SEMICONDUCTOR COMPONENT WITH AN EDGE TERMINATION STRUCTURE

Abstract

Semiconductor device comprising: a semiconductor body (100) having a first main surface (101), an edge surface (103), an inner region (110) and an edge region (120) arranged between the inner region (110) and the edge surface (103); a first semiconductor region (11) of a first doping type arranged in the inner region (110), and a second semiconductor region (21) of a second doping type arranged in the inner region (110) and the edge region (120), wherein a pn junction is formed between the first semiconductor region (11) and the second semiconductor region (21); and an edge termination structure (30) comprising: a third semiconductor region (31) of the first doping type arranged in the edge region (120) and adjacent to the first semiconductor region (11), wherein a doping dose of the third semiconductor region (31) decreases towards the edge surface (103); a surface portion (33) of the second semiconductor region (21) adjacent to the first main surface (101); and an amorphous passivation layer (32) having a specific resistance of higher than 10 9 Ωcm, which is formed on the first main surface (101) and which adjoins the third semiconductor region (31) and the surface portion (33) of the second semiconductor region (21), wherein the edge termination structure (30) in the semiconductor body has a width (w30) (100) in the lateral direction, wherein an electrically active doping dose of the third semiconductor region (31) at a lateral position (x1) spaced from the first semiconductor region (11) by 50% of the width (w30) of the edge termination structure (30) is at least Q BR / q, where Q BR is the breakdown charge of the semiconductor material of the semiconductor body (100) and q is the elementary charge, and wherein the amorphous passivation layer (32) is realized such that a density of states N F in the passivation layer (31) is given by: NF ≥ 1 ε ⋅ (QBRE g ) 2 , where Q BR is the breakdown charge of the semiconductor material of the semiconductor body (100), E g is the band gap of the semiconductor material of the semiconductor body (100), and ε=ε0·ε r is the dielectric constant of the semiconductor material of the semiconductor body (100).

Description

This description relates generally to a semiconductor device, in particular a semiconductor device having an edge termination structure.Power semiconductor devices such as power diodes, power MOSFETs, power IGBTs, or power thyristors are designed to withstand high reverse voltages. These power devices comprise a pn junction formed between a p-doped semiconductor region and an n-doped semiconductor region. The device disables (is off) when the pn junction is reverse biased by applying an appropriate voltage to the pn junction. In this case, a depletion region (space charge region) spreads in the p-doped region and the n-doped region. Usually, one of the p-doped region and the n-doped region is more lightly doped than the other of the p-doped region and the n-doped region, so that the depletion region mainly propagates in the more lightly doped region, which substantially takes over the voltage applied to the pn junction. The more lightly doped region taking on the reverse voltage is usually referred to as a base region in a diode or a thyristor and as a drift region in a MOSFET or an IGBT.The ability of a pn junction to take high voltages is limited by the avalanche breakdown phenomenon. As a voltage applied to the pn junction increases, an electric field increases in the semiconductor regions forming the pn junction. The electric field results in an acceleration of mobile carriers induced by thermal generation in the space charge region. Avalanche breakdown occurs when, due to the electric field, the charge carriers are accelerated in such a way that they generate electron-hole pairs by impact ionization. Carriers generated by impact ionization generate new carriers, so that a multiplication effect is present. When avalanche breakdown begins, a considerable current flows across the pn junction in the reverse direction. The electric field at which avalanche breakdown begins is referred to as critical electric field. The absolute value of the critical electric field is mainly dependent on the type of semiconductor material used for forming the pn junction and is weakly dependent on the doping concentration of the more lightly doped semiconductor region. A reverse voltage withstand capability of the semiconductor device is the voltage applied to the pn junction at which the critical electric field occurs in the semiconductor device. This voltage is often referred to as a breakdown voltage.The reverse voltage resistance is not only dependent on the type of semiconductor material and its doping, but also on the specific geometry of the semiconductor component. A power semiconductor device comprises a semiconductor body of limited size, which is bounded in lateral directions of the