Abstract: A peripheral structure for a monolithic power device, preferably planar, includes front and rear surfaces, connected respectively to a cathode and an anode, two junctions respectively reverse-biased and forward-biased when a direct and adjacent voltage is respectively applied to the two surfaces and at least an insulating box connecting the front and rear surfaces. The structure is such that when a direct voltage or a reverse voltage is applied, generating equipotential voltage lines, the insulating box enables to distribute the equipotential lines in the substrate.

Abstract: A structure of a cross guard ring along the edge of a semiconductor chip is disclosed. A first guard ring, a second guard ring and a third guard ring are formed along the edge of a semiconductor chip. Each guard ring comprises several rectangle shaped vias which are positioned along the edge of the chip structure, wherein each rectangle via is separated from an adjacent rectangle via by a gap. Further, each rectangle via of the second guard ring is positioned opposite the said gap of the first guard ring and are crossed over and have some overlay with rectangle vias of the first guard ring which are separated by the said gap as shown in FIG. 2. Similarly the third guard ring is positioned with respect to the second guard ring.

Abstract: A silicon-based semiconductor component includes a high-efficiency barrier junction termination. In the semiconductor component, a silicon semiconductor region takes on the depletion region of an active area of the semiconductor component. The junction termination for the active area is formed with silicon with a doping that is opposite to that of the semiconductor region, and the junction termination surrounds the active area on or in a surface of the semiconductor region. The junction termination is doped with a dopant that has a low impurity energy level of at least 0.1 eV in silicon. Preferably Be, Zn, Ni, Co, Mg, Sn or In are used as acceptors and S, Se or Ti are provided as donors.

Abstract: A pin diode includes an inner zone, a cathode zone and an anode zone. A boundary surface between the inner zone and the anode zone is at least partly curved and/or at least one floating region having the same conduction type and a higher dopant concentration than in the inner zone is provided in the inner zone. The turnoff performance in such geometrically coupled power diodes, in contrast to the turnoff performance of pin power diodes (in the Read-diode version) with spaced charge coupling, is largely temperature-independent. Hybrid diodes with optimized conducting-state and turnoff performance can be made from such FCI diodes. FCI diodes are preferably used in conjunction with switching power semiconductor elements, as voltage limiters or free running diodes.

Abstract: A photodetector circuit incorporates an avalanche photodiode (APD) 300 produced by epitaxy on a CMOS substrate 302 with implanted n-well 304 and p-well 306. The n-well 304 has an implanted p+ guard ring 310 delimiting the APD 300. Within the guard ring 310 is an implanted n+ APD layer 312 upon which is deposited an epitaxial p+ APD layer 314, these layers forming the APD 300. The APD may be incorporated in an amplifier circuit 50 providing feedback to maintain constant bias voltage, and may include an SiGe absorption region to provide extended long wavelength response or lower avalanche voltage. Non-avalanche photodiodes may also be used.

Abstract: An apparatus has a head portion for holding a strip of filter, and a support portion for supporting the head portion. The head portion has a guide portion for guiding the filter, a pair of plate members disposed on both sides of the guide portion, with a space left between tips of the plate members and with vicinities of the tips provided at a given angle, and a pair of blades disposed on the tips of the plate members. When the head portion is pressed against a V-shaped channel by the support portion, elastic deformation occurs in the pair of plate members, narrowing the space between the edges of the pair of blades. When the head portion is moved off the V-shaped channel, the space between the edges of the pair of blades expands to an original width.

Abstract: A method of fabricating a self-aligned bipolar junction transistor in a semiconductor structure having a first layer of silicon carbide generally having a first conductivity type and a second layer of silicon carbide generally having a second conductivity type, opposite to the first conductivity type. The method comprises forming a pillar in the second silicon carbide layer, the pillar having a side wall and defining an adjacent horizontal surface on the second layer, forming a dielectric layer having a predetermined thickness on the second semiconductor layer, including the side wall and the horizontal surface. After formation of the dielectric layer, the dielectric layer on a portion of the horizontal surface adjacent the side wall is anisotropically etched while at least a portion of the dielectric layer remains on the side wall, thereby exposing a portion of the horizontal surface.

