US3541404A

US3541404A - Transferred electron oscillators

Info

Publication number: US3541404A
Application number: US655121A
Authority: US
Inventors: Cyril Hilsum
Original assignee: National Research Development Corp UK
Current assignee: National Research Development Corp UK
Priority date: 1966-07-21
Filing date: 1967-07-21
Publication date: 1970-11-17
Anticipated expiration: 1987-11-17
Also published as: GB1205211A

Description

Nov. 17, 1970 c. HILSUM TRANSFERRED ELECTRON OSCILLATORS Filed July 21, 1967 /ILHU H n m V, w P 4 k\ WW .1 P/ A mztal or n+ FIG.

FIG. 2.

FIG 3.

/bzw 6 K m F v P M M pl 0 I. n n m m J P 5 \H m F PL 4 A United States Patent 3,541,404 TRANSFERRED ELECTRON OSCILLATORS Cyril Hilsum, Malvern, England, assignor to National Research Development Corporation, London, England v Filed July 21, 1967, Ser. No. 655,121 Claims priority, application Great Britain, July 21, 1966, 32,773/ 66 Int. Cl. H01l 11/00 US. Cl. 317-435 2 Claims ABSTRACT OF THE DISCLOSURE A transferred electron oscillator includes means for creating a depletion layer in the transferred electron effect material. The means comprises an additional electrode adjacent to the conventional electrodes which is connected to a voltage source for reverse biasing it with respect to the conventional electrodes.

The present invention relates to transferred electron oscillators.

The well-known transferred electron eifect is the effect by which pieces of certain semiconductors such as cadmium telluride, gallium arsenide and indium phosphide which may exist in two difference resistance states depending on the voltage gradient applied across them may lose their homogeneity when the voltage gradient is between certain limits and exhibit two resistance states in series within the same piece of material. A device exhibiting this effect has been proposed as a microwave generator at frequencies of 30 mc./s. up to K band, depending upon the dimensions of the device (the smaller the length, the higher the frequency).

It is apparent that it is not always an advantage for the precise frequency of operation to depend upon the physical dimensions of the device. For example, in some applications it may be convenient for a relatively long device to work at relatively high frequencies in order that heat may be more easily dissipated. Alternatively it may be convenient for the frequency of a device to be varied.

A transferred electron oscillator typically includes an anode and a cathode connected to a piece of material which exhibits the transferred electron effect, whereby when the correct voltage is applied across the electrodes a high resistivity domain is formed in the material and this domain travels from the cathode to the anode, a further domain then forming at the cathode. Such an oscillator is described in the specification of United Kingdom patent application No. 5434/66.

According to the present invention there is provided a transferred electron oscillator including a first electrode and a second electrode connected to a piece of material which exhibits the transferred electron effect wherein the improvement comprises means for creating a depletion layer in said material in the path between said first electrode and said second electrode, said means including a third electrode in the neighbourhood of said first electrode and said second electrode and means for reverse biasing said third electrode relative to said first electrode and said second electrode.

Embodiments of the invention will be described by way of example with reference to the accompanying drawings, in which FIGS. 1 and 2 are a cross-sectional diagram and a plan view respectively of a transferred electron oscillator;

FIG. 3 is a plan view of an alternative transferred electron oscillator;

FIG. 4 is a cross-sectional diagram of a further alternative transferred electron oscillator; and

FIGS. 5 and 6 are a cross-sectional diagram of a plan 3,541,404 Patented Nov. 17, 1970 view respectively of a further transferred electron oscillator.

FIGS. 1 and 2 are a cross-sectional diagram and a plane view respectively of a transferred electron oscillator. A layer M of n-type semiconducting material is epitaxially deposited upon a semi-insulating substrate S. A suitable material is gallium arsenide. Two electrodes viz, an anode A and a cathode K are either metallic or diffused n+ contacts in the layer M of an electrode P between the anode A and cathode K is a diffused or alloyed p-type region.

The action of the oscillator is conventional except for the provision of the electrode P. When this is reverse biased relative to the cathode K a depletion layer is created across most of the anode-cathode path. The depletion layer will not be of uniform thickness along the whole path since the bias will not be of uniform thickness along the whole path since the bias between the electrode P and the anode A is less than that between the electrode P and the cathode K. The narrowest path, and hence the point of highest electric field will move slightly as the reverse bias changes and this will change the frequency of operation.

FIG. 3 is a plan view of an alternative, more sensitive, transferred electron oscillator. In this arrangement the electrodes A and K are set at an angle to each other in plan with the electrode P set across the end where the electrodes A and K are closest to each other.

The operative end of this device is the position where the two contacts are closest to each other since it ishere that the electric field is highest. Since the electrode P is near to the area of closest spacing, the reverse bias will create a depletion layer which drives the operating region away to a position where the spacing between electrodes A and K is greater. Thus the frequency falls.

An alternative arrangement is shown in FIG. 4, where a layer M of n-type material is epitaxially deposited upon a p-type substrate P. The anode A and cathode K are diffused or alloyed into the opposite side of the layer M.

In this arrangement the substrate itself acts as the frequency controlling contact creating and controlling a depletion layer by a reverse bias in a similar Way to the frequency controlling contact P in FIGS. 1 and 2.

FIGS. 5 and 6 are a cross-sectional diagram and a plan view respectively of a further transferred electron oscillator. A layer M of n-type semiconducting material is epitaxially deposited upon an n+ type substrate which acts as the anode. The cathode K is a metallic or diffused n+ region in the surface of the layer M opposite the anode A. The electrode P is an annular p-type region made as a guard ring surrounding the cathode K.

The action is as follows. The electrode P acts in three separate ways. It isolates the oeprating parts of the oscillator from the quiescent regions outside the guard ring; this is designed to give higher efficiency. It controls the frequency by controlling the extent and shape of a depletion layer as described with reference to other embodiments above. It also provides a distributed capacitance along the oscillator which will also be voltage dependent and thus affect the frequency.

I claim:

1. A transferred electron oscillator including a first electrode and a second electrode connected to a piece of material which exhibits the transferred electron effect wherein the improvement comprises means for creating a depletion layer in said material in the path between said first electrode and said second electrode, said means including a third electrode in the neighborhood of said first electrode and said second electrode and means for reverse biasing said third electrode relative to said first electrode and said second electrode, said first electrode and said second electrode being in the same face of said piece of material and said third electrode being in the same face as said first electrode and said second electrode exhibits the transferred electron eflect wherein the improvements comprises: means for varying the frequency 10 of oscillations between said first electrode and said second electrode, said means comprising a field effect electrode on saidsame face between said first electrode and said second electrode which produces a depletion layer in said material and forms a p-n junction with said material and in which a multiplicity of alternative path lengths is formed by said first electrode and said second electrode having varying proximity and in which is bias on said field eflEect electrode varies said depletion layer in a direction lateral of said multiplicity of path lengths between said first electrode and said second electrode.

References Cited v UNITED STATES PATENTS 3,275,908 9/1966 Grosvalet 317235 3,439,236 4/1969 Blicher 317-235 3,365,583 1/1968 Gunn 317--23s FOREIGN PATENTS 1,498,778 9/1967 France;

, OTHER REFERENCES IBM Technical Discl. Bulletin, Electrical Shock Wave Device]? by Morgan, vol. 8, No. 9, February 1966, p. 1302.

Electronics Engineering, A Gunn Effect Epitaxial Device for Microwave Applications, September 1965, p.

JERRY D. CRAIG, PrimaryExaminer U.S. c1} X.R.