semiconductor body by edge surfaces. A vertical power semiconductor device is a semiconductor device in which the pn junction extends mainly in a horizontal plane of the semiconductor body. In such a vertical power semiconductor device, the pn junction typically does not extend as far as the edge surface of the semiconductor body. Instead, the pn junction is spaced apart from the edge surface of the semiconductor body in a lateral direction. In this case, a semiconductor region (edge region) of the semiconductor body adjacent to the pn junction in the lateral direction also needs to withstand the voltage applied to the pn junction.The edge region may be realized with a planar edge termination structure, which may include a passivation layer formed on a surface of the semiconductor body in an edge region of the semiconductor body. In a manufacturing process of the semiconductor device and / or during operation of the semiconductor device, the passivation layer may be contaminated by ions such as alkali ions, hydronium (H 3 O +)- ions or hydroxide (OH -)- ions. These ions may influence the electric field distribution in the edge region by an electrostatic induction effect such that a reverse voltage strength of the semiconductor device is reduced.DE 10 2005 004 355 A1 describes a semiconductor component having a semiconductor body which has an inner region and an edge region, wherein an n-doped semiconductor region is arranged in the inner region and in the edge region. In the inner region, the component also has a p-doped region which forms a pn junction with the n-doped region. In the edge region, the device has an edge termination structure comprising a p-doped VLD region and a passivation layer on the VLD region.DE 103 58 985 B3 and DE 10 2014 108 986 A1 each describe a semiconductor component having an edge termination structure which comprises an amorphous semi-insulating passivation layer above a p-doped semiconductor region in the edge region of a semiconductor body.U.S. Pat. No. 5,093,693 A describes a semiconductor component having a p-doped guard ring (guard ring) which surrounds a p-doped guard ring arranged in an inner region of a semiconductor body. The guard ring has a doping of about 1E15 cm -3 and extends between 40 micrometers and 80 micrometers into the semiconductor body.The object on which the invention is based is to provide a semiconductor component having a planar edge termination which is robust with respect to parasitic ions (charges).This object is achieved by a semiconductor component according to claim 1.Examples are explained below with reference to the drawings. The drawings are intended to illustrate certain principles, so that only aspects necessary to understand these principles are illustrated. The drawings are not to scale. In the drawings, like reference numerals designate like features. FIG. 1A shows a vertical cross-sectional view of a portion of a semiconductor device comprising a pn junction in an inner region and an edge termination structure with an amorphous passivation layer in an edge region; FIG. 1B illustrates an example of a doping dose profile in the semiconductor device illustrated in FIG. 1A ; FIG. 2 illustrates an example of a plan view of the semiconductor device shown in FIG. 1 ; FIG. 3A illustrates the critical electric field at an abrupt pn junction formed between two complementarily doped semiconductor regions depending on a doping concentration of a lower doped one of the two semiconductor regions; FIG. 3B illustrates the breakdown charge relative to the elementary charge of a more heavily doped one of the two semiconductor regions forming the pn junction, depending on a doping concentration of the less heavily doped one of the two semiconductor regions; FIGS. 4A and 4B illustrate the dependence of a reverse voltage withstand capability of the semiconductor device depending on a doping dose profile of a semiconductor region of the edge termination structure in a semiconductor device of the type illustrated in FIG. 1 and in a conventional semiconductor device; FIGS. 5A and 5B illustrate the dependence of a reverse voltage withstand capability of a semiconductor device of the type illustrated in FIG. 1 on a slope of a doping dose profile and the doping dose; FIGS. 6A and 6B illustrate another example of a dependence of a reverse withstand voltage of a semiconductor device of the type illustrated in FIG. 1 on a slope of a doping dose profile and the doping dose; FIGS. 7A and 7B illustrate a dependence of a lateral electric field in a semiconductor device of the type illustrated in FIG. 1 on a slope of a doping dose profile; FIG. 8 is a vertical sectional view of a portion of a semiconductor device realized as a diode; FIG. 9 shows a vertical sectional view of a portion of a semiconductor component configured as a transistor component; and FIG. 10 illustrates an example of a method for forming a doped semiconductor region of the edge termination