Short-circuit semiconductor device and method of operating the same

Abstract

The short-circuit semiconductor device has a semiconductor body, in which a back base region and a front base region of a first conductivity type and an internal region of a second conductivity type complementary to the first conductivity type are arranged. The back and front base regions are conductively connected to the back and front electrodes. At least one front and / or back conductive structure is embedded in the front and / or base region and at least partially covered by the front and / or back electrodes. The conductive structure is an emitter structure of the second conductivity type, which is conductively contacted with the corresponding electrodes respectively and can be turned on by means of at least one trigger structure, which is activated by means of an electrical conduction signal. In the activated state, the trigger structure injects a surge current into the semiconductor body, which irreversibly destroys the first semiconductor junction formed between the conductive structure and the base region in which it is embedded and / or the second semiconductor junction formed between the base region and the internal region. It also relates to a method for operating such a short-circuit semiconductor device.

Description

Technical Field

[0001] The invention relates to a short-circuit semiconductor component according to the preamble of claim 1 and to a method for operating such a short-circuit semiconductor component. Background Art

[0002] After a certain fault or disturbance, for example in a converter system, all components and in particular power semiconductors may be destroyed due to excessive short-circuit currents or overload currents, which may be as high as 100 kA. In order to reliably avoid the destruction of components in the system, in particular semiconductor components, various protective devices have been used.

[0003] When a fault occurs, a limited, permanent low-resistance connection (short circuit) can be generated, for example, by a safety switch in the form of a short-circuiting device. Here, the protection device can be triggered once and remain permanently short-circuited after being triggered by an electrical or optical trigger signal.

[0004] In modern applications for energy transmission and in high-power electric drives, there is an additional requirement to short-circuit individual submodules in series in the event of a fault in order to continue to ensure operation of the system. The short-circuiting of these submodules also serves to protect the application from damage and is therefore of great importance for protecting the application.

[0005] An alternative application is, for example, short-circuit breakers and fast earthing devices in high-voltage and medium-voltage equipment. Currently, such devices are implemented as mechanical switches, driven by spring force. They are used, for example, to prevent equipment overload / destruction due to arc faults, by causing a fast direct short circuit or a fast earthing. In addition, high-voltage high-power fuses (HH fuses) may also melt in the event of an equipment overload that the fuse has not yet safely switched (e.g. due to a more distant arc fault) and short-circuit safely.

[0006] Typically, mechanical switches are used for short-circuiting, which have a switching time (from opening to closing) of approximately 1 ms to 10 ms. Since a short-circuit of a submodule occurs only in the event of a fault, it is more likely that the switch will be used only rarely (e.g. once every 10 years). Mechanical switches therefore have problems with safe switching operation after long periods of non-use. Due to the movable parts in the mechanical switch and, if necessary, the use of grease and oil, safe switching operation can no longer be guaranteed after correspondingly long (standstill) times.

[0007] For example, DE10323220B4 discloses a short-circuit circuit for a faulty partial converter, in which an electronic semiconductor device absorbs a short-circuit current in the event of a fault or is activated according to this short-circuit current and then permanently melts. As an electronic semiconductor device, an active semiconductor switch is proposed, which can be constructed in particular as a conventional thyristor. In the event of a fault, the thyristor is triggered, and the short-circuit current flowing through the thyristor thereafter causes local overheating of the semiconductor material, which can destroy the short-circuit thyristor. This will then form a permanent short circuit. The contact of the short-circuit thyristor is designed to be short-circuit-proof so that no arc is formed in the short-circuit thyristor due to the high short-circuit current. For this reason, pressure contacts are particularly proposed.

[0008] DE10333798A1 also describes a method for short-circuiting a faulty partial converter, wherein in the event of a fault, a conductive IGBT power semiconductor switch (insulated gate bipolar transistor) is turned on for a short time, so that the IGBT power semiconductor switch is permanently short-circuited. In order to ensure the safe short-circuiting of the semiconductor switch, an increase in the gate-emitter voltage of the IGBT power semiconductor switch is also proposed. The IGBT usually has a MOS control region, in which a conductive channel is generated between the IGBT source and the IGBT drain or the IGBT emitter and the IGBT collector by a small voltage applied to the IGBT gate, which causes the IGBT to turn on. However, the IGBT-connections "gate" and "emitter" are always electrically insulated from each other by an insulator (oxide layer) that only enables an "extreme" increase in the gate-emitter voltage. When a higher voltage is applied to the gate electrode that far exceeds the usual operating rating, a breakdown will occur, causing the IGBT to turn on.

[0009] In addition, EP3001525A1 discloses a short-circuit switch having a controllable semiconductor switch for short-circuiting a voltage between two conductors. The semiconductor switch is a conventional press-pack thyristor having planar contact electrodes on opposite contact sides. When the semiconductor component is installed, its contact electrodes are contacted by planar connecting electrodes, which act on each other with high rigidity elasticity. In this way, planar contact of the semiconductor component is also ensured by high-flowing power current under high thermal stress. The thyristor is arranged in a mechanical press-pack structure, which constructs a protective cover surrounding the thyristor. The thyristor is designed to be irreversibly destroyed as a controllable semiconductor switch by means of a short-circuit current flowing through the thyristor during a short circuit, wherein conductivity is maintained between its contact electrodes. The short-circuit current should be at least 50 times the rated current of the thyristor for its switch design.

[0010] The disadvantage of using conventional thyristors as short-circuit switches is that it is not possible to ensure at what current the thyristor will be destroyed and form a defined short circuit, because the surge current value of conventional thyristors deviates greatly upward.

[0011] The "short circuit fault" state achieved by the targeted short circuit as described above can be ensured to be safe by a suitable module structure, such as a pressure contact. Here, the semiconductor device is located in a suitable housing, such as a ceramic unit or a ceramic capsule, and its external current contacts, usually the cathode and anode terminals, are made of thick copper dies (or other conductive metals). Through suitable pressure bonding, it is ensured that the two contact dies are in contact with the corresponding metal contact electrodes of the semiconductor device with a sufficiently high pressure so as to establish a particularly low electrical and thermal transition between the semiconductor device and the copper die of the housing. When the melting position is located in the pressurized contact area, the damage area of ​​the semiconductor device remains unchanged on the pressurized contact surface not only during the fault period in which melting or short circuit occurs, but also during the continuous current flow after the fault. The housing will not or will hardly be damaged from the outside, and the energy conversion is carried out inside the semiconductor component during and after the fault situation, and there it remains limited to the area of ​​the pressure contact. It is decisive that the electrical connection in the component will not be separated due to the fault situation, such as occurs in the case of welded components, and the current flows through the bonding wire. In this case, in the case of high currents, the terminals in the component are electrically separated due to the melting of the bonding wires or the soldered connections. Due to the very high energy in the event of a fault, complete destruction of the entire semiconductor component is inevitable, which also poses a high risk to the equipment in the surroundings of the component. The bonding wires are relatively open in the housing surrounding the semiconductor component, usually also embedded in a gel, so that explosive destruction can occur due to arcs between the interrupted bonding wires. In contrast, in components with pressure contacts, the energy conversion or melting range or arc in the silicon is limited to the area contacted by the contact plates on the anode side and the cathode side, so that even at the highest current densities, explosions or destruction of the immediate surroundings of the component are avoided as much as possible.

[0012] If conventional thyristors are loaded with their surge current limit value, they will temporarily, i.e. reversibly, lose their blocking and isolating capabilities partially or completely due to large-area heating in their cathode surface until the blocking layer temperature drops back to a range permissible for continuous operation. The permissible surge current density at the current maximum of a single 10 ms pulse is usually 1.0 kA / cm 2 and 1.5kA / cm 2The thyristor is designed to be able to withstand a current density of 0.1 to 0.1 V at which no irreversible damage occurs. If this limit value is exceeded significantly, heating results in melting (melting) in the cathode surface, the thyristor is destroyed and loses its blocking capability or control function irreversibly. As long as the melting is located in the area contacted and pressed by the copper contact pads or copper stamping of the capsule and at the same time sufficiently far away from the edge area of ​​the thyristor, the capsule and the immediate surroundings remain as intact as possible.

[0013] However, the main disadvantage of conventional thyristors as short-circuit protection devices described in DE102008049678A1 or DE102009045216A1, for example, is their large cathode surface, which is designed to minimize conduction losses or penetration losses or maximum surge current capabilities. Its diameter is roughly equivalent to the diameter of the contact disk of the copper die or is usually slightly larger (up to about 0.5 to 2 mm). If the desired melting is caused on the cathode surface due to high short-circuit current, it cannot be ensured that the melting occurs far enough from the edge area of ​​the thyristor. The edge area can be understood as a radially outer area of ​​the device, which is not covered or pressed by a contact disk such as a copper die that is usually in pressure contact with the contact electrodes (especially cathode and anode terminals) of the semiconductor device. The molten area is located anywhere within the cathode surface and can also extend into the edge area. In this case, the destruction of the ceramic housing and the escape of plasma may result. Plasma escape may also occur in the center of a semiconductor device when, for example, a gate electrode or an auxiliary thyristor structure (amplifier gate structure) is arranged in the center of the semiconductor device, and the gate or auxiliary thyristor structure is also not covered or pressed by the contact pad of the copper stamper that is in pressure contact with the semiconductor device.

[0014] Moreover, the edge area of ​​the thyristor is more susceptible to the overvoltage that occurs than the remaining area that is directly contacted or covered by the copper contact plate. This failure, together with the short-circuit current that flows subsequently, can lead to the destruction of the capsule due to the very high energy density, especially in the case of high blocking voltages, as described above. Although the strength against such overvoltages can be significantly improved by various measures, this has serious disadvantages. For example, a much thicker silicon starting wafer will significantly improve the blocking capacity, and thus the possibility of edge failures caused by overvoltages can be avoided to a large extent. However, due to the excessively high dynamic losses and penetration losses in the conventional thyristor structure, this cannot be meaningful, because the thyristor can no longer meet the requirements thereof, in particular the lowest penetration losses and the maximum surge current characteristics and the lowest dynamic losses.

[0015] If, for example, a thyristor is subjected to a blocking voltage U RIf the thyristor fails, for example due to overvoltage or blocking instability caused by the passivation of its edge profile, the fault location is mostly located in the edge area of ​​the thyristor component and therefore outside the area directly contacted by the copper contact disk. A very high short-circuit current flows in the blocking direction, wherein the undisturbed area does not participate in the current flow, and melting occurs at the fault location. An arc occurs inside the shell of the capsule or at the fault location, the molten material evaporates and generates hot plasma, which, depending on the intensity, can lead to the destruction of the capsule. Due to this explosive discharge, the ceramic capsule may be torn in the vicinity of the melting, and then the hot plasma escapes from the ceramic capsule. Such edge faults may also occur when the blocking voltage U is applied. D occurs. Summary of the invention

[0016] Against this background, the object of the present invention is to provide a short-circuit semiconductor component with which a voltage applied to the semiconductor component can be short-circuited reliably, safely and permanently (long-term stable) in response to a trigger signal or a conduction signal, without plasma escaping explosively from the semiconductor component. Furthermore, the blocking losses in the blocking operating mode of the semiconductor component should be so low that the semiconductor component can be operated without active cooling. Furthermore, a method for operating a short-circuit semiconductor component is to be provided.

[0017] This object is achieved by a short-circuit semiconductor component having the features of claim 1 and by a method for operating a short-circuit semiconductor component having the features of claim 24. Further particularly advantageous embodiments of the invention are disclosed in the dependent claims.

[0018] It should be pointed out that the features listed individually in the following description can be combined with one another in any desired technically meaningful manner and disclose further embodiments of the invention. The description, in particular in conjunction with the drawings, additionally characterizes and explains the invention in detail.

[0019] According to the invention, a short-circuit semiconductor device has a semiconductor body in which, starting from a back side with a back side center determined by its geometric center of gravity, to a front side opposite to the back side with a front side center determined by its geometric center of gravity, a back side base region of a first conductivity type (e.g., p-type), an inner region of a second conductivity type (e.g., n-type) complementary to the first conductivity type, and a front side base region of the first conductivity type are successively arranged in a vertical direction. In addition, the back side base region is electrically conductively connected to a back side electrode (metallization, e.g., aluminum) applied on the back side, the back side electrode having a lateral front side electrode width provided by its lateral extension. The front side base region is electrically conductively connected to a front side electrode (metallization, e.g., aluminum) applied on the front side, the front side electrode having a lateral back side electrode width provided by its lateral extension. In addition, according to the invention, at least one front side conductive structure is embedded in the front side base region and is at least partially covered by the front side electrode, and / or at least one back side conductive structure is embedded in the back side base region and is at least partially covered by the back side electrode. According to the invention, the respective conduction structure is an emitter structure of the second conductivity type (e.g. n-type), which is in conductive contact with an electrode applied to the base region in which it is embedded, and which can be turned on by means of at least one trigger structure electrically acting on it through the semiconductor body, which in turn can be activated by means of an electrical turn-on signal supplied to it. Furthermore, the trigger structure is constructed and arranged such that in its activated state it injects a surge current with a predetermined energy input, e.g. approximately 2 Ws, into the semiconductor body between the trigger structure and the front and / or back conduction structure, which in its activated state irreversibly destroys a first semiconductor junction, e.g. an np junction, formed between the respective conduction structure and the base region in which it is embedded, and / or a second semiconductor junction, e.g. a pn junction, formed between the base region and the inner region.

[0020] It should be noted that the conjunction “and / or” between two features and connecting them to one another is always to be interpreted in such a way that in a first embodiment of the subject matter according to the invention only the first feature may be present, in a second embodiment only the second feature may be present and in a third embodiment both the first feature and also the second feature may be present.

[0021] Here, a direction substantially perpendicular to the vertical extension direction of the semiconductor body is understood to be a lateral or radial extension direction of the semiconductor body. Thus, in the case of a semiconductor body which is designed, for example, as a cylinder (with a small height), the vertical extension direction corresponds to the direction of the cylinder height and the lateral direction corresponds to the radial direction of the cylinder. The terms "lateral" and "radial" are used synonymously in this document.

[0022] Furthermore, it should be noted that the front center and the back center of the short-circuited semiconductor component can be different, but this is not necessary and is not usually the case. The two centers usually determine the symmetry axis of the semiconductor body, from which the radius of the semiconductor body is determined. When the center is mentioned below with respect to the front part of the short-circuited semiconductor component, this center is always understood to be the front center, unless otherwise explicitly specified here. Similarly, unless otherwise explicitly stated in this text, in the relationship between the center of the semiconductor body and the back part, the back center should be understood accordingly.

[0023] It is understood that an emitter structure in conductive contact with a front electrode is a front-side conductive structure, wherein an emitter structure of a second conductivity type is embedded in a front base region of a first conductivity type adjacent to the front side of the semiconductor body in such a way that the emitter structure is at least partially conductively contacted by applying the front electrode to the front side of the semiconductor body. Similarly, this relates to an emitter structure in conductive contact with a back electrode, which is correspondingly a back-side conductive structure, wherein an emitter structure of a second conductivity type is embedded in a back base region of a first conductivity type adjacent to the back side of the semiconductor body in such a way that the emitter structure is at least partially conductively contacted by applying the back electrode to the back side of the semiconductor body. Contact of the emitter structure with the respective front or back electrode is preferably carried out here by direct contact between the components.

[0024] The lateral width of an overall structure, such as a conductive structure or an electrode, is derived from the distance between the inner side which delimits the overall structure radially furthest inwards and the outer side which delimits the overall structure radially furthest outwards. This is obvious for structures of a semiconductor body, which are arranged at a distance from its radial center (r=0), i.e. the geometrical centroid or center of the front and / or back side of the semiconductor body. In the special case where the overall structure has no distance from the radial center, i.e. the radial inner side of the overall structure reaches the radial center, the lateral width of the overall structure corresponds to the distance between the outer side which delimits the entire structure radially outwards and the radial center. In other words, in this case, the lateral width corresponds to the radius of the overall structure (measured from the radial center) or half its diameter.

[0025] Therefore, the semiconductor body of the short-circuit semiconductor device according to the present invention (without considering the conduction structure) has a pnp or npn structure between its front electrode or front metallization and its back electrode or back metallization, which reliably and stably blocks the device for applying a voltage between the front electrode and the back electrode in both directions or polarities for a long time. This situation is referred to herein as the normal non-short-circuit operating state of the short-circuit semiconductor device according to the present invention.

[0026] In contrast, if an overload due to an excessively high current of up to several hundred kA occurs in the electrical device to be protected by the short-circuit semiconductor device according to the present invention, at least the first semiconductor junction (e.g., np junction) formed between the conduction structure and the base region embedded in the conduction structure and / or the second semiconductor junction (e.g., pn junction) formed between the base region and the inner region is irreversibly destroyed (also referred to herein as pre-destruction) by the prior activation of the trigger structure and the subsequent surge current generated by the trigger structure. The blocking effect of the two semiconductor junctions between the conduction structure, the base region, and the inner region is thereby eliminated.

[0027] If now after this pre-destruction a sufficiently high main current or load current flows directly between the external metal terminals of the semiconductor component, that is to say between the front and back electrodes, this will further increase the degree of this destruction and, due to the high current density, lead to permanent, irreversible, low-ohmic melting or melting between the two electrodes contacting the semiconductor body (permanent short circuit). The degree or magnitude of this melting ultimately depends on the current density or main current between the front electrode and the back electrode. The short-circuit semiconductor device according to the invention is therefore no longer able to block. This state is referred to herein as the short-circuit operating state or short-circuit state of the short-circuit semiconductor device according to the invention.

[0028] Activation of the trigger structure by supplying a predeterminable switch-on signal to the semiconductor body or to the trigger structure results in a targeted, one-time creation of a permanent, irreversible, electrically conductive, low-ohmic connection between the front electrode and the back electrode, ie, a permanent short circuit.

[0029] In this case, a permanent short circuit is produced after the preliminary destruction of at least one of the two semiconductor junctions by the targeted conduction of the short-circuit current on a locally limited surface in the component, so that there subsequently a melting occurs in the conductive area due to the high current density. In the melting, which is also called a short circuit, the material of the external metal terminal (for example aluminum, molybdenum and / or copper) and the silicon of the semiconductor body melt. After the melting process, a highly conductive metal area is formed, which permanently short-circuits the external metal terminal and can, in principle, also maintain this short-circuit effect for a longer time under current load, for example until the next repair.