Abstract: A semiconductor device has: a semiconductor substrate having a surface which has a predetermined pattern, in which an insulating layer is embedded; an interlayer insulator film formed on the substrate, the interlayer insulator film having a protective coat for protecting the substrate; and an electrode formed on the interlayer insulator film. In addition, a method for manufacturing a semiconductor device comprises the steps of: forming a semiconductor substrate having a surface which has a groove in which an insulating layer is embedded; forming a protective coat for protecting the surface of the semiconductor substrate, on the upper surface of the insulating layer embedded in the groove; and forming an electrode on the protective coat.

Abstract: A method is provided for inserting dummy conductive channels along with the interconnected conductive channels. The dummy channels have an approximately even metal weight distribution to provide better plating uniformity, minimize CMP dishing, improve process heating uniformity, improve spin-on process properties, and increase etch and lithography uniformity.

Abstract: A trench DMOS device has a gate insulating layer on the bottom and sidewalls of the trench. The upper edges of the trench have an impurity injection region and are rounded. In addition, a first conductive layer is formed on the gate insulating layer, and a second conductive layer is formed on the first conductive layer and filled in the trench. The second conductive layer has different crystallization from the first conductive layer. As such the first conductive layer acts as a buffer between the gate insulating layer and the filled in second conductive layer. A method for fabricating a trench DMOS device includes the steps of forming an epitaxial layer on a semiconductor substrate. Then an impurity is injected into the epitaxial layer to form an impurity injection region. Then a trench is formed in the semiconductor substrate passing through the impurity injection region. Then a dry etching process is used to round the upper edges of the trench.

Abstract: An anode electrode metal layer composed of aluminum is formed in a region on the inner side than an anode layer formed on a main surface of a semiconductor substrate. Thus, an impurity diffusion region from the innermost circumferential surface of said surface of field limiting innermost circumferential layer to the outermost circumferential surface of the anode electrode metal layer may be used as an electrical resistance. As a result, the hole density distributed from the bottom side of the field limiting innermost circumferential layer to a cathode layer when forward bias is applied may be reduced. As a result, when a reverse bias is applied, locally great recovery current passed from a cathode layer to the bottom of field limiting innermost circumferential layer may be restrained. Therefore, a diode capable of preventing destruction of a field limiting innermost circumferential layer when a reverse bias is applied may be provided.

Abstract: A semiconductor device includes a guard ring in the termination area that is formed using the same processing steps that form the active area of the device and without requiring additional masking steps or a passivation layer. The guard ring is formed in an opening in the field oxide located in the termination area and is electrically connected to a polysilicon field plate that is located atop a portion of the field oxide region. The guard ring increases the rated voltage of the device without the introduction of a passivation layer.

Abstract: A specification is given of a bidirectionally controllable thyristor which is distinguished by improved decoupling between the two thyristor structures. In particular, the intention is that the switched-off structure cannot be triggered in an uncontrolled manner by undesirable migration of charge carriers. This is achieved by virtue of the fact that the degree of shorting of the cathode region increases toward the isolation region. In particular, this can be achieved by virtue of the fact that the density per unit area of the short-circuit regions tends to a maximum value toward the isolation region. The use of a linear, continuous short-circuit region running along the isolation region is particularly favorable. (FIG. 1).

Abstract: A semiconductor component having a high-voltage endurance edge structure in which a multiplicity of parallel-connected individual components are disposed in a multiplicity of cells of a cell array. In an edge region, the semiconductor component has cells with shaded source zone regions. During commutation of the power semiconductor component, the shaded source zone regions suppress the switching on of a parasitic bipolar transistor caused by the disproportionately large reverse flow current density. Moreover, an edge structure having shaded source zone regions can be produced very easily in technological terms, in particular in the case of self-adjusting processes, and can thus be produced cost-effectively.

Abstract: A high voltage semiconductor device such as a gate turn-off thyristor, reduces surface field concentration of a main P-N junction part and attains withstand voltage increase.

Abstract: A semiconductor device includes a first semiconductor layer of a first conductivity type having first and second main surfaces, a second semiconductor layer of a second conductivity type selectively formed on the first main surface of the first semiconductor layer, the second semiconductor layer including a first region having a relatively high injection efficiency and a second region having a relatively low injection efficiency and the first region being surrounded by the second region, a third semiconductor layer of the first conductivity type formed on the second main surface of the first semiconductor layer, a first electrode selectively formed on the second semiconductor layer of the second conductivity type and connected to at least the first region, and a second electrode formed on the third semiconductor layer of the first conductivity type.

