flowing from the collector to the base. The collector cuirent is the sum of the hole current and the electron current, and is therefore greater than the hole current by a factor commonly designated (1* or a and termed the intrinsic current amplification at the collector, or sometimes called the collector multiplication factor. Due to an amplifying mechanism whose exact nature is not material to the present discussion, each hole reaching the collector may produce a flow of several electrons from the collector, so that the intrinsic current amplification may be very high.

It has been found by experiment that when a single emitter is placed at a point substantially half-way between two similar collectors, the current carriers from the emitter do not divide equally between the collectors, as might be expected, but instead most of the current tends to go to one collector or the other. It is considered that this phenomenon is due to a cumulative effect. To understand this effect, consider that in any physical transistor it is a practical impossibility to place an emitter electrode exactly half-way between two collectors; likewise, it is impossible to bias the two collectors with exactly equal potentials; similarly, it is impossible to give two collectors identical forming treatments so as to obtain equal current amplification and equal back resistances at both collectors. Consequently, in any physical set-up, there is always a. slight unbalance in favor of one collector. That slight unbalance produces an unbalance in the electric field within the semi-conductive body, which unbalance favors one collector and tends to attract more than its share of carriers from the emitter. If the total emitter current is constant, this additional emitter current flowing to one collector must be taken away from the unfavored collector. These changes in hole currents flowing to the collectors in turn produce changes in the electron currents flowing away from the collectors. If the intrinsic current amplification (o8 or a at each collector is big enough, then these changes in electron current overcome any changes in conductivity due to the injected holes and the electric fields emanating from each collector will change in the same directions as the electronic currents. Thus, the favored collector will have a larger field emanating from it and the unfavored collector a smaller field.

These changes in field will tend to produce a further unbalance in the division of the incoming emitter current. This in turn will produce a change in the fields emanating from the collectors, and so forth.

In the transistor illustrated in the drawing, the situation just described exists with respect to the emitter 3e and the collectors c and 60. These two collectors should be made to have characteristics as nearly equal as possible. This symmetrical situation is disturbed by the presence of the additional emitter 42 located nearer the collector 60 than is the emitter 3e. The bias potentials on the two collectors, or other controllable factors, should be selected so that when there is no current flowing through the emitter 42, most of the holes from the emitter 3e flow to the collector 50, with a small current flowing through collector 60. As soon as a current starts to flow from the emitter 4e, it is attracted to the collector do by the electric field of that collecor. This hole current, as it arrives at the collector, increases the electron current from the collector in a ratio determined by the intrinsic amplification factor of, previously mentioned. This increased electron flow increases the electric field of collector 60, which increased field attracts more of the holes from emitter 3e, and the increased flow of holes further amplifies the electron flow from collector 66.

This attraction of holes from the collector 3e by the electric field due to the electron current flowing from emitter 4e may be described as an internal feedback mechanism. This mechanism is similar, if not identical, to that present in the transistor disclosed in the Rutz application previously mentioned.

The present transistor has, however, another internal feedback mechanism operating which further enhances its overall current amplification factor. Referring to the collector 50, it may be seen that as the proportion of holes from emitter 3-2 flowing to this collector gets smaller.

the electron current flow from collector 5c is reduced, the electric field due to that electron flow decreases and collector 5c consequently tends to receive an even smaller proportion of the holes from emitter 3e.

It may therefore be seen that a transistor constructed as described above has an overall current amplification factor even greater than that of the Rutz type of transistor. Furthermore, it has been observed that the back resisatnce of the transistor described above is maintained better than is the back resistance of the Rutz transistor.

It should be further noted that the transistor described herein is not subject to any limitation with regard to the lifetime of the excess carriers in the semi-conductive ma terial, nor as to the diffusion length. The electrode spacings should, on the other hand, be smaller than the diffusion length, rather than greater as in the Rutz transistor.

It is desirable to keep a high bias voltage on the emitter 3e, and a high resistance in series with it so that there is a plentiful, but constant, supply of current carriers continuously available from that emitter. The bias voltage on the emitter 3e and that on the collector Sc may be made variable, as indicated in the drawing, in order that the current amplification factor can be controlled as desired.

While a resistor 12 is shown in series with battery 11, this resistor may in many cases be omitted, and indeed it is preferable to omit it. It might be expected that the best balance between collectors 5c and 6c would be obtained by making resistor 12 equal to resistor 13 and the potential of battery 11 equal to that of battery 14. However, as explained above, an accurate balance between the emitters cannot be practically attained, and is not in fact utilized in the operation of the transistor, except as a transitional condition during a transfer from an unbalance favoring one collector to an unbalance favoring the other. The desired operation then is to pass as quickly as possible from one condition of unbalance to the other. Resistors 12 and 13, like all resistors, have a tendency to maintain the current flow through them constant, and therefore tend to be detrimental to the desired operation of the circuit. The load resistor 13 must be retained in the circuit, but resistor 12 may readily be omitted. If resistor 12 is omitted, then in order to obtain nearly equal voltages at the collectors 5c and 6c, the potential of battery 14 should be made greater than the potential of battery 11, in order to compensate for the potential drop across resistor 13.

While I have shown and described a preferred embodiment of my invention, other modifications thereof will readily occur to those skilled in the at, and I therefore intend my invention to be limited only by the appended claims.

I claim:

1. A current multiplication transistor circuit, comprising a transistor having a body of semi-conductive material of uniform conductivity type, a first electrode in ohmically conductive contact with one side of said body, second, third, fourth and fifth electrodes in asymmetrically conductive contact with the opposite side of said body, a load resistor and a first source of unidirectional electrical energy connected in series between said first and second electrodes, said source being poled to reversely bias said second electrode, a second source of unidirectional electrical energy, means connecting said second source between said third electrode and said first electrode, said second source being poled to bias said third electrode reversely, said fourth electrode being located substantially equally distant from said second and third electrodes, a third source of unidirectional electrical energy, means connecting said third source between said first and fourth electrodes, said third source being poled to bias said fourth electrode forwardly, said fifth electrode being located closer to said second electrode than said fourth electrode, and a source of variable input signal potential connected between said first and fifth electrodes, said third source of energy cooperating with said fourth electrode continuously to introduce minority current carriers into said semi-conductive body, said first and second sources biasing said second and third electrodes so that they competitively attract said minority carriers, said second and third electrodes and their respective connections being unbalanced in favor of said third electrode so that it tends to attract the greater proportion of said carriers, said signal source being operable at times to inject additional minority carriers into said body, said second electrode being thereupon effective to attract substantially all said additional carriers, the flow of said additional carriers being effective to modify the electric field in said body to reverse the unbalance in favor of said third electrode and to establish an unbalance in favor of said second electrode, thereby further increasing the minority carrier flow to said second electrode and thereby greatly increasing the current multiplication at said second electrode, and output means connected to said second electrode.

2. A current multiplication transistor circuit as defined in claim 1, in which said means connecting said second source between said third electrode and said first electrode comprises direct connections substantially without impedance.

3. A current multiplication transistor circuit as defined in claim 1, in which said means connecting said third source between said fourth electrode and said first electrode comprises high impedance means effective to maintain the current flow through said fourth electrode substantially constant.

4. A current multiplication transistor circuit as defined in claim 1, in which at least one of said second and third sources is variable in potential to control said increased current multiplication.

References Cited in the file of this patent UNITED STATES PATENTS 2,592,683 Gray Apr. 15, 1952 2,655,607 Reeves Oct. 13, 1953 2,660,624 Bergson Nov. 24, 1953 2,679,619 Grassl May 25, 1954

Images