[0030] Compared to the initial non-destructive triggering or switching of the conduction structure, the particular advantage of the pre-destruction by the trigger structure according to the invention is that the pre-destruction is introduced at a point in time which must only be in the vicinity of the actual fault event. Subsequent large fault load currents can then also occur at a later point in time which can be, for example, a few milliseconds after the activation of the trigger structure.

[0031] In contrast, in the case of a "non-destructive" (conventional) switching of the conduction structure by means of a trigger structure, a conventional trigger pulse must be generated directly or immediately before the fault load current, in particular within a time range of only very few microseconds. Since as long as a conventional trigger pulse of several tens of microseconds to 100 μs (depending on, for example, the principle of the generator generating the conduction signal) is still present or has its effect, a fault load current of several hundred kiloamperes can cause the desired destruction in the conduction structure and produce a permanent short circuit in the semiconductor structure.

[0032] The short-circuit semiconductor component according to the invention can therefore allow a greater compensation in time between the generation of the switch-on signal (predestruction) and the subsequent complete melting of the semiconductor body caused by the fault load current.

[0033] Furthermore, by using the semiconductor-based short-circuit switch according to the invention, the switching time is reduced from the conventional approximately 1 ms to 10 ms to less than 100 μs. Thus, according to the invention, there are no movable mechanical parts in the short-circuit switch, and furthermore the switching time is so short that the safety or reliability of the short-circuit semiconductor device is significantly improved.

[0034] It is emphasized again that two circuits are always provided in the short-circuit semiconductor device according to the invention:

[0035] 1) Trigger circuit causing pre-destruction and

[0036] 2) The load current loop between the external metal terminals of the semiconductor device, that is, between the front electrode and the back electrode.

[0037] According to another aspect of the present invention, a method for operating a short-circuit semiconductor device having the features of the present invention described herein is also disclosed. In particular, with respect to the definitions of terms related to the method and the effects and advantages of the features related to the method, the explanations of the reasonable definitions, effects and advantages of the device of the present invention set forth herein should be fully applicable. That is, as long as it is not explicitly excluded here, the disclosure of the device according to the present invention herein should also be used in a similar manner to define the method according to the present invention. Similarly, the disclosure of the method of the present invention herein should also be used in a similar manner to define the device of the present invention, as long as this is also not explicitly excluded here. In this regard, for a more compact description, the repetition of the explanation of the meaningful and identical features, their effects and advantages of the device according to the present invention disclosed herein and the method according to the present invention disclosed herein is largely abandoned.

[0038] In a method for operating a short-circuit semiconductor device according to the present invention, the short-circuit semiconductor device has a semiconductor body in which, starting from a back side having a back side center determined by its geometric center of gravity to a front side opposite to the back side having a front side center determined by its geometric center of gravity, a back side base region of a first conductivity type, an inner region of a second conductivity type complementary to the first conductivity type, and a front side base region of the first conductivity type are successively arranged in a vertical direction, wherein the back side base region is electrically conductively connected to a back side electrode applied on the back side with a lateral back side electrode width given by its lateral extension, and the front side base region is electrically conductively connected to a back side electrode applied on the front side. The front electrode is conductively connected with a lateral front electrode width given by its lateral extension, wherein at least one front conduction structure is embedded in the front base region and is at least partially covered by the front electrode, and / or at least one back conduction structure is embedded in the back base region and is at least partially covered by the back electrode, wherein the corresponding conduction structure is an emitter structure of the second conductivity type, which is respectively conductively contacted with the electrode applied to the base region in which it is embedded, and the emitter structure can be turned on by means of at least one trigger structure electrically acted on by the semiconductor body, wherein the trigger structure can be activated by means of an electrical conduction signal supplied to it.

[0039] In the method according to the present invention, a trigger structure is activated by means of a conduction signal, and a surge current with a predefined energy input is injected into the semiconductor body between the trigger structure and the front and / or back conduction structure, thereby irreversibly destroying the first semiconductor junction formed between the corresponding conduction structure and the base region embedded in it, and / or the second semiconductor junction formed between the base region and the internal region.

[0040] According to an advantageous embodiment of the invention, the energy input is predetermined in the range between approximately 1 Ws and approximately 3 Ws, for example 2 Ws. Thus, a pre-destruction of at least one of the two semiconductor junctions can be reliably ensured.

[0041] The term "approximately" used herein is related to the disclosed invention and represents the tolerance range generally understood by those skilled in the art. The term "approximately" should be understood as the tolerance range of the relevant parameter, with a maximum of ±20%, preferably a maximum of ±10%.

[0042] According to a further advantageous embodiment of the invention, the semiconductor body has a resistance of at most about 10Ω, preferably at most 5Ω or even less (greater than 0Ω), in the region of the surge current injection between the trigger structure and the front and / or back conductive structure in the radial direction. This means that there is a relatively good conductive connection between the trigger structure and the conductive structure, so that correspondingly no high voltages can be applied between the trigger structure and the conductive structure. Energy is thus input into the semiconductor body, and depending on the charge quantity provided by the trigger structure and the associated height of the surge current, a surge current flows between the activated trigger structure on one side and the corresponding conductive structure on the other side. However, as described here, the charge quantity or the intensity of the surge current can be determined and set very well.

[0043] Another advantageous embodiment of the present invention provides that the front conductive structure has a lateral front conductive structure width given by its lateral extension, and the back conductive structure has a lateral back conductive structure width given by its lateral extension, wherein the ratio of the lateral conductive structure width to the corresponding lateral electrode width of the corresponding electrode contacting the corresponding conductive structure is less than 1, preferably less than 4 / 5, preferably less than 3 / 4, further preferably less than 2 / 3 and further more preferably less than 1 / 2.

[0044] A conductive structure configured as an emitter structure with a certain lateral conductive structure width offers the advantage that the current plasma can expand on the surface of the emitter structure before melting begins, as a result of which the melt has a larger lateral width and thus a larger cross section overall. The larger conductive cross section of the melt region in the semiconductor body achieved in this way advantageously leads to a reduction in the electrical resistance of the short-circuit region formed in the semiconductor body by melting. As a result, the short-circuit region can permanently carry higher short-circuit currents with low conduction losses. Advantageously, active cooling of the semiconductor device according to the invention can be dispensed with.

[0045] Taking into account the relationship between the width of the lateral conductive structure and the corresponding lateral electrode width, according to the invention, the lateral conductive structure width of the front conductive structure (if it is present in the semiconductor body) is always compared with the lateral electrode width of the front electrode, and the lateral conductive structure width of the back conductive structure (if it is present in the semiconductor body) is always compared with the lateral electrode width of the back electrode. Accordingly, this also refers to the allocation of the front conductive structure to the front electrode and to the allocation of the back conductive structure to the back electrode.

[0046] As already explained above, a permanent short circuit is formed on a locally limited surface in the semiconductor component, which can currently be determined essentially by the corresponding ratio of the lateral front conductive structure width to the lateral front electrode width (if a front conductive structure is present in the semiconductor body) or the lateral back conductive structure width to the lateral back electrode width (if a back conductive structure is present in the semiconductor body), so that there, due to the high current density, a fusing (melting) occurs in the current-carrying area. In the fusing, which is also called a short circuit, the material of the external metal terminal (for example aluminum, molybdenum and / or copper) and the silicon of the semiconductor body melt. After the melting process, a highly conductive metal area is formed, which permanently short-circuits the external metal terminal and can, in principle, also maintain this short-circuit effect for a longer time under current load, for example until the next repair.

[0047] Compared to conventional thyristors, in which the ratio of the width of the lateral conduction structure to the width of the lateral electrode is substantially 1 in order to achieve a maximum surge current and minimum conduction losses, in the short-circuit semiconductor device according to the invention, the current carrying capacity of the front and / or back conduction structures is limited in such a defined manner due to the ratio being less than 1, respectively defined therein, that the pre-destruction of the short-circuit semiconductor device according to the invention safely leads to subsequent complete melting. After melting, the short-circuit semiconductor device is in a permanent, irreversible, low-ohmic current-carrying state between the front and back electrodes, so that a voltage applied between the two electrodes can be short-circuited permanently and in a long-term stable manner. In other words, the voltage drop between the front electrode and the back electrode of the short-circuit semiconductor device according to the invention in the case of a short circuit is so small (low-ohmic current-carrying state) that the electrical losses in the short-circuit semiconductor device according to the invention are also small in a long-lasting short-circuit operation, and active cooling can therefore be advantageously omitted.

[0048] In the sense of the invention, the condition "less than 1" should always be designed such that production-related dimensional deviations between the width of the respective lateral conducting structure and the lateral electrode width of the electrode respectively corresponding to the conducting structure under consideration, which dimensional deviations are not included in the condition "less than 1" used here, are within the manufacturing tolerances defined for the respective production method of the short-circuit semiconductor device according to the invention. In other words, according to the above definition, in the sense of the invention, two compared lateral widths can only be regarded as "less than 1" if their dimensions differ so much from one another that the ratio of the lateral conducting structure width to the lateral electrode width of the respective electrode, so that the dimensional differences are no longer attributable to production-related tolerance deviations of the relevant features, but are the result of the target operation.

[0049] It has been shown that, given a vertical thickness of the semiconductor inner region of the semiconductor body, the melting current is substantially proportional to the area of ​​the front and / or back conductive structure and can therefore be determined. In addition, a lateral conductive structure width of the respective conductive structure that is (substantially) small compared to the lateral electrode width of the individual electrodes allows local limitation of melting on the short-circuited semiconductor component, the expansion of the melting being able to be substantially limited in the lateral or radial direction to the surface of the front and / or back of the conductive structure. As a result, the area through which the melting occurs can in any case be sufficiently far away from the radial outer edge region of the short-circuited semiconductor component according to the invention. As a result, the melting occurs only in the inner region of the short-circuited semiconductor component covered by the front and back electrodes, so that in the short-circuited operation, mechanical destruction of the short-circuited semiconductor component does not occur in the edge region of the short-circuited semiconductor component, which mechanical destruction could lead to the escape of plasma from the short-circuited semiconductor component and thus to the destruction of the housing surrounding the short-circuited semiconductor component.

[0050] It should also be noted here that for each arbitrarily observed position of the short-circuit semiconductor device according to the invention, the ratio of the lateral front conductive structure width to the lateral front electrode width is less than 1 or the ratio of the lateral back conductive structure width to the lateral back electrode width is less than 1. If the short-circuit semiconductor device according to the invention is, for example, constructed as a completely rotationally symmetrical cylindrical structure with front and / or back conductive structures and corresponding electrodes (front and / or back), the observation of the above ratios at any position on the short-circuit semiconductor device also directly applies to all other positions on the short-circuit semiconductor device. However, if in the exemplary selected cylindrical short-circuit semiconductor device the lateral conductive structure width and / or the lateral electrode width vary depending on the observed position along the circumference of the short-circuit semiconductor device, that is, the short-circuit semiconductor device is not constructed rotationally symmetrically with respect to the observed conductive structure and / or the electrode associated therewith, the ratio of the lateral conductive structure width to the lateral electrode width can also vary depending on the observed circumferential position. However, according to the above definition, in the sense of the invention, this should be less than 1 at each arbitrarily observed position on the short-circuit semiconductor device.

[0051] Furthermore, it should be understood that, in the sense of the present invention, in the case where the conductive structure does not essentially have a constant conductive structure width in the radial direction, for the ratio between the conductive structure width and the electrode width of the electrode corresponding to the relevant conductive structure, the maximum conductive structure width occurring in the radial direction is regarded as the conductive structure width to be used for comparison.

[0052] Particularly preferably, the ratio of the lateral conduction structure width to the lateral electrode width of the corresponding electrode associated with the front or back conduction structure is selected to be less than 4 / 5, for example also selected to be less than 3 / 4, 2 / 3, 1 / 2, 1 / 3, 1 / 4, 1 / 5. Therefore, the lateral conduction structure width of the front or back conduction structure is selected to be substantially smaller than the lateral electrode width of the corresponding electrode in the short-circuit semiconductor device according to the present invention. As a result, the fuse current can be determined to a desired small value, at which the melting of the short-circuit semiconductor device according to the present invention is safely caused. It is thus ensured that even relatively small fault currents or short-circuit currents conducted by the short-circuit semiconductor device according to the present invention between the front electrode and the back electrode safely melt the device in the desired manner and thus establish an irreversible, low-ohmic current-carrying state (short-circuit operating state).

[0053] Another advantageous embodiment of the invention provides that the conductive structure is completely covered by the front electrode and the back electrode, wherein this is to be understood as covering in the lateral (radial) direction. Complete front and back covering of at least one conductive structure offers the advantage that the probability of molten plasma escaping from the semiconductor body is further reduced, since the melting in the short-circuiting semiconductor component according to the invention can be essentially locally limited so that it occurs in the area covered by the two electrodes, as described above. Therefore, due to the blocking of the plasma escape from the short-circuiting semiconductor component in this way, damage to the housing that may surround the short-circuiting semiconductor component during melting is reliably prevented. Preferably, the covering of the corresponding conductive structure by the front or back electrode in the radial outer edge region (i.e. in the direction towards the radial outer edge of the short-circuiting semiconductor component according to the invention) can be in the range of at least several millimeters, for example approximately 1 mm to 5 mm or more.

[0054] Furthermore, at least one conductive structure can preferably be arranged close to the center in the semiconductor body. This is to be understood as meaning that the radial distance from the inner side of the conductive structure bounded radially inwards to the corresponding center of the semiconductor body, i.e. the geometrical center of gravity or center of the front and / or back side of the semiconductor body (also referred to as radial center) is smaller than the distance from the outer side of the conductive structure bounded radially farthest outwards to the radial outer edge of the semiconductor body. This measure also further reduces the probability of plasma escape in the edge region of the semiconductor body in the event of melting of the short-circuited semiconductor component according to the invention, since the melting is limited to a close to the center region of the semiconductor body between the front and back electrodes, which region is just covered by these electrodes. As a result, possible destruction of the housing surrounding the short-circuited semiconductor component due to melting is also reliably prevented.

[0055] A further advantageous embodiment of the invention provides that the front electrode and the back electrode cover the center of the front side or the back side of the semiconductor body and that the conductive structure is arranged in this center. As a result, the distance from the outer side of the conductive structure bounded radially outwards to the radial outer edge of the semiconductor body can be maximized and thus, depending on the selected lateral electrode width of the front and back electrodes, the edge region of the semiconductor body and further regions between the conductive structure and the radial outer edge of the semiconductor body are protected from undesired plasma escapes due to melting of short-circuited semiconductor components by covering the front side and / or the back side of the semiconductor body, which lateral electrode width is always greater than the lateral conductive structure width according to the invention.

[0056] Another advantageous embodiment of the present invention provides that an electrically insulating insulation layer is arranged between a conductive structure and an electrode conductively contacting the conductive structure, wherein the insulation layer extends in a radial direction on the conductive structure and in a circumferential direction of the semiconductor body along an inner edge of the electrode located inwardly in the radial direction.

[0057] It should be understood that the insulating layer does not extend so wide in the planar direction that the electrode is completely electrically insulated from the conductive structure and the base region embedded in the conductive structure. Rather, at least one region between the electrode and the conductive structure and the base region remains free of the insulating layer, so that an electrically conductive connection still exists between the electrode and the conductive structure and the base region.

[0058] The insulating layer may preferably have a thickness of at least 10 nm to at most 100 nm, more preferably at least 50 nm to at most 100 nm.

[0059] The insulating layer may preferably be a silicon dioxide layer.

[0060] Furthermore, according to a further advantageous embodiment of the invention, the insulation layer can have openings which are spaced apart from one another in the radial direction and which produce a direct electrically conductive connection between the electrode and the base region adjoining it.

[0061] According to a further advantageous embodiment of the invention, the trigger structure is a gate electrode and the turn-on signal is a surge current injected into the semiconductor body by the trigger structure.

[0062] An exemplary gate structure includes a gate spring line contacting a gate electrode from the outside. The gate spring line is pressed onto a gate terminal of a semiconductor body with its inherent stress.

[0063] It is conceivable that another form of spring gate structure could be used in place of the spring gate structure which creates a more planar transition between the gate supply line and the gate contact, for example to keep the contact transition resistance as low as possible.

[0064] The transition resistance or contact resistance to the gate electrode is critical, especially in the case of current pulses which are still very high compared to the usual gate currents (in the range of 10 A), and which are in the range of up to several hundred amperes, if not kiloamperes. The contact resistance to the silicon must be kept low and stable by a suitable embodiment. For this purpose, suitable contact materials or surface coatings must be selected. Therefore, these surfaces must not be corroded.

[0065] The contact force must also be set particularly precisely in order to avoid excessively high transition resistances in the case of these high current pulses.

[0066] If this is unsuccessful, this may lead to failure directly at the transition of the gate supply line to the gate metal (risk of plasma escape since this area is not in a pressurized region).

[0067] In order to ensure the long-term stability of the low-ohmic contact transition, a surface coating is considered in the transition area from the gate supply line to the gate metal on the semiconductor surface. If, for example, nickel plating is insufficient, suitable long-term stable or particularly corrosion-resistant surface coating materials for coating the gate supply line are, for example, ruthenium or rhodium. Further improvements can be made using precious metal coatings such as Au. Thus, not only the gate line can be coated, but also the metal surface of the gate contact of the semiconductor. In order to improve adhesion, adhesion promoters such as titanium can also be used in addition.

[0068] Furthermore, it is helpful to provide the (gate) metallization of the semiconductor with a particularly thick (soft) silver layer, which helps to reduce the contact resistance with the usually hard or relatively stiff gate supply line (also due to elastic effects). In other words, the gate electrode can be coated preferably on its metal contact side located on the outside in the vertical direction with a silver layer which is relatively thick on the one hand and significantly softer on the other hand than the outer gate electrode contact line (spring line) contacting the silver layer.

[0069] Since these measures can be cost-intensive, the goal is to have a contact system that is as simple as possible, i.e. without additional expensive coatings. This is easier to achieve the lower the current amplitude of the discharge current / surge current. Other measures described below (e.g. small diameter, small distance between gate and conductive structure / electrode and small circumference of conductive structure) have the effect that the necessary discharge current / surge current can be generated and therefore also the energy input that is as low or as small as possible.