Abstract: In a surface of a silicon substrate of one conductivity type, there are formed a plurality of depressions or recesses, gate regions of opposite conductivity type are formed at bottoms of respective recesses, gate electrodes are provided on respective gate regions, and an electrically conductive block is joined to the surface of the semiconductor substrate. Between the surface of the semiconductor substrate and the electrically conductive block a contact region having a high impurity concentration and/or an electrically conductive material layer may be provided in order to improve electrical and mechanical properties of the contact between the semiconductor substrate and the electrically conductive block. The gate region can have a high impurity concentration and a distance between a channel region and the electrically conductive block can be very small.

Abstract: A hybrid schottky injection field effect transistor is provided. A first diffusion region of a second conductivity type and a second diffusion region of a first conductivity type are separately formed at a main surface of a silicon layer. A third diffusion region of a first conductivity type is formed within the first diffusion region. An insulating layer covers part of the second diffusion region and the third diffusion region. A gate electrode is formed on the insulating layer and is situated over the first and third diffusion regions and the silicon layer. A cathode electrode is commonly connected to the third diffusion region and the first diffusion region. An anode electrode comprises a trench filled with electrode material and is formed in the silicon layer along side of the second diffusion area and a gate insulating layer.

Abstract: An optically-triggered silicon controlled rectifier (SCR) (21) having a number of semiconductor layers (23, 24, 31) diffused into an N type substrate (22). Specifically, the SCR is formed by diffusing a first P+ layer (23) into an upper surface of the substrate. Then, an N+ layer (24) is diffused into a portion of an upper surface of the first P+ layer. An oxide layer (25) which is permeable to optical radiation is formed on the first P+ layer. A conductive cathode terminal (26) is then deposited on the N+ layer. Therefore, a trench (30) is etched in the lower surface of the substrate. The trench is defined by a depth and a surface. A second P+ layer (31) is diffused into the surface of the trench. The depth of the trench substantially defines a spacing between the first and second P+ layers. The chip is soldered onto a pedestal (33) formed on a lead frame (34). The solder is deposited in the trench and contacts the second P+ layer to form an anode terminal (36).

Abstract: A semiconductor device comprises a monocrystalline silicon wafer having a major surface lying in the <100> crystal plane. Disposed on the surface is a mesa having a generally square cross-section with generally rounded corners. The mesa has four main side walls each having a slope of around 45 degrees with respect to the base plane of the mesa, and the horizontal edges of the main side walls are disposed at an angle of at least around 12 degrees to the <110> directions on the wafer surface. The corners of the mesa each comprises a number of surfaces also having slopes of around 45 degrees and one surface having a slope of around 54 degrees. A high-low (N.sup.+ N.sup.- or P.sup.+ P.sup.-) junction is disposed within the mesa and makes a continuous line intercept with the mesa side walls around the entire periphery of the mesa. Except for exceptionally small deviations of no great significance, the high low junction intercept is at a constant height location entirely around the mesa periphery.

Abstract: In a surface of a silicon substrate of one conductivity type, there are formed a plurality of depressions or recesses, gate regions of opposite conductivity type are formed at bottoms of respective recesses, gate electrodes are provided on respective gate regions, and an electrically conductive block is joined to the surface of the semiconductor substrate. Between the surface of the semiconductor substrate and the electrically conductive block a contact region having a high impurity concentration and/or an electrically conductive material layer may be provided in order to improve electrical and mechanical properties of the contact between the semiconductor substrate and the electrically conductive block. The gate region can have a high impurity concentration and a distance between a channel region and the electrically conductive block can be very small.

Abstract: An insulated gate field effect transistor comprising a semiconductor substrate having one side on which a cell area is composed of a plurality of first wells of a first conductivity type, each of the first wells containing a source region of a second conductivity type. A channel region is defined in the surface portion of the semiconductor substrate adjoining to the source region, and a gate electrode is formed, via a gate insulating film, at least over the channel region. A source electrode is in common contact with the respective source regions of the plurality of first wells. The semiconductor substrate has a drain electrode provided on another side. A current flows between the source electrode and the drain electrode through the channel being controlled by a voltage applied to the gate electrode. A guard ring area is disposed on the one side of the semiconductor substrate so as to surround the cell area.