structure.In the following detailed description, reference is made to the accompanying drawings. The drawings form a part of the specification and show by way of illustration examples of how the invention may be used and practiced. Of course, the features of the various exemplary embodiments described herein can be combined with one another, unless explicitly stated otherwise.FIG. 1A schematically illustrates a vertical cross-sectional view of a portion of a semiconductor device. The semiconductor device comprises a semiconductor body 100 having a first main surface 101, an edge surface 103, an inner region 110 and an edge region 120. The edge region 120 is arranged between the inner region 110 and the edge surface 103. The semiconductor body 100 may be made of a conventional semiconductor material such as silicon (Si), silicon carbide (SiC), gallium nitride (GaN), gallium arsenide (GaAs), or the like.FIG. 2 illustrates a top view of the first surface 101 of the semiconductor body 100 according to an example to illustrate a position of the inner region 110 relative to the edge region 120 in the semiconductor body. Referring to FIG. 2, the edge region 120 surrounds the inner region 110 and separates the inner region 110 from the edge surface 103. Merely for the purpose of illustration, the semiconductor body 100 is rectangular in the example shown in FIG. 2. However, this is only an example. According to another example (not shown), the semiconductor body 100 is circular.Referring to FIG. 1A, the semiconductor device further comprises a first semiconductor region 11 of a first doping type, wherein the first semiconductor region 11 is arranged in an inner region 110 of the semiconductor body 100. In the example illustrated in FIG. 1, the first semiconductor region 11 adjoins the first surface 101. However, this is only an example. According to further examples, which are explained further below, it is also possible to realize the first semiconductor region 11 such that it is spaced apart from the first surface 101. The semiconductor device also comprises a second semiconductor region 21 of a second doping type, wherein the second doping type is complementary to the first doping type. The second semiconductor region 21 is arranged in the inner region 110 and in the edge region 120. In addition, the second semiconductor region 21 in the inner region 110 adjoins the first semiconductor region 11 such that a pn junction is formed between the first semiconductor region 11 and the second semiconductor region 21. Referring to FIG. 1, the second semiconductor region 21 may be adjacent to the edge surface 103 of the semiconductor body 100. According to an example, the first semiconductor region 11 is a p-type region and the second semiconductor region 21 is an n-type region.Referring to FIG. 1A, the semiconductor device further comprises an edge termination structure 30. this edge termination structure 30 comprises a third semiconductor region 31 of the first doping type. This third semiconductor region 31 is arranged in the edge region 120 and adjoins the first semiconductor region 11. The third semiconductor region 31 may be adjacent to the first surface 101. In addition, a doping dose of the third semiconductor region 31 decreases towards the edge surface 103. This includes that at least a portion of the third region 31 is present in which the doping dose decreases with decreasing distance to the edge surface 103. However, there may also be portions in which the doping dose is substantially constant over a certain length in a direction towards the edge surface 103. "The doping dose" is the integral of the doping concentration of the third semiconductor region 31 in a vertical direction y of the semiconductor body 100, wherein the vertical direction y is a direction perpendicular to the first surface 101. The doping dose of the third region 31 is in particular the effective doping dose, which is the integral of the effective and electrically active doping concentration of the third region 31.The third semiconductor region 31 adjoins the second semiconductor region 21 in the edge region 120 of the semiconductor body 100. In addition, the edge termination structure 30 comprises a portion 33 of the second semiconductor region 21 adjoining the first surface 101. This portion 33 of the second semiconductor region 21 is referred to below as the surface portion 33 of the second semiconductor region 21 or fourth semiconductor region 33.Additionally, the edge termination structure 30 comprises an amorphous passivation layer 32 on the first surface 101 in the edge region 120. The amorphous passivation layer 32 adjoins the third semiconductor region 31 and the surface portion 33 of the second semiconductor region 21 and can also cover a portion of the first semiconductor region 11 as illustrated in FIG. 1. The amorphous passivation layer 32 is semi-insulating and has a resistivity higher than 1E9 (=10 9) Ωcm (ohm·cm). According to an example, the resistivity is higher than 1E10Ωcm or higher than 1E11Ωcm. In addition, the resistivity is less than 1E14Ωcm, according to one example.A pn junction between the first region 11 and the second region 21 may be reverse biased by applying an appropriate voltage between the first region 11 and the second region 21. When the pn junction is reverse biased, a space charge region (depletion region) spreads in the first region 11 and the second region 21. According to an example, the first region 11 is doped higher than the second region 21. According to an example, an (average) doping concentration of the first region 11 is at least 10 times or at least 100 times an (average) doping concentration of the second region 21. According to an example, a doping dose in this implantation process is such that an (average) doping concentration of the first region 11 is selected from between 1E16 cm -3 and 1E18 cm -3, wherein the doping concentration is substantially given by the doping dose divided by a depth of the first region 11 in a vertical direction of the semiconductor body 100, which is a direction perpendicular to the first surface 101. The second region 21 may be formed by an epitaxial growth process, wherein the first region 11 and the doped regions of the edge termination structure 30 may be formed by implanting dopant atoms into the epitaxial layer. A doping concentration of the second region 21 is selected from between 1E 12 cm -3 and 1E 17 cm -3. for example.A pn junction is also formed between the third region 31 and the second region 21. When the pn junction between the first region 11 and the second region 21 is reverse-biased, the pn junction between the third region 31 and the second region 21 is also reverse-biased, so that a depletion region also spreads in the third region 31 and the second region 21 in the edge region 120. Avalanche breakdown occurs when an amount of voltage that poles the pn junction in the reverse direction is such that an amount of electric field at a position reaches a critical value, commonly referred to as E crit. This critical value E crit depends on a doping concentration of the lower doped region 21 and, of course, the type of semiconductor material. FIG. 3A illustrates the critical value E crit depending on the doping concentration of the second region 21 in a semiconductor device comprising a monocrystalline silicon (Si) semiconductor body 100. As can be seen from FIG. 3A, the critical value E crit increases as the doping concentration increases. More specifically, for example, as the doping concentration increases from 1E12cm -3 to 1E17cm -3 the critical value E crit increases from about 1.7E5V / cm to about 6.0E5V / cm. For example, when the semiconductor body 100 is made of 6H-SiC type silicon carbide (SiC), the critical value E crit is higher and increases from about 1E6V / cm to about 4E6V / cm as the doping concentration of the second region 21 increases from 1E14 cm -3 to 5E17 cm -3.A "blocking voltage stability" of the semiconductor component is given by a voltage level of the reverse-polarity voltage at which the critical value E crit is reached. The "reverse-biasing voltage" is the voltage that reverse-biases the pn junctions between the first region 11 and the second region 21 and between the third region 31 and the second region 21. In the inner region 110, the pn-junction is substantially planar and parallel to the first surface 101, while the pn-junction is curved in the edge termination structure 30. It is generally known that the same voltage applied to a planar pn junction and to a curved pn junction causes a higher electric field at the curved pn junction than at the planar pn junction. Therefore, the reverse blocking voltage strength in the inner region 110 is usually higher than in the edge region, wherein the lower reverse blocking voltage strength in the edge region 120 defines the overall reverse blocking voltage strength of the semiconductor device. Therefore, the edge region 120 is intended to achieve a certain reverse voltage strength of the semiconductor device. However, the edge region 120 does not improve other parameters of the semiconductor device, such as conduction losses. In principle, it is desirable to (a) realize the edge region 120 with a small area compared to an area of the inner region 120; (b) achieve a high blocking voltage resistance of the semiconductor device in the edge region (120); and configure the edge termination structure 30 to be robust with respect to parasitic charges, such as alkali, hydronium or hydroxide ions, which may be introduced into the passivation layer 33 in a manufacturing process of the semiconductor device and / or during operation of the semiconductor device.In the edge termination structure illustrated in FIG. 1, this is achieved by realizing the third semiconductor region 31 such that the doping dose decreases towards the edge surface 103 but still has a