[0070] Another advantageous embodiment of the invention provides a capacitor with a predetermined capacitance corresponding to a predetermined energy input, which capacitor provides a conduction signal in a charged state, wherein the capacitor can be electrically switched between the gate electrode and the front electrode and / or the back electrode by means of an electrical switch.

[0071] The capacitor can be charged, for example, by a special switching device, so that the capacitor energy comes from the load circuit applied to the semiconductor device. The charging of the capacitor can also be realized by the gate electronic control device or the capacitor is also located on the primary circuit side of the trigger transformer.

[0072] In conventional thyristor structures, the diameter of the gate center is in the range of 10 mm to 20 mm, i.e. from the midpoint (center) of the semiconductor device to the start of the cathode emitter (front-side conduction structure), typically 15 mm for larger thyristors with a device diameter greater than 3". Due to this spatial extent, the discharge energy and thus the energy density must be chosen to be large enough to be able to produce a safe pre-destruction.

[0073] On the other hand, however, the destructive energy stored by the capacitor cannot be chosen to be arbitrarily large. The energy stored in the capacitor is (C×U^2) / 2=Q^2 / (2×C). In order to obtain a high current for destruction, a sufficiently high capacitance must be chosen, for example 4mF. For a charging voltage of 30V to 40V, an energy of 1.8Ws to 3.2Ws can thus be achieved. From this, the achievable current (ΔQ / Δt) can be estimated. Assuming a (discharge) time unit of 100μs and determining the amount of charge ΔQ stored in the capacitor from the energy, a high current pulse / surge current in the range of 1.2kA to 1.6kA will result. Assuming that this current flows radially symmetrically from the gate contact to the inner edge of the electrode (e.g. cathode) (assumption: the current flows directly from the gate metal to the circular inner edge of the cathode metal, but does not flow into any gate fingers), a current density of about 20 to 30A / mm for a specific line is generated at the start of the inner circular cathode emitter (circumference of the circular cathode inner edge: about 50mm). If 50 μm or 0.05 mm is assumed for the depth of the conductive base region, a current density in the range of 500 A / mm2 is obtained at the start of the internal cathode emitter. This value must ultimately be large enough to produce the desired pre-destruction in the region of the internal cathode emitter, wherein the very short current duration must also be taken into account. The shorter the current duration or pulse length, the higher the applied current or charge amount must be selected.

[0074] In conventional thyristor structures, the region of the p-type region with high ohms relative to the metal is located between the metal edge of the gate contact and the inner edge of the cathode emitter, and one or more ring-shaped amplifier gates are additionally present. Both lead to an increase in resistance in this region, which causes a correspondingly high voltage drop during this pulse-shaped current load. However, this can lead to failures in the amplifier gate region or to undesired pre-destruction in this region. However, this must be avoided in any case, because this region is not located in the pressure-bearing area of ​​the contact electrode. Melting at this location leads to a "blow-out" of the plasma from the gate feed (the location in the ceramic capsule through which the gate power supply line is conveyed).

[0075] Therefore, the semiconductor device according to the present invention preferably has the following features:

[0076] The distance between the radially outer metal edge of the gate terminal G and the radially inner edge of the conduction structure (emitter structure) is minimal to avoid damage in the area between the gate terminal and the conduction structure, i.e. in a possibly unstressed area;

[0077] ·Omit the amplifier gate level

[0078] The diameter of the inner edge of the circular gate metal terminal and the conduction structure is as small as possible (e.g., only 1.75 mm to a few mm)

[0079] Do not use long gate fingers, as this will reduce the specific current density or energy density that causes damage, where the current-carrying surface is the product of the circumference and depth of the base region (about 0.05 mm);

[0080] The inner diameter of the emitter or the lead-through structure should be as small as possible. This also offers the advantage that the energy input for pre-destruction can be reduced and thus the capacitance of the capacitor can also be reduced. For example, instead of a circular lead-through structure, only a ring segment-shaped lead-through structure can also be used;

[0081] By providing a high-ohmic resistance region in the base region between the inner edges of the gate metal and the cathode metal, the discharge current / surge current is directed in a targeted manner only to a small portion of the inner edge of the conductive structure; as a result, the discharge current / surge current flows primarily only between the gate metal and the conductive structure or a small region of the conductive structure, which leads to an increase in the energy density there.

[0082] Particularly preferably, at least one region of the base region containing the emitter structure, which forms an electrically operative connection between the gate electrode and the emitter structure, can be designed such that an electrical signal supplied to the gate electrode can be used to switch on / pre-destroy the emitter structure, the electrical signal having a current strength of at least approximately 1 A, preferably at least approximately 2 A and even more preferably between approximately 5 A and approximately 10 A. It is to be understood that a region of the base region embedded in the conduction structure between the trigger structure and the corresponding conduction structure can be designed such that the conduction structure can be switched on by means of the trigger structure with an overload factor of at most 10, preferably at most 5, more preferably at most 3, wherein a minimum conduction current that causes switching on / pre-destruction of the conduction structure is at least approximately 1 A, preferably at least approximately 2 A and even more preferably between at least approximately 5 A and at most approximately 10 A.

[0083] The trigger current or conduction current of a conventional thyristor depends on the surface resistivity of its cathode-side base region, in which the cathode-side emitter is embedded, and must be set in terms of process technology in such a way that the trigger current value or conduction current value is typically in the mA range, in particular assuming values ​​between 30 mA (125° C.) and 400 mA (room temperature), and accordingly in a relatively narrow range. It can thus be ensured that, for example, a conventional thyristor safely withstands high conduction voltages when switching very high currents at very high current slopes at the start of conduction, when it is controlled by a conventional trigger or conduction current generator with a precisely defined trigger or conduction current profile (trigger / conduction current amplitude and duration).

[0084] Since the short-circuit semiconductor device according to the present invention should be molten after a trigger, a significantly higher minimum on-state current is preferably selected for the short-circuit semiconductor device, especially in the ampere range, for example at least 1A or at least 2A or even at least 5A to about 10A, because on the one hand the cost for controlling the surface resistivity of the base region containing the emitter structure when producing the short-circuit semiconductor device can be significantly reduced, and on the other hand significantly lower requirements can be placed on the trigger or on-state current generator that generates the electrical conduction signal, thereby making it possible to implement a more cost-effective control scheme.

[0085] Furthermore, it should be understood that, despite the selection of a relatively high trigger current or forward current in the short-circuit semiconductor device according to the invention according to this embodiment, the short-circuit semiconductor device according to the invention is still controlled with only a slight overmodulation, i.e., it is operated with an overmodulation of, for example, 10 A / 5 A=2. In contrast, if a trigger current or forward current is used in the ampere range in conventional thyristors, for example in the range between 5 A and 10 A, instead of in the mA range as described above, this results in an overmodulation of the conventional thyristors of more than one or two orders of magnitude, with an overmodulation factor of, for example, 10 A / 100 mA=100. A large overmodulation factor enables a conventional thyristor to be turned on quickly.

[0086] In the case of the short-circuit semiconductor component according to the invention, an overmodulation which is selected to be weak (but with a high trigger or forward current in the ampere range) promotes the desired failure of the component, i.e. the desired one-time irreversible melting. In the sense of this embodiment, the provision of the maximum overmodulation ensures in particular a minimum energy input through the trigger structure which is necessary for the desired pre-destruction.

[0087] Furthermore, a high on-current brings other advantages. For example, a simple detection can be provided to check the readiness of the short-circuited semiconductor device by a continuous test current between the gate electrode and the electrode (front or back electrode) in contact with the emitter structure to be switched on / destroyed by the gate electrode, wherein the test current is selected to be much smaller than the on-current. Furthermore, a high on-current also provides a higher EMC protection, for example with respect to false triggering due to very high voltages and rapid changes of currents in the vicinity of the short-circuited semiconductor device, since the sensitivity to such false triggering increases with decreasing trigger or on-current.

[0088] In addition to the above-mentioned forward current strengths, in certain cases the forward current can also be set in the short-circuit semiconductor component according to the invention to be clearly less than 1 A, for example less than 1 A up to less than approximately 500 mA or even less than approximately 100 mA.

[0089] According to another advantageous embodiment of the invention, the trigger structure is a breakdown structure and the switch-on signal is a voltage applied between the front electrode and the back electrode. The breakdown structure is characterized in that, depending on the voltage applied at the external electrodes of the short-circuited semiconductor component, which voltage causes a corresponding voltage drop across the breakdown structure, a sharp current increase in the form of an avalanche current is allowed when a breakdown voltage that can be determined for the breakdown structure is reached or exceeded, which current increase is preferably not limited by a protective resistor, by which a pre-destruction of at least one of the two semiconductor junctions by the breakdown structure configured as a trigger structure can be achieved.

[0090] The breakdown structure can be configured, for example, as a BOD structure (BOD: Break Over Diode). Other breakdown structures with similar functions can also be used, for example breakdown structures known per se from conventional light-triggered thyristors (LTTs), for example breakdown structures with conventional du / dt protection and / or breakdown structures with conventional circuit commutation turn-off time protection with local increase of carrier lifetime. In addition, the breakdown structure is another measure for avoiding blocking failures in the radial outer edge region of the short-circuited semiconductor component in the event of an unforeseen overvoltage. The level of the breakdown voltage of the breakdown structure is set in such a way that it takes its lowest value of the entire semiconductor body at the precise position of the breakdown structure, thereby safely preventing breakdown in the edge region. For this purpose, the breakdown structure is preferably arranged in the semiconductor body away from the edge, in particular close to the center. For example, the breakdown structure can be configured as a circumferentially closed structure surrounding the center, but (slightly) spaced apart from the center, for example as a ring and / or polygonal structure. However, in other cases, it can also be arranged directly in the center of the semiconductor body. In this case, a circumferential direction is understood to be a direction extending parallel to the front side or rear side of the semiconductor body and substantially perpendicular to a radial direction or a lateral direction of the semiconductor body.

[0091] In an embodiment of the invention which is particularly easy to manufacture in terms of its construction, the conductive structure has a closed circumference, for example, due to the trigger structure arranged in the center of the semiconductor body, the conductive structure is arranged around the center and is spaced apart from the trigger structure. The conductive structure can be designed, for example, as a closed annular and / or polygonal structure. In this case, the extension direction of the conductive structure parallel to the front side or back side of the semiconductor body and substantially perpendicular to the radial direction or lateral direction of the semiconductor body is understood as a circumferential direction.

[0092] According to another advantageous embodiment of the present invention, the conduction structure is spaced apart from the trigger structure in the radial direction and has at least one conduction structure segment, which has a conduction structure width extending in the circumferential direction of the semiconductor body, which conduction structure width is smaller everywhere than the closed circumference through the conduction structure segment in the circumferential direction. In other words, in this embodiment, the conduction structure is not closed in the circumferential direction, as, for example, in the case of a closed annular structure and / or polygonal structure. That is, the circumferential conduction structure width of at least one conduction structure segment is in any case smaller than the circumferential conduction structure width that would correspond to the closed annular structure and / or polygonal structure of the conduction structure and therefore to a circumferential angle of 360 degrees of the closed annular structure and / or polygonal conduction structure, which conduction structure width, for example, could correspond to only one ring segment. Since during the production process the conductive structure which is radially spaced apart from the center of the semiconductor body and has a polygonal and / or annular structure which is closed in the circumferential direction has a certain minimum area, since, for example, the lateral conductive structure width cannot be produced in terms of process technology to be smaller than a certain lateral minimum structure width, which minimum structure width can be, for example, at least approximately 50 μm, the area of ​​the conductive structure which is effective for the fusing of the semiconductor body can be further reduced by means of the currently advantageous embodiments of the short-circuit semiconductor device according to the invention, whereby also particularly low fusing currents can be achieved, which lead to a safe short circuit of the semiconductor body. Thus, for example, it is possible to provide a minimum individual conductive structure section in the semiconductor body which has a lateral conductive structure width of approximately 50 μm and a circumferential conductive structure width, i.e. a circumference, which is likewise approximately 50 μm, which allows the design of a short-circuit semiconductor device according to the invention with a fusing current in the order of several mA.

[0093] According to another advantageous embodiment of the invention, at least two mutually separated conductive structure segments are provided in the circumferential direction of the semiconductor body. These semiconductor components are preferably spaced equidistantly from one another in order to achieve a substantially circumferentially symmetrical embodiment of the short-circuit semiconductor component, which not only leads to a uniform current distribution of the conductive current on the semiconductor body for conducting or predestruction of the conductive structure, but also to a uniform current distribution of the melting current on the semiconductor body for the generation of melting on the semiconductor body, and thus to a uniform heat distribution, that is to say a uniformly distributed mechanical load, over the entire semiconductor body, in particular during a short circuit.

[0094] A further advantageous embodiment of the invention provides that, at least in a part of the radial gap between the conduction structure and the trigger structure (also referred to herein as the resistance region of the base region), in addition to at least one forward channel electrically coupling at least one conduction structure segment with the trigger structure, the resistance in the base region containing the conduction structure is increased. As a result, the conduction current flowing from the trigger structure to the conduction structure can be directed in a targeted manner to the individual conduction structure segments in order to safely conduct or pre-destroy them if necessary in order to initiate melting of the short-circuited semiconductor component according to the invention. By directing the conduction current in a targeted manner by means of the flow channel to at least one conduction structure segment, the total conduction current that is required for the conduction / pre-destruction of all conduction structure segments provided by the trigger structure can be additionally reduced in an advantageous manner compared to a conduction current distribution that is otherwise uniformly distributed in all radial directions.

[0095] An increased resistance in the resistance region of the base region compared to the resistance value in the through-current path is understood to be a resistance size deviation which is not determined by a manufacturing-related size deviation due to generally acceptable manufacturing tolerances, but is the result of a targeted process. For example, the resistance in the resistance region can be increased in a targeted manner by significantly reducing the concentration of charge carriers of the same conductivity type as in the rest of the base region and / or by introducing a so-called pinch resistor of a conductivity type complementary to the conductivity type of the base region into the resistance region of the base region, the resistance in the resistance region of the base region also being increased in a targeted manner due to a reduction in the effective conducting cross section in the base region.

[0096] In the case of using only small (ring-shaped) conduction structure sections, these should be controlled in the event of a fault so that complete melting occurs there. The problem with such structures is that, due to the relatively high specific conductivity of the base region, the resistance between the metal edges of the gate contact and the electrode contact is so low that most of the current, which should actually destroy the conduction structure locally, flows into the entire metal contact of the electrode metallization, and the energy density of the discharge current pulse / surge current is thereby reduced so strongly that no destruction occurs. Therefore, as described herein, an additional resistor or a resistor region can be introduced into the base region.

[0097] As an alternative or in addition to the additional resistance area, it is also conceivable to apply the very thin insulating layer already described above in the nanometer range in the entire area of ​​the electrode contact. For example, a full-surface SiO2 layer can be produced on the semiconductor surface (front or back) of the semiconductor body. With the help of photolithography, this insulating layer can be removed in the gate area or in the inner edge of the conductive structure and in small areas of the conductive structure, so that direct contact is produced between the silicon surface of the semiconductor body and the metallization of the electrode only in these areas without the insulating layer. Therefore, the largest area of ​​the electrode metallization does not directly contact the silicon / semiconductor body surface. In this way, the current is "forced" to flow directly to the segmented conductive structure area, which now also corresponds to the area not covered by the additional insulating layer. In this way, the additional resistance area can also be omitted.

[0098] However, using this insulating layer, other embodiments are possible as will be described below.

[0099] Typically, the conduction structure must not be covered radially inwards by the electrode metallization or the contact pad, because otherwise the conduction current flows from the gate contact G directly via the base region to the electrode terminal without switching on or triggering the conduction structure, because the electrode contact would in this case bridge or short-circuit the conduction structure. That is, for all semiconductor devices without an insulating layer, the conduction structure protrudes radially inwards under the electrode contact.

[0100] With the aid of the additional locally applied insulating layer described here, preferably at the end of all high-temperature steps for producing the semiconductor body, a very thin thermal oxide layer (e.g. SiO2) is produced, which is then photolithographically structured, and the electrode contact or contact pad can now reach the radially inner inner edge of the conductive structure without bridging this inner edge. As a result, the pre-destructed area is always located below the compressed contact area. This measure is a further improvement. Whether the end is necessary depends on whether the protrusion of the conductive structure or the emitter radially inwards towards the gate area under the electrode contact has a critical influence. In the worst case, the emitter layer reaches the gate contact far enough below the electrode contact that the pre-destruction occurs there and at the end the plasma can escape outwards towards the gate power contact (the pressure wave is "blown" towards the gate contact).

[0101] If the insulating layer is applied over a large area or only at a position where the conductive structure should be triggered or pre-destructed, and the fault load current at this position produces melting afterwards, there is no direct electrical contact transition between the metallization of the corresponding electrode and the semiconductor surface at this position, which is a local area of ​​the base region or the conductive structure constructed as an emitter. This may lead to a breakdown between the insulating layer and the metallization, because the blocking current in the semiconductor body has a wide path in the insulating region until the next directly coupled contact region. As a result, a high transverse (lateral) voltage parallel to the semiconductor surface appears in the base region under the insulating layer, which can then cause electrical breakdown. In order to avoid this situation, the insulating layer should have an opening or groove described herein at a certain distance, similar to the short-circuit structure in the large-area conductive structure. Then the blocking current can flow out to the corresponding electrode terminal when the blocking voltage is applied, and no excessive voltage will appear in the base region to which it belongs. This can be an annular groove or a short-circuit hole, which must be sufficient in size to avoid the undesirable lateral voltage drop in the base region caused by the reverse current. Each notch must be of sufficient size to prevent an undesired lateral voltage drop in the base region, which is caused by blocking current. Such interruptions can also be omitted if the lateral width of the insulating layer formed as an insulating ring is not too wide.

[0102] Furthermore, with the aid of an additional insulating layer, the lateral position of the conduction structure can be moved arbitrarily far from the inner edge of the electrode contacting the base region or moved outwards by any distance. This can be of interest when there is a risk that, if a fault occurs with currents in the 100 kA range, melting can extend to the non-pressure contact area of ​​the gate terminal (there is a risk of blowing in the area of ​​the gate wiring).