Abstract: The present invention provides a reverse conducting (RC) thyristor of a planar-gate structure for low-and-medium power use which is relatively simple in construction because of employing a planar structure for each of thyristor and diode regions, permits simultaneous formation of the both region and have high-speed performance and a RC thyristor of a buried-gate or recessed-gate structure which has a high breakdown voltage by the use of a buried-gate or recessed-gate structure, permits simultaneous formation of thyristor and diode regions and high-speed, high current switching performance, and the RC thyristor of the planar-gate structure has a construction which comprises an SI thyristor or miniaturized GTO of a planar-gate structure in the thyristor region and an SI diode of a planar structure in the diode region, the diode region having at its cathode side a Schottky contact between n emitters or diode cathode shorted region and the thyristor region having at its anode side an SI anode shorted structure fo

Abstract: A P.sup.+ layer is formed on the lower surface of an N.sup.- substrate, and recesses are defined in the upper surface of the N.sup.- substrate. Then, P.sup.+ gate regions and bottom gate regions are formed in side walls and bottoms of the recesses. The N.sup.- substrate and an N.sup.- substrate are ultrasonically cleaned to remove impurities therefrom, then cleaned by pure water, and dried by a spinner. Then, while lands on the upper surface of the N.sup.- substrate are being held against the surface of the N.sup.- substrate, the N.sup.- substrate are joined to each other by heating then at 800.degree. C. in a hydrogen atmosphere.

Abstract: An insulated gate type semiconductor device has a gate electrode which controls current flow between two regions of the same conductivity type in a semiconductor substrate. A main electrode has a first portion contacting a first one of the two regions, a second portion extending above the gate electrode and a third portion providing a raised external contact surface to contact an external electrode. The gate electrode is insulated above and below by insulating films. To prevent damage to the gate electrode and the lower insulating films due to the pressure of the external electrode, there is a supporting insulating layer on the surface of the substrate underlying the contact portion of the main electrode and having a thickness substantially greater than the thickness of the insulating film below the gate electrode and the contact surface is more remote from the substrate than the second portion of said main electrode.

Abstract: A novel semiconductor device with a pair of main surfaces is disclosed, in which at least three semiconductor layers are formed adjacently to each other. The device comprises a main thyristor portion for supplying a main current, an auxiliary thyristor portion, a pilot thyristor portion and a breakover portion. The breakover portion, in turn, includes a semiconductor layer having a high impurities concentration formed on one of the main surfaces, and a plurality of semiconductor layers having a high impurities concentration of opposite conduction type formed adjacently to the semiconductor layer and in spaced relationship from each other.

Abstract: A lateral transistor (14) is configured as a reverse protection diode that allows low and high current modes of operation while maintaining low forward voltage drop. The base region (38) of the lateral transistor is formed inside a collector ring (34) and adjacent to the emitter region (36). In low current mode, the transistor operates as a conventional diode. In high current mode, the excessive number of minority carriers injected into the base region causes the device to enter conductivity modulation that effectively increases the doping concentration and lowers the bulk resistance. The lower bulk resistance keeps the forward voltage drop low. By having the base region inside the collector ring, the bulk resistance is kept low to aid in the onset of conductivity modulation. Thus, the transition between low current mode and high current mode is minimized.

Abstract: A semiconductor integrated circuit device has a test component associated with a dummy test pattern for evaluating corresponding circuit components of the integrated circuit, and the test component and the dummy test pattern is surrounded by a peripheral dummy pattern so that a micro loading effect on the test component is equivalent to the corresponding circuit components.

Abstract: To provide thermal relief, particularly of the edge of disk-shaped gate-turn-off GTO thyristors (GTO) as are used in converters in power electronics, at least one cooling segment which is isolated from a GTO cathode metallization of the GTO thyristor segment (GTO) by a gate electrode metallization of a gate electrode is arranged on the edge and laterally adjacent to the GTO thyristor segment (GTO). An insulation layer is provided between a cooling segment metallization and the gate electrode metallization. Cooling segments in an lo outer annular zone can be alternately arranged with GTO thyristor segments (GTO) or offset towards the outside in the radial direction or perpendicular direction thereto. Instead of cooling segments, a p.sup.+ -type GTO emitter layer of the GTO thyristor segments (GTO) can be shortened at the edge in the outer annular zone.

Abstract: A process for manufacturing a semiconductor switching device (such as a thyristor device) comprises: etching a face of a semiconductor body to provide islands and channels which define a mesa-contoured surface; diffusing dopant of a first conductivity type through said surface so that the lines of equal concentration of the dopant in said body follow substantially the mesa-contoured surface; and diffusing dopant of a second conductivity type into said islands to form p-n junctions with said dopant of a first conductivity type. The diffusion of said dopant of a first conductivity type is followed by an out-diffusion step so that the dopant concentration of said dopant of a first conductivity type is at a maximum at a depth below said surface.

Abstract: A semiconductor device is provided with a surface-inactivated semiconductor layer provided on the surface of a compound semiconductor on which surface a semiconductor layer forming the depletion layer is provided, the semiconductor layer forming the depletion layer being of a conduction type opposite that of the compound semiconductor, and having a carrier density and thickness being capable of forming a depletion layer on the compound semiconductor. When a depletion layer is formed on the surface of the compound semiconductor by the semiconductor layer forming the depletion layer, the depletion layer has no charge so that the concentration of electrical fields is relaxed, the surface of the semiconductor is stabilized, and excellent dielectric breakdown performance is obtained.

Abstract: For IGBT, MCT or like devices, the substrate is formed with P+, N+ and N- layers and PN diffusions to define body and source regions in the N-layer and a MOS-gated channel at the upper surface. The N-layer is sized and doped (.about.10.sup.14 /cm.sup.3) to block reverse bias voltage. The N+ layer is >20 .mu.m thick and doped below .about.10.sup.17 /cm.sup.3 but above the N- doping to enhance output impedance and reduce gain at high V.sub.ce conditions. Or the N+ layer is formed with a thin (.about.5 .mu.m) highly doped (>10.sup.17 /cm.sup.3) layer and a thick (>20 .mu.m) layer of .about.10.sup.16 /cm.sup.3 doping. A platinum dose of 10.sup.13 to 10.sup.16 /cm.sup.3 is ion implanted and diffused into the silicon to effect lifetime control. Gate and source contacts and body and source diffusions have an inter-digitated finger pattern with complementary tapers to minimize current crowding and wide gate buses to minimize signal delay.

Abstract: Disclosed herein is an insulated-gate thyristor comprising a base layer of a first conductivity type, having first and second major surfaces, a first main-electrode region of the first conductivity type, formed in the first major surface of the base layer, a second main-electrode region of a second conductivity type, formed in the second major surface of the base layer, at least a pair of grooves extending from the first main-electrode region into the base layer, and opposing each other and spaced apart by a predetermined distance, insulated gate electrodes formed within the grooves, and a turn-off insulated-gate transistor structure for releasing carriers of the second conductivity type from the base layer.

Abstract: A MOS semiconductor device which exhibits high switching operations including high turn-on and an excellent self-cooling capability. The device prevents damage to insulation films and electrodes thereof. An IGT includes a multi-layer structure having a p type emitter layer, an n type base layer, a p type base layer and an n type emitter layer superimposed therein. A gate electrode and an overlying gate oxide film are disposed on a recessed surface of the multi-layer structure. A cathode electrode is located only in and around a cathode surface so that most of the top surface of the gate electrode is uncovered. Via an intervening cathode distortion snubbering plate, the cathode electrode is in pressure contact with a cathode electrode body. The gate and the cathode electrodes have a reduced capacitance therebetween. The cathode electrode body serves to cool the cathode electrode. The gate electrode and the gate oxide film are protected from stress, and hence, will not be damaged by stress.

Abstract: There is disclosed a semiconductor device wherein a p layer (7) is formed in an isolating portion (Z) and portions (1a, 1b) of an n-type base layer (1) lie on opposite sides of the p layer (7), the upper surfaces of the p layer (7) and the portions (1a, 1b) lying in the same plane as the upper surface of a p layer (3). The presence of the p layer (7) provides for high resistance to breakdown and high formation accuracy of the p layers (2, 3, 7) as compared with a structure in which the isolating portion (Z) lies in the bottom of a the recess, whereby the semiconductor device is less susceptible to short-circuit between the p-type base layer (2) and the p layer (3).