[0103] The formation of the insulating layer using a thermal oxide, such as SiO2, has the advantage that it can be produced by conventional diffusion processes at the end of a high-temperature step or before the deposition of the metal layer, i.e. by thermal oxidation in an oxygen atmosphere at high temperatures in the range of 800° C. to 900° C. However, if adhesion of the metal coating is unfavorable due to various process-related reasons, another insulating layer, such as Si3N4 (silicon nitride), can also be deposited. This insulating layer can also be produced by vapor deposition at temperatures far below 1000° C. using a suitable process, such as LPCVD (low-pressure chemical vapor deposition). As a pretreatment for metal deposition, etching chemicals can be used to safely remove the natural oxide layer (in the range of 20 nm to 40 nm) on the Si surface, such as a buffered HF etching mixture. Unlike thermal oxides, nitride layers are resistant to HF-based etching solutions.

[0104] For all deposited insulating layers, it is important that their thickness is not chosen too large compared to the thickness of the metal layer of the electrode, in order to avoid steps at the edge of the insulating layer. Therefore, the insulating layer thickness should be chosen in the range of 50 nm to 100 nm. Compared to the thick electrode metallization, which is typically in the range of 10 μm to 30 μm, the steps at the transition to the insulating layer are negligible.

[0105] The insulating layer should be applied only locally, but always in such a way that the electrical contact between the conductive Si surface of the semiconductor body and the electrode metallization locally has an extremely high impedance or is completely avoided, but as described above, there must (still) be sufficient direct electrical contact between the electrode and the corresponding base region.

[0106] Since the breakdown structure designed as a trigger structure or a conduction structure as described above obtains its conduction signal from the voltage applied to two electrodes outside the semiconductor body, it is not necessary to provide the short-circuited semiconductor device with other external conduction signals, such as optical or electrical conduction signals. If the conduction structure or the trigger structure assigned to the conduction structure is designed only as a breakdown structure, a diode housing can be used in a particularly advantageous manner to accommodate the short-circuited semiconductor device designed in this way, which has no additional connection for the additional external conduction signal. In this case, the breakdown structure can be particularly preferably arranged in the center of the semiconductor body, and the front electrode and the back electrode can extend completely in the center and thus on the conduction structure, because no groove for providing an external conduction signal needs to be provided.

[0107] According to a further advantageous embodiment of the invention, the conduction structure has at least one conduction structure finger extending radially outwards from the rest of the conduction structure in a lateral direction (also referred to as fingering of the conduction structure). In other words, at least one conduction structure finger projects radially outwards from the rest of the conduction structure. As explained above, the minimum melting current integral, at which melting of the short-circuited semiconductor component should occur, is essentially determined by a lateral conduction structure width of the corresponding conduction structure, for example by a lateral emitter structure width of the emitter structure or by a lateral breakdown structure width of the breakdown structure, and similarly also by an effective area of ​​the conduction structure when the semiconductor body melts. In other words, the present invention aims to achieve a small ratio between the lateral conduction structure width of the corresponding conduction structure and the lateral electrode width of the electrode applied to the base region having the conduction structure, so that the desired melting current is small and the short-circuited semiconductor device is melted safely under the melting current, so that the relatively small fault current or short-circuit current conducted by the short-circuited semiconductor device between the front and back electrodes safely melts the semiconductor device in the expected manner, thereby producing an irreversible low-ohmic power-on state (short-circuit working state).

[0108] On the other hand, the effective area of ​​the conducting structure, for example the effective area of ​​the emitter structure or the breakdown structure, and thus the entire area of ​​the semiconductor body should generally be matched to the intensity of the short-circuit current that is continuously received by the short-circuited semiconductor device in the case of a short circuit, in order to prevent the melting process from starting explosively just at the beginning of the current flow when the area of ​​the conducting structure is too small, which may lead to mechanical destruction of the entire short-circuited semiconductor device and thus may endanger the surrounding environment. A larger area of ​​the conducting structure can have a favorable influence on the size of the lateral expansion of the melting area, wherein a further lateral expansion of the melting area advantageously leads to a reduction in the resistance of the short-circuit area formed by melting in the semiconductor body due to the larger conductive cross section. The fingering of the conducting structure advantageously achieves a compromise between the minimum possible area of ​​the conducting structure to reduce the relative requirements of the short-circuit or melting current required for safe melting and a sufficiently large area to safely conduct the short-circuit current without mechanical destruction of the short-circuited semiconductor, because the current-carrying area of ​​the short-circuited semiconductor device that is connected by the conduction signal can be distributed to a greater extent by the finger-shaped conducting structure.

[0109] The finger-like structure of the conductive structure also helps to keep the melting zone away from the center of the short-circuited semiconductor device. This is particularly advantageous when there is a trigger structure in the center of the semiconductor body, which is provided with an external turn-on signal (e.g. an optical or electrical signal) for turning on the conductive structure or the short-circuited semiconductor device. In this case, the area where the turn-on signal is provided to the trigger structure is not covered by an electrode (e.g. a cathode or an anode) that provides the main electrical connection for the short-circuited semiconductor device and is applied to the base region with the conductive structure, so that when melting occurs, plasma may escape from the uncovered front and / or back areas of the semiconductor body, which may damage the housing around the short-circuited semiconductor device and thus endanger the environment.

[0110] Another advantageous embodiment of the invention provides that, if at least one trigger structure and a finger-shaped conductive structure as described above are present, the trigger structure has at least one trigger structure finger which corresponds to or follows the contour or shape of at least one conductive structure finger. This ensures that the trigger structure is spaced uniformly from the conductive structure over the entire length of the conductive structure, for example in the transverse or radial direction and / or in the vertical direction, so that the trigger structure can be uniform over the entire length of the conductive structure in the sense of the invention.

[0111] According to another preferred embodiment of the present invention, at least one of the conductive structure fingers has a maximum lateral (i.e. radial) finger length that is less than 4 / 5 of the lateral electrode width of the electrode applied to the base region having the conductive structure. Thus, for example, the maximum radial finger length may be less than 3 / 4, 2 / 3, 1 / 2, 1 / 3, 1 / 4, 1 / 5 of the lateral electrode width. If the trigger structure is also provided with a finger structure, as described above, which corresponds to or follows the finger structure of the conductive structure, then at least one of the trigger structure fingers may also have a finger length corresponding to the finger length of each conductive structure finger.

[0112] According to a further advantageous embodiment of the invention, the thickness of the inner region in the vertical direction is greater than 400 μm, more preferably greater than 600 μm, and even more preferably greater than 800 μm or 900 μm or even 1000 μm. It can generally be determined that in the event of an overvoltage, the risk of edge failure of the short-circuited semiconductor device is prevented by the significantly thicker inner region of the semiconductor body compared to conventional thyristors. In practice, there is no maximum thickness for the inner region as in the case of conventional thyristors, since for the short-circuited semiconductor device according to the invention there are no conduction losses or dynamic losses which limit the thickness of the inner region upwards. Therefore, for the required voltage levels, for example V DRM / V RRM =2.2kV, thickness W will be used n1 The conventional thyristor has a silicon resistivity of 85 Ω·cm and a thickness of about 320 μm. However, for the short-circuit type semiconductor device of the present invention, for example, a thickness W of about 850 μm may be selected. n For the semiconductor structure of the semiconductor body according to the invention, with a resistivity of 270 Ω·cm, more than twice the symmetrical blocking capability (V DRM / V RRM =5.2kV). With a corresponding design of the thickness of the inner region, a self-blocking capability of >10kV (medium voltage grid) is possible. By means of a high radiation dose for electron radiation, the blocking current can be set very low and the lifespan adjusted accordingly. Therefore, the blocking capability of the short-circuit semiconductor device can be set much higher than the maximum overvoltage that occurs when the short-circuit semiconductor device is used, and this has no adverse effect on the short-circuit semiconductor device and its application. Therefore, a blocking current that is as small as possible is very advantageous for the short-circuit semiconductor device according to the present invention, because the blocking voltage is actually continuously applied to the short-circuit semiconductor device, so that a blocking current that is as small as possible can be obtained, and the resulting blocking losses can be kept as small as possible, so that the short-circuit semiconductor device can be operated without active cooling. This measure also substantially eliminates the occurrence of faults in the edge region of the short-circuit semiconductor device that are critical to the shell rupture current.

[0113] According to another advantageous embodiment of the present invention, the front electrode and the back electrode are respectively contacted by conductive pressure via a metal contact pad. Each contact pad preferably has the same lateral width as the corresponding electrode, so that the contact pad completely covers the corresponding electrode. The contact pad can be firmly connected to the corresponding metal electrode of the semiconductor body, for example, by melting, or only loosely abuts on the latter. The contact pad can be designed as a pure molybdenum disk or composed of a combination of a molybdenum disk and a copper disk, preferably the molybdenum is in direct contact with the metallization of the semiconductor body. Here, as mentioned above, the present invention ensures in an advantageous manner that the area between the front electrode and the back electrode of the short-circuited semiconductor device produced by melting in the case of a short circuit is always located in the area of ​​the short-circuited semiconductor device contacted by pressure by the metal contact pad.

[0114] The thickness of the molybdenum disk or the combination of molybdenum and copper is preferably between 1 mm and 5 mm. It will be appreciated that the thickness of the contact disk can be adapted to the intensity of the short-circuit current flowing through the short-circuit semiconductor device. The melting in the short-circuit operation fuses the silicon of the semiconductor body and the metal area of ​​the contact disk to each other, so that a metal area is formed. Depending on the application or the intensity of the maximum short-circuit current to be permanently conducted, the thickness of the contact disk is designed so that the side of the contact disk away from the corresponding electrode of the semiconductor body does not show much change after the occurrence of the molten part. In particular, this is intended to maintain such a pressure contact during the short-circuit operation of the short-circuit semiconductor device, which can ensure a conductive, low-resistance, thermal coupling, and the pressure contact is located between the contact disk in contact with the metallized conductor and the corresponding contact die (for example, a copper die), which is a die of the housing that contacts the short-circuit semiconductor and has sufficient force to act on the contact disk, so that arcing can be reliably prevented outside the area contacted by the contact disk after the short-circuit semiconductor device is triggered. During the melting process, the molten zone must not penetrate the (copper) contact die of the housing under any circumstances.

[0115] It should be understood that according to the present invention, the above-mentioned different types of conduction structures can be combined with each other in a single short-circuit semiconductor device, especially if more than one conduction structure is provided in the short-circuit semiconductor device. If more than one trigger structure is provided in the short-circuit semiconductor device, the above-mentioned different types of trigger structures can also be combined with each other in a single short-circuit semiconductor device according to the present invention. Similarly, the above-mentioned different types of trigger structures can be combined with different types of conduction structures.

[0116] For example, the short-circuit semiconductor device according to the present invention can be turned on by one or more electrical conduction signals, and a corresponding conduction structure is set in the semiconductor body. The conduction signal can be only on one side, such as on the front side of the semiconductor body, or it can also be on both sides, that is, on the front side and back side of the semiconductor body. The short-circuit semiconductor device is provided with a corresponding trigger structure on the corresponding side of the semiconductor body.

[0117] Furthermore, the short-circuit semiconductor component according to the invention can have one or more breakdown structures as trigger structures or conduction structures, wherein a plurality of existing breakdown structures can be designed for the same breakdown voltage or respectively for different breakdown voltages.

[0118] Breakdown structures used as trigger structures, such as BOD structures, as well as breakdown structures with du / dt protection and breakdown structures with circuit commutation switch-off time protection with local increase of carrier lifetime, can be arranged within the amplification gate structure, i.e. laterally spaced apart from the front or back electrode and correspondingly arranged outside the region electrically contacted by these electrodes. However, the arrangement within the amplification gate structure is not absolutely necessary. The breakdown structure used both as trigger structure and as conduction structure can also be arranged vertically below the front electrode or back electrode, for example in the base region electrically contacted by the corresponding electrode.

[0119] Furthermore, the triggering and conduction structures described here can be arranged only on one side of the semiconductor body of the short-circuit semiconductor component according to the invention, ie only in the front or back base region, or also on both sides, ie in both base regions.

[0120] In summary, the definitions of terms used herein are listed again in summary form below:

[0121] Ⅰ.Conduction structure

[0122] This structure is used to produce melting in the semiconductor body of the short-circuit semiconductor device according to the present invention so that after it is turned on or pre-destroyed, a low-ohmic conductive state (short circuit) is established between the main electrical terminals (cathode / anode) outside the short-circuit semiconductor device that are electrically connected to the short-circuit semiconductor device and electrically separated from the semiconductor body under normal operation (non-short-circuit condition).

[0123] Possible implementations are:

[0124] a. Emitter structure

[0125] b. Breakdown structure (e.g. in the form of BOD structure, du / dt structure, circuit commutation off-time protection structure)

[0126] Ⅱ.Conduction signal:

[0127] A signal used to initiate conduction or pre-destruct conduction structures.

[0128] Possible implementations are:

[0129] a. A separate electrical signal applied externally to the semiconductor body according to the invention

[0130] b. Voltage / operating voltage applied to the main electrical terminals (cathode / anode) of the short-circuited semiconductor device according to the invention

[0131] Ⅲ. On-current:

[0132] A current obtained from an enable signal and supplied to a enable structure designed as an emitter structure in order to trigger or enable the enable structure.

[0133] IV. Fusing current:

[0134] An electric current that causes melting in the semiconductor body.

[0135] Ⅴ. Trigger structure:

[0136] An optional auxiliary structure is used to transfer the conduction signal that is always supplied to the auxiliary structure to the conduction structure.

[0137] Possible implementations are:

[0138] a. The turn-on signal provided to the gate electrode: II.a

[0139] b. The turn-on signal delivered by the breakdown structure (e.g. in the form of a BOD structure, du / dt structure, circuit commutation off-time protection structure): II.b BRIEF DESCRIPTION OF THE DRAWINGS

[0140] Other features and advantages of the present invention are derived from the following description of embodiments of the present invention, which should not be understood as limiting and are explained in detail below with reference to the accompanying drawings. In these drawings, schematically shown:

[0141] Figure 1 shows a vertical cross-sectional view of a thyristor according to the prior art;

[0142] Figure 2 A vertical cross-sectional view of an embodiment of a short-circuit semiconductor device according to the present invention having a turn-on structure and a trigger structure is shown;

[0143] Figure 3 A top view showing another embodiment of a short-circuit semiconductor device according to the present invention having a finger-shaped conductive structure;

[0144] Figure 4 A vertical cross-sectional view of another embodiment of a short-circuit semiconductor device according to the present invention having two conduction structures and one trigger structure is shown;

[0145] Figure 5 A vertical cross-sectional view of another embodiment of a short-circuit semiconductor device according to the present invention having two conduction structures and two trigger structures is shown;

[0146] Figure 6 A vertical cross-sectional view showing another embodiment of a short-circuit semiconductor device according to the present invention having a turn-on structure and a trigger structure;

[0147] Figure 7 A vertical cross-sectional view of another embodiment of a short-circuit semiconductor device according to the present invention having two conduction structures and two trigger structures is shown;

[0148] Figure 8 Shows Figure 2 Two circuit diagrams of various embodiments of a circuit of a short-circuited semiconductor device;

[0149] Fig. 9 A vertical cross-sectional view of another embodiment of a short-circuit semiconductor device according to the present invention having two conduction structures and two trigger structures is shown;

[0150] Fig.10 A vertical cross-sectional view of another embodiment of a short-circuit semiconductor device according to the present invention having two conduction structures and one trigger structure is shown;

[0151] Fig.11 In views (a), (b) and (c), the concentration profiles Na, Nb or Nc shown therein are shown. Fig.10 The concentration distribution of the short-circuited semiconductor device shown;

[0152] Fig.12 A vertical cross-sectional view of another embodiment of a short-circuit semiconductor device according to the present invention having two conduction structures and two trigger structures is shown;

[0153] Fig.13 A partial top view of another embodiment of a short-circuit semiconductor device according to the present invention is shown, which has a conduction structure with a conduction structure segment and a trigger structure which are open in the circumferential direction of the semiconductor body;

[0154] Fig.14 A partial top view of another embodiment of a short-circuit semiconductor device according to the present invention is shown, which has a conduction structure with a conduction structure segment and a trigger structure which are open in the circumferential direction of the semiconductor body;

[0155] Fig.15 Shown along section line XX through Fig.14 A vertical cross-sectional view of a short-circuited semiconductor device;

[0156] Fig.16 A partial top view of another embodiment of a short-circuit semiconductor device according to the present invention is shown, which has a conduction structure with a conduction structure segment and a trigger structure which are open in the circumferential direction of the semiconductor body;

[0157] Fig.17 A partial top view of another embodiment of a short-circuit semiconductor device according to the present invention is shown, which has a conduction structure with a conduction structure segment and a trigger structure which are open in the circumferential direction of the semiconductor body;

[0158] Fig.18 Shown along the section line YY through Fig.17 Vertical cross-section of a short-circuited semiconductor device. DETAILED DESCRIPTION

[0159] In the various figures, functionally identical components are always provided with the same reference symbols, so that these components are generally also described only once.

[0160] Figure 1 1 shows a vertical section through a thyristor according to the prior art, the thyristor 1 having a semiconductor body 2 which has essentially the shape of a very flat cylinder, the base of which extends perpendicularly to the vertical direction v. Each direction perpendicular to the vertical direction v is referred to herein as a radial or lateral direction, wherein Figure 1 The lateral direction is denoted by the reference symbol r.

[0161] The semiconductor body 2 is formed from a semiconductor material, for example silicon, and has p-type and n-type conductively doped semiconductor regions which determine the electrical properties of the thyristor in a manner known per se. The thyristor 1 and / or the semiconductor body 2 are designed rotationally symmetrically with respect to a center axis 3 extending in a vertical direction v.

[0162] In the semiconductor body 2 of the illustrated thyristor 1 according to the prior art, a p-type doped back base region 6, an n-type doped inner region 7 and a p-type doped front base region 8 are arranged successively in the vertical direction v starting from the back side 4 to the front side 5 opposite the back side 4.

[0163] Adjacent to the front side 5 of the semiconductor body 2 , an n-doped front side emitter 9 is embedded in the p-doped front side base region 8 , wherein the front side emitter 9 has a total lateral emitter width W given by its lateral extension. s1 Here, the lateral total structure width of the emitter 9 is to be understood as the distance between the inner side of the emitter structure 9 which is radially bounded furthest inwards and the outer side of the emitter structure 9 which is radially bounded furthest outwards. Figure 1 In the conventional thyristor 1 shown, the emitter 9 has a circumferentially closed annular structure concentrically surrounding the central axis 3 , wherein the semiconductor body 2 or the thyristor 1 is substantially configured as a cylinder that is rotationally symmetrical with respect to the central axis 3 .

[0164] In addition, from Figure 1As can be seen in FIG. 1 , the n-doped front emitter 9 is electrically conductively connected to a metallization layer 10 (e.g. aluminum), which is applied on the front side 5 and forms the front electrode 10 of the thyristor 1, wherein in the shown example of a conventional thyristor 1 the front electrode 10 can also be referred to as a cathode electrode, which has a cathode terminal K. The front electrode 10 has an overall lateral front electrode width W given by its lateral extension. E1 The p-doped back base region 6 is conductively connected to a metallization layer 11 (for example aluminum) applied on the back side 4, which metallization layer 11 forms a back electrode 11 of the thyristor 1, wherein in the shown example of a conventional thyristor 1, the back electrode 11 can also be called an anode electrode, which has an anode terminal A.

[0165] In addition, as in Figure 1 It can be seen that the n-doped front emitter 9 is penetrated in a generally known manner by a columnar short circuit 12 (also called cathode short circuit) with a doping complementary to the front emitter 9, which short circuit electrically connects the p-doped front base region 8 directly to the front electrode 10.

[0166] also, Figure 1 The thyristor 1 is shown to have a trigger structure in the form of a gate electrode 14 in a center 13 (r=0) of the semiconductor body 2, which is determined by the geometric center of gravity of the front side 5, by means of which an electrically conductive low-ohmic connection between the front electrode 10 (cathode here) and the back electrode 11 (anode here) can be established in a generally known manner, depending on an electrical trigger or conduction signal (gate current). In addition, the geometric center of gravity of the back side 4 of the semiconductor body 2 also determines a center 13, which coincides with the lateral position of the center 13 of the front side due to the symmetrical structure of the semiconductor body 2, and therefore, the two centers are given the same reference numeral 13. The two centers 13 determine the course of the central axis 3 (also called the axis of symmetry).

[0167] For example, when there is a positive potential between the back electrode 11 and the front electrode 10, the blocking gate direction (U D ) is triggered or started by gate current (positive current direction from gate 14 to front electrode 10). When turned on, the current plasma starts from the side of the n-type highly doped front emitter 9 facing the center 13, and moves radially or laterally outward in the entire front electrode area W at a speed of about 0.1 mm / μs. E1 Diffusion, according to the diameter of the thyristor 1, until the entire front electrode area W E1After about 1 to 4 ms, the current is conducted and a high load current flows between the front electrode (cathode) 10 and the back electrode (anode) 11. In the conducting state of the thyristor 1, the applied through-voltage then assumes its minimum value. Only when the current flowing through the thyristor 1 reaches zero again does the thyristor "extinguish" and switch to the blocking state (U between the back electrode 11 and the front electrode 10). R or negative potential). In between these two states, such as Figure 1 The intact conventional thyristor structure shown can be switched arbitrarily frequently.

[0168] The blocking voltage or cut-off voltage of the thyristor 1 is determined by the minimum thickness W of the n-doped inner region 7. n1 The minimum area of ​​the n-doped front emitter 9 determines the height of the permissible surge current and the conduction losses (conduction losses). n1 It cannot be chosen to be arbitrarily large, for example to increase the blocking capability, since this has a negative impact on surge current and conduction losses. n1 The conventional thyristor 1 is adapted accordingly to the respective application.

[0169] like Figure 1 As further shown, the lateral width W of the front emitter 9 S1 is substantially equal to the lateral width W of the front electrode 10 E1 Therefore, in the conventional thyristor 1, the lateral front emitter width W S1 The width of the horizontal front electrode W E1 The ratio is essentially equal to 1, or even slightly greater than 1.

[0170] Figure 2 FIG. 1 shows a vertical cross-section through an embodiment of a short-circuit semiconductor device 15 according to the present invention. The short-circuit semiconductor component 15 has a semiconductor body 16, wherein Figure 1 In a similar manner to the conventional thyristor 1 shown in , a p-doped back base region 6, an n-doped inner region 7 and a p-doped front base region 8 are arranged successively in the vertical direction v starting from the back side 4 to the front side 5 opposite the back side 4.

[0171] like Figure 2 As shown, the rear base region 6 is electrically conductively connected to a rear electrode 11 applied to the rear side 4, which rear electrode 11 has a total lateral rear electrode width W given by its lateral extension. ER , and the front base region 8 is electrically conductively connected to a front electrode 10 applied to the front side 5, which has a total lateral front electrode width W given by its lateral extension EV .

[0172] Adjacent to the front side 5 of the semiconductor body 16 , a front side conductive structure 17 is embedded in the p-doped front side base region 8 , which has a lateral front side conductive structure width W given by its lateral extension. SV .exist Figure 2 In the embodiment shown in FIG. 1 , the front conductive structure 17 is configured as an n-type doped front emitter structure 17 , which can be configured as a circumferentially closed annular and / or polygonal structure, for example. Figure 2 As can be seen in FIG. 1 , the emitter structure 17 is electrically conductively connected to a metallization layer 10 (eg aluminum) applied on the front side 5. As can be seen further, the front electrode 10 at least partially covers the front emitter structure 17. In particular, in FIG. Figure 2 , at least in terms of magnitude, the coverage of the conductive structure 17 by the front electrode 10 in the outer edge region, i.e. in the direction of the radial outer edge (r>0) of the short-circuiting semiconductor component 15, is at least in the millimeter range, for example at least approximately 1 mm to 5 mm or more. Figure 2 As shown in FIG. 1 , the emitter structure 17 protrudes slightly beyond the metallization layer 10 on the radial inner side thereof facing the center 13 . Figure 2 In the exemplary embodiment shown in , the n-doped front emitter structure 17 is likewise penetrated by a columnar short circuit 12 doped complementarily to the emitter structure 17 , which short circuit electrically connects the p-doped front base region 8 to the front electrode 10 .

[0173] The shape of the short circuit 12 can vary depending on the desired strength of the short circuit effect. Usually, a circular or polygonal, for example hexagonal, small dot-shaped recess is formed throughout the emitter structure 17 through the lateral front emitter structure width W. SV Within a defined continuous emitter structure surface, the short circuit 12 can have a size of the order of magnitude of approximately 100 μm. However, other forms of such short circuits 12 are also possible, such as strip-shaped breaks or ring-shaped breaks that divide the emitter structure 17 into individual concentric rings. The width of such breaks is typically in the range of approximately 100 μm.

[0174] and Figure 1 Compared with the conventional thyristor 1 shown in FIG. 1 , in the short-circuit semiconductor device 15 according to the present invention, the short-circuit effect of the short-circuit 12 can be greatly improved by targeted design. For example, the short-circuit opening of the short-circuit 12 is configured to be larger than that in FIG. 1 . Figure 1 , and / or the density of the short circuits 12 is selected to be significantly higher relative to the emitter structure surface of the emitter structure 17, i.e. the distance between adjacent short circuits 12 is selected to be significantly smaller than in the conventional thyristor 1. As a result, the short-circuit semiconductor device 15 according to the invention can be further reduced in terms of kA / cm 2A surge current is specified per unit area (in contrast to conventional thyristors 1 in which the surge current is known to be always maximized) in order to reliably ensure melting of the semiconductor body 16 as described here in the event of a short circuit.

[0175] The two pn junctions or np junctions provided in the semiconductor body 16 in the manner described above produce a short-circuit semiconductor component 15 that blocks at both voltage polarities.

[0176] Figure 2 The short-circuit semiconductor device 15 shown in FIG. Figure 1 The difference between the conventional thyristor 1 shown in FIG. 1 and FIG. 2 is the width W of the lateral conduction or emitter structure. SV and the lateral electrode width W EV which, in the case of a short-circuited semiconductor component 15 , is significantly smaller than 1 and can preferably be selected to be smaller than ½ or even smaller, but this is not mandatory.

[0177] In addition, from Figure 2 It can be seen that the front electrode 10 and the back electrode 11 are electrically conductively contacted, in particular pressure-contacted, by means of a metallic front contact pad 18 and a back contact pad 19, respectively. The contact pads 18 and 19 can be constructed as pure molybdenum pads or as a combination of molybdenum pads and copper pads, wherein the molybdenum is preferably in direct contact with the corresponding electrode 10 or 11 of the semiconductor body 16. The thickness of the contact pads 18 and 19 is preferably between 1 and 5 mm.

[0178] In addition, Figure 2 1 shows a front trigger structure 20 of the short-circuit semiconductor component 15, which is designed as a gate electrode 20 and contacts the front base region 8 in an electrically conductive manner and can be activated according to an electrical conduction signal supplied to the gate electrode 20 via its gate terminal G, and can thus switch on or pre-destroy the front conduction structure or emitter structure 17, which is operatively connected to the gate electrode 20 via the front base region 8. After switching on / pre-destroying the conduction structure or emitter structure 17, such a structure once and for all establishes a permanent, irreversible, electrically conductive, low-ohmic connection between the front electrode 10 and the back electrode 11. The gate electrode 20 is contacted during operation via an electrically conductive wire (not shown) which is pressed by means of a spring action.

[0179] As already mentioned above and Figure 21 , it can be seen that the emitter structure 17 protrudes slightly beyond the metallization layer 10 on the radial inner side of the metallization layer 10 facing the center 13. This is necessary in order not to short-circuit the emitter structure 17. In the event of such a short circuit, the current of the conduction signal delivered by the trigger structure or gate electrode 20 can flow directly from the trigger structure or gate electrode 20 to the electrode 10, which can, however, at least partially negatively affect the desired function of the conduction structure or emitter structure 17.

[0180] Furthermore, in operation, the short-circuit semiconductor device 15 is usually accommodated in a tightly sealed ceramic capsule (not shown) or plastic capsule configured as a housing. In order to convey an external electrical conduction signal or gate signal, a metal tube is installed in the ceramic / plastic capsule, which is connected to the metallized gate area 20 via a line for coupling the external conduction signal. In addition, the ceramic / plastic capsule has two copper contact dies, which are placed directly on the contact pads 18 or 19 of the front electrode 10 or the back electrode 11 on both sides of the short-circuit semiconductor device 15. The outward electrical connection of the short-circuit semiconductor device 15 in the ceramic / plastic capsule is realized by a corresponding mechanical pressure connector, which presses the two copper contact dies together with sufficient surface pressure, and for control by means of the conduction signal, the metal tube in the ceramic / plastic capsule is connected to the electrical conductor outside the capsule by suitable plugging or firm welding.

[0181] The main function of the short-circuit semiconductor device 15 according to the invention is that, in the case of a single pulse of a conduction or gate signal with a sufficient current level, that is, when the trigger structure is activated by means of the conduction signal, a predetermined energy input is injected into the semiconductor body between the trigger structure and the front conduction structure, and as a result, the first semiconductor junction formed between the corresponding conduction structure and the base region embedded in the conduction structure, the second semiconductor junction currently formed between the n-type emitter and the p-base region, and / or between the base region and the inner region is irreversibly destroyed, which in turn Figure 2 The semiconductor body 16 is then connected to the electrode terminal E by a flow of current. V and E R The fault current is destroyed, so that a fuse or melting 21 is formed around the region of the initial triggering of the semiconductor body 16, and the size or radial / lateral extension of the fuse or melting is also related to the terminal E V and E R After melting and cooling, a metal region is formed, which is composed of the molten silicon of the semiconductor body 16 and the metal of the contact pad 18 or 19, and provides two main terminals E V and E RThe height of the fuse current is determined by the maximum area to be provided for the n-doped front emitter structure 17 after the short-circuit current has been permanently absorbed, wherein for small fuse currents the smallest possible ratio W is sought. SV / W EV , and the lateral width W of the metal contact pad 18 or the front electrode 10 EV The height of the short-circuit current is determined such that the spatial extension of the melt region 21 does not reach the outer edge region of the short-circuit semiconductor component 15. In order to avoid edge failures caused by possible overvoltages, the thickness W of the n-doped inner region 7 of the short-circuit semiconductor component 15 is determined. n It can be selected almost arbitrarily, wherein the resistivity of the inner region 7 is additionally adapted accordingly to the desired level of the blocking capability.

[0182] In the case of interference, Figure 2 An electrical conduction signal is generated once in the short-circuit semiconductor device 15 shown in the figure, and the electrical conduction signal is output to the gate electrode 20 in the center 13 of the semiconductor body 16 through electrical transmission to the gate terminal G. As a result, the npnp structure 75 of the semiconductor body 16 is pre-destroyed, whereby a current flows between the front electrode 10 and the back electrode 11, which leads to a fuse 21 in the current-carrying area of ​​the semiconductor body 16. Thereafter, the short-circuit semiconductor device 15 is in a permanent, irreversible, low-ohmic current-carrying state.

[0183] The conductivity of the n-doped front emitter structure 17 is relatively Figure 1 The n-doped cathode-side emitter 9 of the conventional thyristor 1 shown in FIG. 1 is very limited because its lateral emitter structure width W is limited by the n-doped front emitter structure 17 . SV The total width W of the area contacted by the metallization or front electrode 10 and the contact pad 18 EV ratio to determine.

[0184] The n-doped front emitter structure 17 holds the molten region 21 near the initial ignition at a location near the center 13 of the semiconductor body 16 and prevents the plasma from leaving the region of the semiconductor body 16 contacted by the front electrode 10 and the contact disk 18 to reach the radially outer edge regions. Figure 2 The structure of the embodiment of the short-circuit semiconductor device 15 shown in FIG. 1 ensures that the melt 21 always remains within the area covered by the electrode 10 and the contact pad 18, that is, within the electrode width W EV Within.

[0185] Figure 3FIG. 2 shows a top view of another embodiment of a short-circuit semiconductor device 22 according to the present invention having a finger-shaped conductive structure 23. Figure 3 2, a front electrode 24 can be seen, which is applied to the front side of the semiconductor body 25 in the structure according to the invention. In the center 13 of the short-circuit semiconductor component 22, a gate electrode 26 designed as a trigger structure is applied to the semiconductor body 25. Figure 3 As can be seen in FIG. 1 , the front electrode 24 is radially spaced apart. Figure 3 As can be seen, in the short-circuit semiconductor device 22 shown, the gate electrode 26 has four trigger structures or gate fingers 27 extending radially outward from the remaining part of the gate electrode 26, that is, from the central area. The front electrode 24 has a recess corresponding to the gate finger 27 to allow the gate finger 27 to extend radially outward and be spaced apart from the electrode 24. Corresponding to the recess of the electrode 24, the conductive structure 23 embedded in the semiconductor body 25, which is located below the electrode 24 and at least partially covered by the electrode and is electrically contacted by the electrode 24, also has a course that follows the contour of the electrode recess. The conductive structure is an emitter structure here.

[0186] The emitter structure 23 and its orientation are Figure 3 The width of the electrode recess is represented by the contour line drawn in FIG. SV Intuitively, it basically corresponds to Figure 3 The width of the contour line drawn in FIG. 2 is shown in cross section. The emitter structure 23 can be similar to Figure 2 The emitter structure 17 in FIG. 1 can also be constructed in the manner of an emitter structure 17 and can also have a short circuit 12 . However, this is not absolutely necessary.

[0187] exist Figure 3 As can be seen in FIG. 1 , the emitter structure 23 has four conductive structure fingers 28 extending radially outward from the remaining portion of the emitter structure 23 near the center, that is, from the region near the center of the emitter structure 23 in the lateral direction r. Figure 3 In the exemplary embodiment of the short-circuit semiconductor component 22 shown, the conduction or emitter structure 23 is designed to follow the contour of the electrode recess in a linear manner. Such a linear conduction structure 23 can preferably have a lateral structure width W of approximately 500 μm. SV (See Figure 2 ). Therefore, in Figure 3 The conduction or emitter structure 23 shown in FIG. 2 is partially constructed as a circumferentially closed annular structure and partially as a circumferentially closed polygonal structure (emitter finger 28 ), which extends radially outward from the annular structure.

[0188] It should be understood that the fingers of the conduction structure 23 and the corresponding fingers of the trigger structure 26, here the gate electrode 26, can of course be implemented in many other conceivable ways. Therefore, the number of the conduction structure fingers 28 or the trigger structure fingers 27 is not limited to Figure 3 There may be more or fewer fingers. The orientation of the fingers 28 or 27 is also not limited to Figure 3 The straight line direction of the embodiment shown in . For example, this direction can also be curved. In addition, it is conceivable that starting from each finger 28 or 27, one or more branched finger sections can also be connected from each in Figure 3 The fingers 28 or 27 shown in the figure extend outwards and thus form one or more branched fingers.

[0189] from Figure 3 As can be seen from the diagram of FIG, the trigger structure 26 has a trigger structure finger 27 that corresponds to (i.e. follows) the shape or contour of the conductive structure finger 28. Since the spacing between the trigger structure 26 and the conductive structure 23 is uniform over the entire course of the finger-shaped conductive structure 23, especially in the lateral direction, this embodiment ensures a uniform electrical effect of the trigger structure 26 on the conductive structure 23.

[0190] The finger-shaped conductive structure 23 provides a significant advantage, that is, the conductive or triggered area of ​​the short-circuited semiconductor device 22 can be distributed over a larger range. Preferably, the conductive structure finger 28 is extended laterally from the remaining part of the conductive structure 23 near the center, that is, the radial finger length L of the conductive structure SF Select less than W EV / 2. In the exemplary embodiment of the short-circuit semiconductor device 22 shown here, this also applies to the radial finger length of the trigger structure finger 27. This measure helps to make the fuse area 21 (see Figure 2 ) is far from the center 13 of the semiconductor body 22 and thus from the gate electrode 26, and on the other hand is also far enough away from the radial outer edge of the short-circuiting semiconductor component 22. Figure 3 In the embodiment shown, the central region 13 is not contacted or covered by the electrode 24 and therefore is not covered by the thick contact plate 18 ( Figure 2 ) contact or cover, the plasma can be directed in the center 13 to the electrical leads to the gate terminal G and to the metal tube in the ceramic / plastic capsule (not shown) and destroy the capsule there. This is done by Figure 3 The embodiment of the short-circuit semiconductor device 22 shown in FIG. 1 is safely prevented because the melting 21 ( Figure 2 ) always triggers in the event of a short circuit within the region of emitter structure 23 covered by electrode 24 .