Abstract: A readily manufacturable field controlled thyristor with a first semiconductor region of n-type conductivity, a second semiconductor region of p-type in contact with said first region, a void penetrating through said first and second semiconductor regions, a fourth semiconductor region of n-type forming a channel adjacent to said void, a fifth semiconductor region, of p-type, in contact with said third region. The device has a large tolerance for deviations in process parameter precision and accuracy, which enables the device to be produced at a low cost.

Abstract: Semicondctor devices having a curved P-N junction in an active area of the device and an edge passivation region extending from the active area to an edge region of the device include an electrically resistive ribbon that spirals outwardly from the active area to the edge of the device so that a voltage difference between the active area and the edge region is spread along the length of the ribbon. The ribbon may take the form of a linear resistor or may include plural diodes. The distance between radially overlapping portions of the spiralling ribbon and the cross-sectional area of the ribbon may be varied to spread the equipotential lines in the device so as to reduce the effect of the curved P-N junctions on the breakdown voltage of the device.

Abstract: A semiconductor device including a normally-on SI thyristor, and a MOSFET connected in cascade with the SI thyristor. The gate of the SI thyristor is connected to the source of the MOSFET. This arrangement makes it possible to turn the device on and off by controlling only the voltage gate of the MOSFET, obviating a current to maintain the on state of the device. The device needs little driving energy and has a low on state voltage and a high switching speed. It can readily be integrated into one chip.

Abstract: A channel in which electron current is supplied from n.sup.+ type source layer to an n.sup.- type base layer is formed in a thyristor portion by using a first gate electrode to have an electrical connection in a thyristor state. Injection of hole current to a p type base layer, which is necessary to maintain the thyristor state is extracted to a source terminal by a control MOSFET portion including a second gate electrode a turn-off time and the state of this device is changed to the transistor state similar to that in the IGBT so that a short switching time turn-off is realized.

Abstract: A semiconductor body (2) has adjacent a first major surface (3) a first region (5) of one conductivity type part of which defines an active device area (6) of a power semiconductor device (7) having at least two electrodes (8 and 9 or 8 and 10) and active device regions (11) each forming with the first region (5) a pn junction (11a) extending to the first major surface (3). A protection device (12) formed by a series-connected array of semiconductor rectifying elements (13) is provided on an insulating layer (14) on the first major surface (3). The protection device (12) is connected between at least two electrodes (8 and 9 or 10) of the power semiconductor device (7) so as to break down to cause conduction between the two electrodes when the voltage across the protection device (12) exceeds a predetermined limit.

Abstract: A switching device includes a thyristor and a MOSFET, and a voltage clamp circuit. The voltage clamp circuit includes an N.sup.+ type contact region formed in a surface layer of a N type substrate and electrically connected to a gate electrode of a MOSFET, and a P type guard ring surrounding the contact region.

Abstract: A FET structure for use in narrow bandgap semiconductors comprising a narrow bandgap semiconductor substrate 24, an implanted source region 12 of a conductivity type opposite that of the substrate 24, an implanted drain region 12 of the same conductivity type as source region 12 and spaced from source region 12, a first diode guard ring 14 insulatively disposed on the substrate 24 and surrounding source region 12, a second diode guard ring 14 insulatively disposed on the substrate 24 and surrounding drain region 12, a gate region 16 insulatively disposed on the substrate 24 and surrounding the source and drain regions 12 and the first and second diode guard rings 14, an outer periphery transistor guard ring 18 insulatively disposed on the substrate 24 and surrounding the gate region 16 and a field plate region 20 insulatively disposed on the substrate 24 and surrounding the outer periphery transistor guard ring 18.Other devices, systems and methods are also disclosed.

Abstract: The semiconductor protection device has a p.sup.+ -n.sup.- -p-n.sup.+ layer construction, and an n type impurity diffusion region is selectively formed in a surface portion of the pn junction. This n type impurity diffusion region is formed in a linear planar portion where substantially no electric field concentration is generated when a reverse voltage is applied to the pn junction formed between the n.sup.- type semiconductor region and the p type semiconductor region. Further, an electrode is provided in ohmic contact with both of the p type semiconductor region and the n.sup.+ type semiconductor region. This electrode is selectively made in contact with the p type semiconductor region at a position remote from the n type impurity diffusion region and adjacent to a curved planar portion of the pn junction where the electric field concentration tends to occur when a reverse voltage is applied to the pn junction formed between the n.sup.- type semiconductor region and the p type semiconductor region.