[0191] Figure 4FIG. 2 shows a vertical cross-sectional view of another embodiment of a short-circuit semiconductor device 29 according to the present invention having two conduction structures 30 and 31 and a trigger structure 20 (gate electrode). Figure 4 It can be seen that the short-circuit semiconductor device 29 has an n-type doped front-side conduction or emitter structure 30 in the semiconductor body 32, and its lateral emitter structure width W SV The width of the horizontal front electrode W EV Compared to being selected as still Figure 2 The short-circuit semiconductor device 15 in is much smaller. Figure 4 As can be seen in FIG. 1 , the emitter structure 30 in this embodiment does not have the Figure 2 The short circuit 12 shown in FIG. 1 is not necessarily required, however. The conduction or emitter structure 30 can be designed as a circumferentially closed annular and / or polygonal structure, which surrounds the trigger structure 20 at a certain radial distance. The emitter structure 30 also protrudes slightly from the inner side of the electrode 10 radially toward the center 13, as already described in conjunction with the Figure 2 As described in the emitter structure 17 in FIG.

[0192] In addition to the trigger structure 20 formed by the gate 20 and arranged in the center 13 of the semiconductor body 32, the semiconductor body 32 also has a second conductive structure 31 in the form of a breakdown structure 31, which is configured as a circumferentially closed annular BOD structure in this embodiment. By means of the breakdown structure 31, the short-circuit semiconductor component 29 can be connected to the electrode terminal E V and E R The operating voltage on the electrode terminal E is also turned on. In the embodiment shown, a negative polarity is applied to the back electrode 11 relative to the front electrode 10. In addition, it is also used to apply an unforeseen voltage to the electrode terminal E. V and E R In the event of an overvoltage above 0.1 V, blocking failures in the radially outer edge region of the short-circuited semiconductor component 29 are avoided.

[0193] As in Figure 4 As can be seen in FIG. 1 , the breakdown structure 31 is located on the side of the semiconductor body 32 opposite the front emitter structure 30 and thus represents the back-side conductive structure 31 of the short-circuited semiconductor component 29. The level of the breakdown voltage of the breakdown structure 31 is set in this case so that the breakdown structure assumes its lowest value for the entire semiconductor body 32 at the exact position where it is arranged and also prevents breakdown in the outer edge region of the semiconductor body 32. The radius of curvature of the pn junction, i.e. its geometric curvature, is Figure 4 In the region denoted by reference numeral 33, the resistivity of the breakdown structure 31 and the diameter D of the breakdown structure 31 are proportional to each other. i (Inner diameter) and D a (outer diameter) or lateral back-side conductive structure width WSR The ratio of θ and θ determines the level of the breakdown voltage in a manner known per se, as described, for example, in DE 42 15 378 C1. R Overvoltage occurs (back electrode terminal E R In the case of the negative electrode at the main terminal E in region 33 V and E R The blocking current flowing between the two main terminals E increases sharply from a certain voltage. The current density reaches such a high value that at the end V and E R A local melting channel is formed between the npnp structure. The blocking capability of the npnp structure is destroyed and the short-circuit current causes melting in the region of the BOD structure 31. With a breakdown structure according to the type of breakdown structure 31, generally speaking, higher breakdown voltages or very high blocking voltages reaching the kV range can be very well controlled technically.

[0194] exist Figure 4 In the embodiment of the short-circuit semiconductor device 29 shown, since the breakdown structure 31 is a back-side conductive structure embedded in the back-side base region 6, wherein the back-side electrode 11 is applied to the back-side base region 6, the lateral back-side conductive structure width W of the breakdown structure 31 is SR The lateral back electrode width W of the back electrode 11 ER The ratio also satisfies the condition W according to the present invention. SR / W ER <1, this is Figure 4 This is evident in the short-circuited semiconductor device 29 shown.

[0195] like Figure 4 It can be seen that the width W of the lateral conductive structure of the breakdown structure 31 is SR fixed to the lateral region of the base region 6 in which the geometry of the pn junction formed between the base region 6 and the inner region 7 is fixed relative to the geometry of the remaining pn junctions, i.e. relative to the lateral back conductive structure width W SR The geometry of the pn junction outside of a certain region is significantly changed. It should be understood here that the pn junction has a changed geometry within the meaning of the invention only if the change is not attributable to manufacturing-related tolerance deviations when manufacturing the base region 6 and the inner region 7, but is a targeted result.

[0196] Figure 4 The breakdown structure 31 of the short-circuit semiconductor component 29 shown has a ring shape surrounding the center 13 of the semiconductor body 32 and is accordingly radially spaced apart from the center 13. Figure 4In the embodiment of the short-circuit semiconductor device 29 shown in FIG, the breakdown structure or back-side conductive structure 31 is completely covered by the front electrode 10 and the back electrode 11. The complete front and back side coverage of the breakdown structure 31 offers the advantage that the plasma is safely prevented from escaping from the semiconductor body 29 by local confinement of the melting in the area around the breakdown structure 31 covered on both sides by the electrodes 10 and 11.

[0197] However, the breakdown structure 31 acting as a conductive structure can also be arranged in the center 13 of the semiconductor body 32 in principle, as is also described in DE 42 15 378 C1, for example. Figure 4 In this case, which is not shown in FIG. 1 , the breakdown structure 31 is preferably also covered by corresponding electrodes 10 and 11 both on the front side 5 and on the back side 4 of the semiconductor body 32 in order to achieve the same effects as described above with regard to the plasma escape. In other words, a short-circuiting semiconductor component having at least one centrally arranged conduction structure, in particular a breakdown structure designed as a conduction structure, does not require a trigger structure arranged there, for example in Figure 4 The gate electrode 20 with the gate terminal G is shown in FIG. Since the breakdown structure 31 is applied to the electrode terminal E V and E R The turn-on signal is obtained by the voltage of the control circuit, so the supply of an external additional turn-on signal can be abandoned.

[0198] It should be noted here that Figure 4 The breakdown structure 31 shown as a BOD structure in the figure can also be replaced by any other known breakdown structure with a similar effect, as long as the breakdown structure is constructed as a conduction structure in order to initiate the melting of the corresponding short-circuited semiconductor device depending on the conduction signal, which in the breakdown structure integrated in the semiconductor body is always implicitly derived from the voltage applied to the external electrodes 10 and 11. Such other breakdown structures are, for example, breakdown structures with du / dt protection known per se or breakdown structures with circuit commutation turn-off time protection known per se with local increase of the carrier lifetime. Fig.10 and 11 The description of further describes the breakdown structure of another optional breakdown structure 31.

[0199] Figure 5 A vertical cross-sectional view of another embodiment of a short-circuit semiconductor device 34 according to the present invention is shown, which has two conducting structures 35, 36 and two trigger structures 37, 38. The two conducting structures 35 and 36 are respectively configured as n-doped emitter structures, which are respectively embedded in the front base region 8 or the back base region 6 of the semiconductor body 39. The conducting or emitter structures 35, 36 can be configured as annular and / or polygonal structures that are closed in the circumferential direction. Figure 5It can be clearly seen that the two emitter structures 35 and 36 each slightly protrude from the radial inner edge of the respective electrode 10 or 10 towards the respective center 13. The two trigger structures 37 and 38 are each designed with a front or rear gate terminal G V and G R The gate electrodes 37 and 38 are provided, wherein the front gate electrode 37 is arranged in the center 13 of the front base region 8 and is in electrical contact with the base region 8, and the gate electrode 38 is arranged in the center 13 of the back base region 6. Figure 5 In the embodiment of the short-circuit semiconductor device 34 shown in FIG, the two conductive structures 35 and 36 respectively have the same lateral conductive structure width W. SV or W SR , however, this is not mandatory. The lateral electrode width W of the front electrode 10 and the back electrode 11 EV and W ER The short-circuit semiconductor components 34 also have the same size, which is also not absolutely necessary. Therefore, the short-circuit semiconductor components 34 are constructed to be completely symmetrical with respect to the vertical direction. EV and W ER can be significantly larger than the corresponding lateral conductive structure width W SV or W SR , so that the condition W also applies to the short-circuit semiconductor device 34 SV / W EV <1 and W SR / W ER <1, however, this is not mandatory. Therefore, the short-circuited semiconductor device 34 can be connected not only to the front gate terminal G V The electrical conduction signal is transmitted to the back gate terminal G R The electrical conduction signal is turned on.

[0200] Figure 6 FIG. 2 shows a vertical cross-sectional view through one embodiment of a short-circuit semiconductor device 76 according to the present invention. The semiconductor device 76 corresponds substantially to Figure 2 , wherein the semiconductor component 76 additionally has an insulating layer 78 or a contact surface of the conduction / emitter structure 17 between the front electrode 10 and the front side 5 of the semiconductor body 77. It can be seen that the insulating layer 78 extends on the conduction structure 17 in the radial direction r and extends in the circumferential direction u along the inner edge of the electrode 10 located in the radial direction r. Here, the insulating layer 78 is a silicon dioxide layer, which is produced in a thermal process when producing the semiconductor body. It preferably has a thickness of at least 50 nm to a maximum of 100 nm.

[0201] Figure 7FIG. 4 shows a vertical cross-sectional view of another embodiment of a short-circuit semiconductor device 45 according to the present invention having two conductive structures 30 and 46 and trigger structures 20 and 47. In a semiconductor body 48, a back conductive structure 46 is provided as a short circuit 12 and a lateral back conductive structure width W. SR The n-type doped emitter structure 46 is embedded in the back base region 6 adjacent to the back side 4 and is conductively contacted and completely covered by the back electrode 11. The conduction or emitter structure 46 can be constructed as a ring and / or polygonal structure that is closed in the circumferential direction. In the same lateral area of ​​the conduction structure 46, but offset in the vertical direction v, a back trigger structure 47 constructed as a breakdown structure, for example a BOD structure, is arranged. As already mentioned in conjunction with Figure 4 As described, when the breakdown voltage determined for the breakdown structure 47 is reached or exceeded, which breakdown voltage directly depends on the voltage applied to the external electrodes 10 and 11 as a conduction signal, a locally limited sharp current increase is adopted in the region 33, which current increase is in the short-circuited semiconductor device 45. Figure 7 The example shown in FIG. 4 is used to turn on the back emitter structure 46 or in Figure 2 or Figure 6 The pre-destruction of the pre-destruction region 75 shown in FIG. 4 finally initiates the melting of the short-circuited semiconductor device 45. The emitter structure 46 is set as a conductive structure, which is turned on or pre-destructed by the breakdown structure 47. The conductive structure is relative to the Figure 4 The conductive structure 31 shown in the figure as a breakdown structure, in which the melting channel for melting starting from the breakdown structure 31 is locally narrowly defined, has the advantage of lateral expansion of the melting region. Due to the lateral expansion of the melting channel, the emitter structure 46 helps the melting to proceed "softer". In addition, the larger cross-sectional melting region in the semiconductor body 48 allows higher short-circuit currents to be conducted permanently with lower conduction losses due to the reduced conduction resistance.

[0202] Figure 8 The view (a) shows Figure 2 A first circuit diagram of an embodiment of a circuit of a short-circuit semiconductor device 15 is shown in another view (b). Figure 2 A circuit diagram of another embodiment of a circuit of a short-circuit semiconductor device 15 is shown.

[0203] exist Figure 8 In (a), it can be seen that there is a predetermined capacitor C corresponding to the predetermined energy input, which provides a conduction signal in the charged state, wherein the capacitor C can be connected to the gate 20 or the gate terminal G and the front electrode 10 or the front electrode terminal E by means of an electronic switch 79. VThe short-circuit semiconductor switch 15 is used as a unidirectional switch. The semiconductor switch 15 must conduct current in only one direction, and the other current direction can be ensured, for example, by a pressure-contact diode (not shown) arranged in anti-parallel with the semiconductor switch 15.

[0204] exist Figure 8 (b) shows double Figure 8 (a) circuit structure, wherein the circuit structure is anti-parallel. In this way, an overall bidirectional structure can be realized by means of two semiconductor switches 15, which are unidirectional in themselves, wherein preferably a gate control unit (corresponding to capacitor C and electronic switch) is respectively assigned to each semiconductor switch 15. The bidirectional implementation or anti-parallel connection of the two semiconductor switches 15 can therefore conduct current in two directions.

[0205] As in Figure 8 Instead of the bidirectional, anti-parallel connection of two unidirectional semiconductor switches 15 shown in (a), it is also possible to use semiconductor switches that operate bidirectionally directly, for example Figure 5 Semiconductor switch 34. In any case, it is preferred to accommodate two unidirectional semiconductor components or one bidirectional semiconductor component in one housing.

[0206] exist Figure 8 Also shown in FIG. 8 is an impedance 80, which represents a parasitic resistance or inductance in the gate circuit. The lower the impedance, the higher the surge current or energy input available for pre-destruction.

[0207] Fig. 9 FIG. 4 shows a vertical cross-sectional view of another embodiment of a short-circuit semiconductor device 52 according to the present invention having two conduction structures 46 and 53 and trigger structures 47 and 54. Figure 7 The illustrated conduction structure 46 constructed as an emitter structure and the back trigger structure 47 constructed as a breakdown structure electrically connected thereto are embedded in the back side of the semiconductor body 55. On the front side, in a similar manner, a front conduction structure 53 constructed as an n-doped emitter structure is embedded in the front base region 8 adjacent to the front side 5 and is electrically contacted with the front electrode 10. The front emitter structure 53 is assigned a front trigger structure 54, which is also constructed as a breakdown structure and which turns on or pre-destroys the emitter structure 53 during breakdown. The mode of action of the combination of the front emitter structure 53 and the front breakdown structure 54 corresponds to the combination of the back emitter structure 46 and the back breakdown structure 47, and has been combined Figure 7 The conduction or emitter structures 46, 53 and the trigger structures 47, 54 can be respectively constructed as circumferentially closed annular and / or polygonal structures.

[0208] The conduction and triggering structures 46 , 53 , 46 , 54 of the short-circuit semiconductor component 52 can be arranged exactly in the center, that is to say in the corresponding center 13 , but are not limited thereto.

[0209] Fig.10 A vertical cross-sectional view of another embodiment of a short-circuit semiconductor device 56 according to the present invention having two conducting structures 30 and 57 and a trigger structure 20 is shown. Figure 4 In addition to the known front-side conductive structure 30, a back-side conductive structure 57 is embedded in the semiconductor body 58 in another structural form of a breakdown structure. Fig.10 It can be clearly seen in FIG. 5 that the back-side conductive structure 57 is locally limited, especially in the radial direction r, and thus has a lateral structure width W SR , the lateral structure width W SR Significantly smaller than the lateral electrode width W of the assigned back electrode 11 ER This is the case for all lateral conductive structure widths disclosed herein, relative to the corresponding lateral electrode widths of the individual electrodes of the short-circuited semiconductor device according to the present invention.

[0210] from Fig.10 It can also be seen that the back base region 6 has a flat p-type substrate facing the back side 4. ++ region, which is highly p-doped compared to the region towards the inner region 7. SR In the cross section, the n-type doping of the n-type doped inner region 7 is substantially higher than that of the n-type doped inner region 7. ++ The region is embedded in the generally p-doped region of the back base region 6 and reaches the inner region 7 in the vertical direction v. Due to the highly doped p-type of the base region 6 ++ The region is interposed with the back surface 4 contacted by the electrode 11. ++ Area spacing, n ++ The region does not reach the back side 4 of the semiconductor body 58 to which the back side electrode 11 is applied. ++ The concentration of the region is selected to be higher than the n-p region interrupting the base region 6. ++ region to prevent a short circuit between the inner region 7 and the back electrode 11.

[0211] for Fig.10 The breakdown structure 57 shown in FIG. 1 can also be adjusted in a targeted manner, for example by selecting a weakly n-doped inner region 7 and / or a highly doped n-type inner region 7. ++ The doping concentration of the region 57 and / or the p-doped base region 6, in particular its highly doped p ++The doping concentration of the region 57 separating the breakdown structure 57 from the back side 4 of the semiconductor body 58 is selected so that the base region 6 has a doping concentration of 0. ++ Area and / or p ++ The higher the concentration of the region, the lower the breakdown voltage of the breakdown structure 57. In this way, a breakdown voltage can be determined which is in a range well below 10 V up to the maximum possible blocking capability of the semiconductor body 58. Breakdown structures of the type of breakdown structure 57 are technically very advantageous in setting relatively small blocking voltages in the range of several tens of volts up to approximately 100 V.

[0212] Fig.11 In views (a), (b) and (c), the concentration variation lines Na, Nb or Nc shown there are shown through Fig.10 The concentration of the short-circuit semiconductor devices 56 is shown to vary. Fig.11 a shows along Fig.10 The change line shows the change in Na concentration. Fig.11 b shows the concentration variation along the variation line Nb, and Fig.11 c shows the concentration change along the change line Nc. Fig.11 In the concentration variation of a, 11b and 11c, the penetration depth x of the corresponding dopant in μm is plotted on the corresponding abscissa. The ordinate represents the penetration depth x in cm -3 The concentration N of each dopant.

[0213] Fig.11 a shows the Fig.10 The variation line Na shown in FIG. 1 is a concentration variation, which extends from the front side 5 of the semiconductor body 58 through the front side conduction or emitter structure 30 and the front side base region 8 into the inner region 7. Fig.11 Starting from x=0, a flat profile can be seen, that is to say a profile with a small penetration depth (here less than approximately 20 μm), which has a high concentration of phosphorus-doped n in the form of a dotted curve, for example. ++ Emitter structure 30 (up to about 1E20 cm -3 ). With increasing penetration depth x (eg from 20 μm to approximately 40 μm), the concentration profile follows, for example in the form of a continuous curve, with the p-type base region 8 doped with boron. Fig.10 In the short-circuit semiconductor device 56 shown in FIG. 1 , the first p-type region doped with boron in the base region 8 is preceded by a first p-type region doped with aluminum (Al) in the base region 8 having a significantly lower concentration (eg, about 1E13 to about 1E15 cm -3 A second p-type region between Fig.11 The concentration profile of a can be seen by means of the dashed curve. However, this Al profile in the base region 8 is not absolutely necessary. Fig.11 In the concentration profile of a, as the penetration depth x increases further (in this case approximately more than 100 μm), the base region 8 is followed by a region with a substantially constant concentration (in this case approximately 1E13 cm -3 )'s weakly n-doped inner region 7.

[0214] Fig.11 b shows the Fig.10 The variation line Nb shown in FIG. 1 is a concentration variation, which extends from the back side 4 of the semiconductor body 58 through the back side base region 6 and the back side conductive or breakdown structure 57 into the inner region 7. Fig.11 In b, starting from x=0, we can first see that, for example, the p-type base region 6 doped with boron ++ Very flat profile variation of the area, which has preferably about 5E18cm -3 and preferably a penetration depth of about 5 to 20 μm in the form of a solid curve. With increasing penetration depth x, the next is, for example, n-doped with phosphorus or arsenic. ++ The concentration of the conductive structure or breakdown structure 57 changes. Fig.11 It is shown as a dotted curve in b. Fig.11 It can be clearly seen in b that the highly doped n ++ The concentration N of the breakdown structure 57 is significantly lower than that of the upstream front surface p of the base region 6. ++ As the penetration depth x continues to increase, the concentration profile of the breakdown structure 57 transitions into a profile of the weakly n-doped inner region 7 .

[0215] Fig.11 c shows the Fig.10 The variation line Nc shown in FIG. 1 is a concentration variation, which extends from the back side 4 of the semiconductor body 58 through the back side base region 6 into the inner region 7. Fig.11 In c, starting from x = 0, we can first see that the p-type base region 6 doped with boron ++ Very flat profile variation of the area, which has preferably about 5E18cm -3 and preferably a penetration depth of about 5 to 20 μm in the form of a solid curve. Then, as the penetration depth x increases, a region with a significantly lower concentration N (here less than 1E16 cm -3 ) of the remaining p-type doped base region 6 changes in concentration, the concentration change is Fig.11 c is shown as a dotted curve. The back base region 6 and the front base region 8 (see Fig.11 a) also has a direction in front of the inner region 7, for example, a p-type region of the base region 6 with a weak doping of aluminum Al, which has a density of about 1E15 cm -3 concentration up to a penetration depth x of about 100 μm, as in Fig.11 c is shown by means of a dashed curve. However, such an Al-doped p-type region of the base region 6 is not absolutely necessary. Fig.11 As shown in b, the weakly n-doped inner region 7 is reconnected to this region of the base region 6.

[0216] It is particularly preferred that the front and / or back base regions 8 or 6 are arranged Fig.10 In the region of the concentration curves Na and Nc given in FIG. 1 , it has a penetration depth or thickness of at least 50 μm, for example about 100 μm, as can also be seen in FIG. Fig.11 a and 11c.

[0217] As described above Fig.10 As already mentioned, by the additional introduction of the formation Fig.10 The n shown in ++ The amount of donor atoms in the breakdown structure 57 determines the adjustable breakdown voltage of the breakdown structure 57 in the blocking direction. The greater the concentration, the lower the breakdown voltage. ++ The area is transitioning to n ++ The concentration of the breakdown structure 57 and the p ++ The contour transitions to n ++ The concentration change when the structure 57 is broken down (see Fig.11 b) Determine the breakdown voltage: At the p- ++ Region to breakdown structure 57 n ++ The flatter the gradient of the concentration change at the transition of the regions, the greater the blocking capability at the same ratio. Thus, for example, in the case of a flat, highly doped p-type region diffused into the base region 6, ++ After the area, you can use p ++ The driving time of the region adjusts the blocking capability of the breakdown structure 57 because as the driving time increases, the p ++ Region to breakdown structure 57 n ++ The concentration gradient at the transition of the regions becomes flatter, which increases the breakdown voltage. For a very low breakdown voltage of the breakdown structure 57, a very flat base region 6 (small penetration depth x) and a very high doping of the p-type region 6 must therefore be established. ++ For example, in this way, for an assumed breakdown voltage of approximately 30 V, the n of the breakdown structure 57 is obtained. ++ Area of ​​about 5E16cm -3 The average concentration of

[0218] The blocking capability of the breakdown structure 57 is adjustable over a wide range. However, when the concentration becomes very small and approaches the concentration of the weakly n-doped inner region 7, it reaches its boundary, which can have a value of less than 1E14 cm for a high blocking semiconductor device, for example. -3 If, on the other hand, the blocking capability of the breakdown structure 57 differs significantly from the maximum possible blocking capability of the short-circuit semiconductor device 56 (which is limited by the silicon thickness W n and the concentration N of the weakly doped inner region 7), the breakdown voltage of the breakdown structure 57 can be adjusted very well if the maximum possible blocking capability is, for example, less than 100 V, wherein the required blocking capability of the short-circuit semiconductor device 56 is then significantly higher than 1000 V, when the ratio of the maximum possible blocking capability of the short-circuit semiconductor device 56 to the adjusted blocking capability of the breakdown structure 57 is in the range of approximately 50 to approximately 100.

[0219] Fig.10 The breakdown structure 57 described in the present invention can be used as a conduction structure and / or trigger structure in the short-circuit semiconductor device of the present invention in the same manner as other breakdown structures described herein. The conduction or breakdown structure 57 of the short-circuit semiconductor device 56 can be constructed as a circumferentially closed annular and / or polygonal structure.

[0220] Fig.12 FIG. 4 shows a vertical cross-sectional view of another embodiment of a short-circuit semiconductor device 59 according to the present invention having two conduction structures 30 and 57 and trigger structures 20 and 60. In the semiconductor body 61, in addition to Fig.10 In addition to the structure of the short-circuit semiconductor device 56 shown in FIG. 1 , the breakdown structure configured as a trigger structure 60 on the front side is arranged below the front trigger structure 20 configured as a gate electrode 20 in the vertical direction v. The trigger structure 60 has the same structure as the back side conductive structure 57, as shown by Fig.12 In order to embed the breakdown structure 60 into the front base region 8 , the breakdown structure is divided into a highly doped flat p-type region facing the front side 5 like the back base region 6 . ++ region and the usual p region towards the inner region 7, so that p ++ The region is disposed between the gate electrode 20 and the breakdown structure 60 in the vertical direction v. ++ A front emitter structure 30 is embedded in the front base region 8 adjacent to the front side 5 and is in electrically conductive contact with the front side electrode 10 .

[0221] exist Fig.12In the case of the short-circuited semiconductor device 59 shown in FIG. 1 , if the blocking voltage applied thereto and determined by applying to the main electrodes 10 and 11 exceeds the breakdown voltage determined for the breakdown structure 60, the front side conduction structure 30 can be turned on not only by the electrical conduction signal supplied from the outside to the gate terminal G, but also additionally by the trigger structure 60. In this case, the breakdown current increases at the breakdown structure 60 until the breakdown structure 60 is turned on or the conduction structure 30 embedded in the same base region 8 is destroyed, which ultimately leads to the desired melting of the short-circuited semiconductor device 59.

[0222] As in Figure 2 and Figure 6 As schematically illustrated in FIG. 7 , in the pre-destruction region 75 between two semiconductor junctions, Fig.12 In the short-circuited semiconductor device 59, it can also be understood that the pre-destruction of the conductive structure 30 caused by the trigger structure 20 and / or 60 always occurs spatially separated from the corresponding trigger structure 20 or 60 (further outward in the radial direction). This ensures that melting always occurs in the area of ​​the semiconductor body 61 that is covered (especially pressed) by the corresponding electrode (here 10). In order to ensure this, the trigger current generated by the corresponding trigger structure 20 and / or 60 can be additionally limited if necessary, for example by providing a corresponding resistance zone, in order to safely prevent pre-destruction caused by excessively high trigger currents in the gate region / gate electrode 20 or the center 13.

[0223] exist Fig.12 In the case of a combination in which the two trigger structures 20 and 60 shown in FIG. 1 are arranged on or in the same base region (here, the front base region 8 ), it should be understood that the trigger or conduction sensitivity of the conduction structure (here, the emitter structure 30 ) associated with the two trigger structures 20 and 60 and connected downstream is matched to the conduction current that can be provided by the trigger structure 20 or 60 , wherein the minimum conduction current that can be provided by the trigger structure is decisive. In other words, the conduction structure 30 is Fig.12In the example shown in FIG. 1 , a minimum forward current is designed, starting from which the conduction structure 30 is safely conducted, because a trigger structure is present, which is designed as a breakdown structure 60, and which typically only provides a forward current in the mA range, which is in a range significantly below 1A, preferably in a range of approximately 50 mA, so that the blocking current generated by the breakdown structure 60 in the event of a breakdown can safely conduct or pre-destroy the conduction structure 30 without being destroyed itself. This can occur if the triggering or conduction sensitivity of the conduction structure 30 is selected too high. Thus, although the emitter structure 30 can be triggered by an external conduction signal in the region of approximately 1A, which is supplied via the gate electrode 20, the breakdown structure 60 arranged below the gate electrode 20 will be destroyed in this case, because it will be overloaded in the event of a forward current of 1A injected via the gate electrode 20. Since the trigger structure 60 is not located in the region of the semiconductor body 61 covered by the electrode 10 and the contact pad 18, such a destruction of the trigger structure 60 may lead to a plasma escape in the region of the uncovered center 13, which is however prevented by Fig.12 The short-circuiting of the semiconductor embodiment shown is safely prevented by the above-described related design of the conductive structure 30 and its conductive sensitivity.

[0224] If, for example, in terms of process technology, the short-circuit semiconductor component 59 is to reach a certain blocking voltage level (especially for example for a breakdown voltage in the kV range) to form a breakdown structure, it should only be produced with particularly high technical effort or unsafely, it can be replaced by Fig.12 The breakdown structure 60 of the semiconductor device 59 shown in FIG. 5 may also be used as in Figure 4 . As already explained there, this breakdown structure 31 offers particular advantages for setting also very high blocking voltages / breakdown voltages up to the kV range.

[0225] Fig.13 A partial top view of another embodiment of a short-circuit semiconductor device 62 according to the present invention is shown, which has a conducting structure 63 open in the circumferential direction u of the semiconductor body and having at least one conducting structure segment 64 (here a ring segment) and a trigger structure 20. Fig.13 In the top view of FIG. , only the front base region 8 can be seen from the semiconductor body. Fig.13 In the illustration of FIG. 1 , the front electrode 10 is not shown, but only its radial inner edge 65 is shown, so that the lead-through structure 63 arranged below the electrode 10 and designed as an emitter structure can be recognized. Fig.13 As can be seen from FIG. 6 , in this embodiment, the emitter structure 63 or the emitter structure segment 64 has a radial distance from the trigger structure or the gate electrode 20. As can be further seen, in this case, the conductive structure 63 is not closed in the circumferential direction u of the semiconductor body, that is, in Fig.13 The conductive structure 63 of the short-circuited semiconductor device 62 shown in FIG. 6 does not have a closed ring-shaped and / or polygonal structure. The conductive structure segment 64 has a lateral emitter structure width W SV and the circumferential emitter structure width W extending in the circumferential direction u UV Circumferential emitter structure width W UV Smaller than a circumferential lead-through structure width which corresponds to a closed annular and / or polygonal structure of the lead-through structure and thus to a circumferential angle of 360 degrees of the closed annular and / or polygonal lead-through structure.

[0226] exist Fig.13 In the embodiment of the short-circuit semiconductor device 62 shown in FIG. 1 , the conduction current for conducting or pre-destroying the conduction structure 63 is determined to be about 2 A, preferably in a range between about 1 A and about 5 A. SV The conductive structure 63 can now be selected to be large enough so that it does not cause problems in terms of process technology, i.e., it can be formed with a sufficient width W. SV If W SV If the size is chosen too small, there is the problem that the radially inner electrode edge 65 no longer covers the highly n-doped emitter structure 64 reliably and sufficiently accurately, wherein the circumferential structure width W UV It is not so important here. Fig.13 The embodiment shown in FIG. 1 can be realized in a particularly advantageous manner due to the relatively small size W compared to closed annular and / or polygonal structures. SV and W UV The resulting relatively small fusing current, wherein the conduction current required for conducting / pre-destroying the conduction structure 63 is more likely to assume a large value in the range of preferably about 1 A to about 5 A, has the advantages already described herein.

[0227] It is important to understand that, as in Fig.13 The short-circuit semiconductor component 62 shown in the figure may also have more than one conducting structure 63, which are then distributedly arranged in the circumferential direction u of the semiconductor body, preferably arranged at equidistant intervals from each other, so as to achieve a substantially symmetrical structure in the circumferential direction of such a short-circuit semiconductor component with the advantages already described here.

[0228] Fig.14 A partial top view of another embodiment of a short-circuit semiconductor device 66 according to the present invention is shown, which has the Fig.13 The conductive structure 63 and the trigger structure 20 shown in FIG. Fig.14In the illustration of FIG. 1 , the front electrode 10 is not shown, but only its radial inner edge 65 is shown, in order to be able to recognize the lead-through structure 63 arranged below the electrode 10 and designed as an emitter structure.

[0229] like Fig.14 It can be seen that in the short-circuit semiconductor device 66, the p-type doped front base region 8 in which the conducting structure 63 is embedded is provided with a p-type doping level significantly lower than that of the conventional p-type doped base region. -- Section 67, the p-type doped p -- The section exhibits a significantly higher resistance than the rest of the base region 8. At a penetration depth of preferably about 60 to 120 μm, the doping concentration of the resistive section 67 is preferably about 1E15 cm -3 to about 1E16cm -3 The resistor section 67 has a through-channel 68 opposite to the conductive structure section 64 , which is formed by the normally doped remaining p-type region of the base region 8 , and the conductive structure section 64 is directly electrically coupled to the trigger structure or gate electrode 20 via the p-type region of the base region 8 .

[0230] By short-circuiting the semiconductor device 66 Fig.14 The embodiment shown in FIG. 1 realizes that, on the one hand, a very small fuse current for melting can be realized by the segmented construction of the conductive structure 63, for example, also in Fig.13 On the other hand, the conduction current flowing from the trigger structure 20 to the conduction structure 63 is directed to the conduction structure section 64 in a targeted manner so as to safely conduct or destroy the conduction structure section in order to initiate melting when necessary. As a result, the short-circuit semiconductor device 66 can be designed in a targeted manner using the conduction current, which is the same as the conduction current in the short-circuit semiconductor device 62 shown in FIG. Fig.13 The short-circuit semiconductor components 62 shown in FIG. 6 are variously in the mA range, for example at approximately 150 mA.

[0231] The channel 68 can also take various different forms, for example, a trapezoidal form, in which the channel 68 gradually tapers from the trigger structure 20 toward the conductive structure segment 64, so as to achieve a more targeted conduction current flowing through the channel 68 to the conductive structure segment 64, and thus keep the conductive current flowing through the conductive structure segment 64 as small as possible.

[0232] Fig.15 Shown along Fig.14 The section line XX shown in Fig.14 6 is a vertical cross section of a short-circuited semiconductor device 66 shown in FIG. It can be seen that in the semiconductor body 69, especially in the front-side p-doped base region 8, p --The resistive segment 67 extends laterally between the trigger structure or gate electrode 20 and the inner edge 65 of the front electrode 10. The conductive structure segment 64 cannot be seen in the vertical section XX.

[0233] Fig.16 A partial top view of another embodiment of a short-circuit semiconductor device 70 according to the present invention is shown, which substantially corresponds to Fig.14 An embodiment of a short-circuit semiconductor device 66 is shown in Fig.16 The conductive structure segment 64 of the short-circuited semiconductor device 70 in the embodiment of the present invention is directly electrically connected to the trigger structure or gate 20 via a shunt resistor 71 in the form of an n-type doped (e.g., phosphorus-doped) zone arranged in the channel 68. The width of the shunt resistor 71 can preferably be determined to be approximately 55 μm. In this way, the conduction current required for turning on / pre-destroying the conductive structure 63 is reduced relative to Fig.14 The short-circuit semiconductor component 66 can again be slightly increased, for example to approximately 400 mA and preferably in the range of approximately 400 mA to approximately 1000 mA, with otherwise identical dimensions.

[0234] exist Fig.16 The serpentine shape of the shunt resistor 71 shown in FIG. 7 allows the length of the shunt resistor 71 to be varied in a desired manner, so as to thereby adjust the through current in the above-mentioned range almost arbitrarily. Fig.16 As shown in FIG. 1 , the length of the serpentine-shaped shunt resistor 71 within the channel 68 can be determined, for example, to be approximately 200 μm, ie, for example, to a value in the range of approximately 150 μm to approximately 250 μm.

[0235] The shunt resistor 71 is directly in electrical contact with the gate electrode 20 at its radially inner end. Fig.16 In the short-circuit semiconductor component 70 shown in FIG. 8 , the opposite radially outer ends of the shunt resistor 71 lead directly into the emitter structure section 64 .

[0236] Instead of Fig.16 In contrast to the arrangement shown in FIG. 6 in the through-channel 68 , the shunt resistor 71 can also be arranged on the resistor section 67 . In this case, the radially outer end of the shunt resistor 71 is directly in electrical contact with the front electrode 10 , for example at the electrode edge 65 .

[0237] Fig.17 A partial top view of a further embodiment of a short-circuit semiconductor component 72 according to the invention is shown, which has a conduction structure 63 open in the circumferential direction u of the semiconductor body and having at least one conduction structure segment 64 (here a ring segment) and a trigger structure 20 . Fig.17 The diagram basically corresponds to Fig.14 , except for Fig.17In the short-circuit semiconductor component 72 shown in FIG. 8 , an n-doped resistor section 73 is introduced into the p-doped front base region 8 . This resistor section is also referred to as a clamping resistor 73 .

[0238] Fig.18 Shown along the Fig.17 The section line YY shown in FIG. Fig.17 72. In this view, it can be seen that the resistor section 73 is formed by a clamping resistor, which, as an n-type region introduced from the front side 5 and adjacent to it in the semiconductor body 74 and in particular in the front base region 8, significantly reduces the effective conductive cross section of the front base region 8 between the trigger structure 20 and the front electrode 10, whereby the clamping resistor 73 causes an increase in the area-specific resistance in this region of the base region 8. The effects of these resistance increases have already been described above. Figures 14 to 16 describe.

[0239] It should be understood that Figures 13 to 18 The described embodiment of the conduction structure 63 with a separate conduction structure segment 64, in combination with a local increase in resistance in the base region 8 between the trigger structure 20 and the conduction structure segment 64, can be combined as desired with all other conduction structures and / or trigger structures described here and therefore encompasses further advantageous embodiments of the invention not shown in the figures. Thus, for example, two radially spaced-apart, electrically insulated conduction structures can also be provided, for example at least one as in Figures 13 to 18 The conductive structure segment 63 shown in FIG. 1 is closed by a ring and / or polygon arranged radially further outward, for example, similar to that in FIG. Figure 3 The finger-shaped conductive structure shown in FIG. 4 is surrounded by the finger-shaped conductive structure shown in FIG. As a result, the melted region produced after the conductive short-circuited semiconductor component is widened in an advantageous manner in the radial direction within the region of the semiconductor body covered by the front and / or back electrodes using the advantages already described here.

[0240] The short-circuit semiconductor device according to the present invention described above is not limited to the embodiments disclosed here, but also includes other embodiments with the same effect, which are derived from other technically meaningful combinations of the features described here of the short-circuit semiconductor device. In particular, an npn layer sequence opposite to the pnp structure of the semiconductor body shown in the embodiment can also be considered. In this case, the weakly n-doped semiconductor inner region consists of a weakly p-doped inner region, and the front and back base regions are respectively composed of n-doped layers, etc. In this case, the conduction type of the conductive structure described herein must also be reversed.

[0241] In a preferred embodiment, the short-circuit semiconductor component according to the invention is used as contact protection or personnel protection and for device protection.

[0242] Reference numerals:

[0243] 1 Existing thyristors

[0244] 2 Semiconductor body

[0245] 3 Central axis

[0246] 4 Back

[0247] 5 Front

[0248] 6 Back base region

[0249] 7 Internal Area

[0250] 8 Positive base region

[0251] 9 Front emitter

[0252] 10. Front electrode

[0253] 11 Back electrode

[0254] 12 Short Circuit

[0255] 13 Front and back center

[0256] 14 Gate electrode

[0257] 15 Short-circuited semiconductor devices

[0258] 16 Semiconductor body

[0259] 17 Front conduction structure / emitter structure

[0260] 18 Front contact plate

[0261] 19 Back contact pad

[0262] 20 Trigger structure, gate electrode

[0263] 21 Melt

[0264] 22 Short-circuited semiconductor devices

[0265] 23 Finger-shaped conduction structure / emitter structure

[0266] 24 Front electrode

[0267] 25 Semiconductor body

[0268] 26 Trigger structure, gate electrode

[0269] 27 Trigger structure finger / gate finger

[0270] 28 23 conductive structure finger

[0271] 29 Short-circuited semiconductor devices

[0272] 30 Front conduction structure / emitter structure

[0273] 31 Backside conduction structure / breakdown structure

[0274] 32 Semiconductor body

[0275] 33 Critical radius of curvature

[0276] 34 Short-circuit semiconductor devices

[0277] 35 Front conduction structure / emitter structure

[0278] 36 Back-conduction structure / emitter structure

[0279] 37 Positive trigger structure / gate electrode

[0280] 38 Back trigger structure / gate electrode

[0281] 39 Semiconductor body 40 41 42 43 44

[0287] 45 Short-circuit semiconductor devices

[0288] 46 Back-conduction structure / emitter structure

[0289] 47 Back trigger structure / breakdown structure

[0290] 48 semiconductor body 49 50 51

[0294] 52 Short-circuit semiconductor devices

[0295] 53 Front conduction structure / emitter structure

[0296] 54 Positive trigger structure / breakdown structure

[0297] 55 semiconductor body

[0298] 56 Short-circuit semiconductor devices

[0299] 57 Backside conduction structure / breakdown structure

[0300] 58 semiconductor body

[0301] 59 Short-circuit semiconductor devices

[0302] 60 Positive trigger structure / breakdown structure

[0303] 61 Semiconductor body

[0304] 62 Short-circuit semiconductor devices

[0305] 63 Front conduction structure / emitter structure

[0306] 64 conduction structure segment / emitter structure segment

[0307] 65 Radial inner edge

[0308] 66 Short-circuit semiconductor devices

[0309] 67 Resistor

[0310] 68 Flow channel

[0311] 69 semiconductor body

[0312] 70 Short-circuit semiconductor device

[0313] 71 Shunt resistor

[0314] 72 Short-circuit semiconductor devices

[0315] 73 Resistor / Clip resistor

[0316] 74 semiconductor body

[0317] 75 Predestruction between the first and second semiconductor junctions

[0318] 76 Short-circuit semiconductor devices

[0319] 77 Semiconductor body

[0320] 78 Insulation layer

[0321] 79 Electronic switch

[0322] 80 Parasitic resistance or inductance in the gate circuit

[0323] A Anode terminal

[0324] Al Aluminum

[0325] C Capacitor

[0326] D a The outer diameter of the BOD structure

[0327] D a* The outer diameter of the BOD structure

[0328] D i Inner diameter of the BOD structure

[0329] D i*The outer diameter of the BOD structure

[0330] E V Front electrode terminal

[0331] E R Back electrode terminal

[0332] G Gate terminal

[0333] G V Front gate terminal

[0334] G R Back gate terminal

[0335] K Cathode connection

[0336] L SF Radial finger length of the conductive structure

[0337] N doping concentration

[0338] Na concentration change line (a)

[0339] Nb concentration variation line (b)

[0340] Nc concentration change line (c)

[0341] r Radial / lateral

[0342] R Limiting resistance

[0343] u Circumferential direction

[0344] v Vertical direction

[0345] x penetration depth

[0346] W E1 The width of the horizontal electrode in the prior art

[0347] W ER Lateral back electrode width

[0348] W EV Horizontal front electrode width

[0349] W n Internal area thickness

[0350] W n1 Prior art inner area thickness

[0351] W S1 The width of the lateral emitter in the prior art

[0352] W SR Lateral back-side conductive structure width

[0353] WSV Width of lateral front conductive structure

[0354] W UV Circumferential positive conduction structure width

[0355] XX Vertical hatch

[0356] YY Vertical hatch line

Claims

1. A short-circuit semiconductor device, comprising: A semiconductor body (16, 25, 32, 39, 48, 55, 58, 61, 69, 74, 77), the semiconductor body comprising, starting from the back side (4) of the semiconductor body along a vertical direction (v), a back side base region (6) of a first conductivity type (p), an inner region (7) of a second conductivity type (n) complementary to the first conductivity type, and a front side base region (8) of the first conductivity type (p) arranged successively, A back side center (13) determined by its geometric center of gravity, facing the front side (5) opposite to the back side (4); The frontal center determined by its geometric center of gravity (13); The back base region (6) is connected to the back electrode (11) in a conductive manner, and the back electrode (11) has a lateral back electrode width (W ER ), The front base region (8) is connected to the front electrode (10) in a conductive manner, and the front electrode (10) has a lateral front electrode width (W EV ), wherein at least one front conductive structure (17, 23, 30, 35, 53, 63) is embedded in the front base region (8), the at least one front conductive structure is at least partially covered by the front electrode (10), and / or at least one back conductive structure (31, 36, 46, 57) is embedded in the back base region (6), the at least one back conductive structure is at least partially covered by the back electrode (11), wherein one or both of the at least one front conductive structure (17, 23, 30, 31, 35, 36, 46, 53, 63) and the at least one back conductive structure are emitter structures (17, 23, 30, 35, 36, 46, 53, 63) of the second conductive type (n), the emitter structure of the at least one front conductive structure is in conductive contact with the front electrode (10, 11), the emitter structure of the at least one back conductive structure is in conductive contact with the back electrode, the emitter structure is constructed to be turned on by at least one trigger structure (20, 26, 37, 38, 47, 54, 60), and the at least one trigger structure is activated by an electrical conduction signal supplied thereto, The at least one front side conductive structure has a lateral front side conductive structure width (W) given by its lateral extension. SV ), and the at least one back side conductive structure has a lateral back side conductive structure width (W given by its lateral extension SR ), wherein the width of the lateral front conductive structure (W SV ) and the corresponding lateral front electrode width (W EV ) is less than 1 and / or the width of the lateral back conductive structure (W SR ) and the corresponding lateral back electrode width (W ER ) is less than 1, It is characterized in that The at least one trigger structure (20, 26, 37, 38, 47, 54, 60) is configured to inject a surge current into a semiconductor body (16, 25, 32, 39, 48, 55, 58, 61, 69, 74, 77) between the at least one trigger structure (20, 26, 37, 38, 47, 54, 60) and at least one front side and / or at least one back side conductive structure (17, 23, 30, 35, 36, 46, 53, 63) so as not to reversibly destroying a first semiconductor junction formed between the at least one front conductive structure (17, 23, 30, 35, 36, 46, 53, 63) and the front base region (6, 8) and / or a second semiconductor junction formed between the front base region (6, 8) and the inner region (7), or irreversibly destroying a back first semiconductor junction formed between the back conductive structure and the back base region and / or a back second semiconductor junction formed between at least one back base region and the inner region, The front electrode and the back electrode each include a contact pad in contact with the semiconductor body, wherein the contact pads of the front electrode and the back electrode each include molybdenum.

2. The short-circuit semiconductor device according to claim 1, characterized in that: The energy input is predetermined to be in the range from 1 Ws to 3 Ws.

3. The short-circuit semiconductor device according to claim 1 or 2, characterized in that: The semiconductor body (16, 25, 32, 39, 48, 55, 58, 61, 69, 74, 77) has a resistance of less than 10Ω in a radial direction (r) in a region where a surge current is injected, between the trigger structure (20, 26, 37, 38, 47, 54, 60) and the front and / or back conductive structure (17, 23, 30, 35, 36, 46, 53, 63).

4. The short-circuit semiconductor device according to any one of the preceding claims 1 to 2, characterized in that: in, The front side transverse conductive structure width (W SV ) and the corresponding lateral front electrode width (W) of each electrode (10, 11) contacting the corresponding conductive structure EV ) is less than 4 / 5, or the width of the back side lateral conductive structure (W SR ) and the corresponding lateral back electrode width (W) of each electrode contacting the corresponding conductive structure ER ) is less than 4 / 5.

5. The short-circuit semiconductor device according to any one of the preceding claims 1 to 2, characterized in that: The at least one front side conductive structure (31, 46, 53, 57) is completely covered by the front side electrode (10) and the at least one back side conductive structure is completely covered by the back side electrode (11).

6. The short-circuit semiconductor device according to any one of the preceding claims 1 to 2, characterized in that: The at least one front side conductive structure (31, 46, 53, 57) is covered by at least 5 mm in a radial direction (r) by a front side electrode (10, 11) contacting the conductive structure.

7. The short-circuit semiconductor device according to any one of the preceding claims 1 to 2, characterized in that: The front electrode (10) and the back electrode (11) respectively cover the front center or the back center (13) of the semiconductor body (16, 25, 32, 39, 48, 55, 58, 61, 69, 74), and the at least one front conductive structure (17, 23, 30, 31, 35, 36, 46, 53, 57, 63) is arranged at the center (13) of the front electrode and the at least one back conductive structure is arranged at the center of the back electrode.

8. The short-circuit semiconductor device according to any one of the preceding claims 1 to 2, characterized in that: An electrically insulating insulating layer (78) is arranged between the at least one front conductive structure (17, 23, 30, 31, 35, 36, 46, 53, 57, 63) and the front electrode (10, 11), wherein the insulating layer (78) extends on the at least one front conductive structure (17, 23, 30, 31, 35, 36, 46, 53, 57, 63) in a radial direction (r) and extends in a circumferential direction (u) along an inner edge of the front electrode (10, 11) located inside in the radial direction (r).

9. The short-circuit semiconductor device according to claim 8, characterized in that: The insulating layer (78) has a thickness of at least 10 nm to at most 100 nm.

10. The short-circuit semiconductor device according to claim 8, characterized in that: The insulating layer (78) is a silicon dioxide layer.

11. The short-circuit semiconductor device according to claim 8, characterized in that: The insulating layer (78) has openings which are spaced apart from one another in a radial direction (r) and which establish a direct electrically conductive connection between the front electrodes (10, 11) and the front base regions (6, 8) which abut against the front electrodes.

12. The short-circuit semiconductor device according to any one of the preceding claims 1 to 2, characterized in that The trigger structure is a gate electrode (20, 26, 37, 38) and the turn-on signal is a surge current injected from the trigger structure (20, 26, 37, 38) into the semiconductor body (16, 25, 32, 39, 48, 55, 58, 61, 69, 74, 77).

13. The short-circuit semiconductor device according to claim 12, characterized in that: A capacitor (C) having a predetermined capacitance corresponding to a predetermined energy input is provided, which provides the on-signal in a charged state, wherein the capacitor (C) can be electrically switched between the gate electrode (20, 26, 37, 38) and the front electrode (10) and / or the back electrode (11) by means of an electronic switch (79).

14. The short-circuit semiconductor device according to claim 12, characterized in that: The gate electrodes (20, 26, 37, 38) are coated with a silver layer on their metal contact sides which are located on the outside in the vertical direction (v).

15. The short-circuit semiconductor device according to any one of the preceding claims 1 to 2, characterized in that The trigger structure is a breakdown structure (47, 54, 60), and the turn-on signal is a voltage applied between the front electrode (10) and the back electrode (11).

16. The short-circuit semiconductor device according to any one of the preceding claims 1 to 2, characterized in that: The at least one front conductive structure (17, 23, 30, 35, 36, 46, 53, 63) is spaced apart from the trigger structure (20, 26, 37, 38, 60) in a radial direction (r) and has at least one conductive structure segment (64) having a conductive structure width (W) extending in a circumferential direction (u) of the semiconductor body (16, 25, 32, 39, 48, 55, 58, 61, 69, 74). UV ), the conductive structure width is smaller at each location than a closed perimeter extending around the center (13) along a circumferential direction (u) through at least one front conductive structure segment (64).

17. The short-circuit semiconductor device according to claim 16, characterized in that: At least two mutually separated conductive structure segments (64) are arranged in a circumferential direction (u) of the semiconductor body (16, 25, 32, 39, 48, 55, 58, 61, 69, 74).

18. The short-circuit semiconductor device according to claim 16, characterized in that: The resistance in the base region (6, 8) containing the conduction structure (17, 23, 30, 35, 36, 46, 53, 63) is increased at least in a section (67, 73) of a radial gap between the conduction structure (17, 23, 30, 35, 36, 46, 53, 63) and the trigger structure (20, 26, 37, 38, 60), except for at least one forward channel (67) electrically coupling at least one of the conduction structure sections (64) with the trigger structure (20, 26, 37, 38, 60).

19. The short-circuit semiconductor device according to any one of the preceding claims 1 to 2, characterized in that: A region of a base region (6, 8) of the conduction structure (17, 23, 30, 35, 36, 46, 53, 63) embedded between the trigger structure (20, 26, 37, 38, 47, 54, 60) and a corresponding conduction structure (17, 23, 30, 35, 36, 46, 53, 63) is constructed so that the conduction structure (17, 23, 30, 35, 36, 46, 53, 63) can be turned on with an overmodulation factor of at most 10 by means of the trigger structure (20, 26, 37, 38, 47, 54, 60), wherein a minimum conduction current causing the conduction structure (17, 23, 30, 35, 36, 46, 53, 63) to be turned on is at least 1A.

20. The short-circuit semiconductor device according to any one of the preceding claims 1 to 2, characterized in that The thickness (W) of the inner region (7) in the vertical direction (v) is n ) is greater than 400μm.

21. The short-circuit semiconductor device according to any one of the preceding claims 1 to 2, characterized in that The thickness of the front base region (8) and / or the back base region (6) in the vertical direction (v) is at least 50 μm.

22. The short-circuit semiconductor device according to any one of the preceding claims 1 to 2, characterized in that The front electrode (10) and the back electrode (11) are each contacted with each other in an electrically conductive manner by means of metal contact pads (18, 19).

23. The short-circuit semiconductor device according to claim 22, characterized in that: The lateral width of each contact pad (18, 19) corresponds to the lateral electrode width (W) of the corresponding pressure contact electrode (10, 11). EV , W ER ).

24. A method for operating a short-circuit semiconductor device (15, 22, 29, 34, 45, 52, 56, 59, 62, 66, 70, 72), the short-circuit semiconductor device comprising a semiconductor body (16, 25, 32, 39, 48, 55, 58, 61, 69, 74), the semiconductor body having, starting from the back side (4) along the vertical direction (v), a back side base region (6) of a first conductivity type (p), an inner region of a second conductivity type (n) complementary to the first conductivity type, and a back side base region (6) of a first conductivity type (p) and an inner region of a second conductivity type (n) complementary to the first conductivity type. (7), and a front base region (8) of the first conductivity type, at least one front conduction structure (17, 23, 30, 35, 53, 63) is embedded in the front base region (8), and / or at least one back conduction structure (31, 36, 46, 57) is embedded in the back base region (6), and at least one trigger structure is coupled to the front base region or the back base region, and the trigger structure turns on the at least one front conduction structure and / or the at least one back conduction structure, the method comprising: The at least one trigger structure (20, 26, 37, 38, 47, 54, 60) is activated by an electrical conduction signal by fusing the semiconductor body to a metal region of a contact pad from the at least one front conductive structure, wherein the semiconductor body (16, 25, 32, 39, 48, 54, 60) is connected between the at least one trigger structure (20, 26, 37, 38, 47, 54, 60) and at least one front conductive structure (17, 23, 30, 35, 36, 46, 53, 63) or between the at least one trigger structure and at least one back conductive structure. 55, 58, 61, 69, 74) to inject a surge current to irreversibly destroy at least one of a first semiconductor junction formed between the at least one front conduction structure (17, 23, 30, 35, 36, 46, 53, 63) and the front base region (6, 8), a second semiconductor junction formed between the front base region (6, 8) and the inner region (7), a back first semiconductor junction formed between the at least one back conduction structure and the back base region, or a back second semiconductor junction formed between the back base region and the inner region.

Images