relatively high doping dose in the middle of the edge termination structure. Referring to FIG. 1, the edge termination structure 30 has a total width w 30 in a lateral direction x, which is a direction from the first region 11 towards the edge region 103. The width w 30 of the edge termination structure 30 is given by a width w 31 of the third region 31 in the lateral direction x and a width w 33 of the surface portion 33 in the lateral direction x.In the example shown in FIG. 1, the semiconductor device comprises a channel stopper 22. the channel stopper is of the second doping type and is doped higher than the second region 21. according to one example, a doping concentration of the channel stopper 22 is at least 1E2times the doping concentration of the second region 21. In this example, the width w 30 of the edge termination structure 30 in the semiconductor body 100 is given by a (shortest) distance between the first region 11 and the channel stopper 22. When the channel stopper 22 is omitted, the surface portion 33 terminates at the edge surface 103. In this case, the width w 30 of the edge termination structure 30 in the semiconductor body 100 is given by a (shortest) distance between the first region 11 and the edge surface 103.The "center" of the edge termination structure 30 is located at a first position x 1, wherein this first position x 1 is spaced apart from the first region 11 by a distance that is 50% of the total width w 30 of the edge termination structure 30. At this first position x 1, the doping dose of the third region 31 is at least Q BR / q, wherein Q BR is the breakdown charge of the semiconductor material of the semiconductor body 100 and q is the elementary charge (1.6E-19C). Q BR / q is also referred to below as breakthrough dose S BR.As stated above, a depletion region spreads in the second region 21 when the pn junction between the third region 31 and the second region 21 is reverse-biased. Such a depletion region is associated with ionization of dopant atoms in the second region 21 and the third region 31. The breakdown charge Q BR is the integral of the ionized dopant atoms per unit area in the second region 21 in a direction perpendicular to the pn junction when the critical value E crit occurs at the pn junction. The breakdown charge Q BR is related to the critical value E crit as follows: where ε 0 is the dielectric constant in vacuum and ε r is the relative dielectric constant of the material of the semiconductor body 100. The relative dielectric constant ε r of silicon is, for example, about 11.9, and the relative dielectric constant ε r of 6H-SiC is, for example, about 9.66. Referring to FIG. 3A, because the critical value E crit is dependent on the doping concentration of the second region 21 and the breakdown charge Q BR is dependent on the critical value E crit, the breakdown charge Q BR is also dependent on the doping concentration of the second region 21. FIG. 3B shows the breakdown dose D BR(= Q BR / q) of a semiconductor device comprising a silicon semiconductor body 100 depending on the doping concentration of the second region 21.The relatively high doping dose of the third region 31, i.e. a doping dose S that is higher than the breakdown dose S BR in the center x 1 of the edge termination structure 30, in combination with the amorphous passivation layer 32, makes the edge termination structure 30 robust with respect to parasitic charges. This is illustrated below with reference to FIGS. 4A and 4B. In FIG. 4A, the curve 201 represents the doping dose S of the third region 31 according to an example. In this example, the doping dose S has a maximum value S 0_301 at a position x0at which the third region 31 adjoins the first region 11. This position x0is referred to below as the beginning of the edge termination structure 30. In the example illustrated in FIG. 4A, the doping dose is substantially constant up to the center x 1 or beyond and then decreases. This maximum value S 0_301 is higher than the breakdown dose S BR.As stated above, forming the first region 11 may include a first implantation process. In this process, an implantation mask (not shown) is formed on the first surface 101 of the semiconductor body such that those regions of the surface 101 below which the first region 11 is to be formed are uncovered by the implantation mask, while other regions are covered by the implantation mask. Furthermore, dopant atoms of the first doping type are implanted into the semiconductor body 100 via the first surface 101 using the implantation mask. Accordingly, forming the third region 31 may include a second implantation process in which dopant atoms of the first doping type are implanted. In this process, an implantation mask covers the inner region 110 while the edge region 120 is at least partially uncovered. An example of a process for forming the third region 31 will be explained below with reference to FIG. 10.Forming the first region 11 and the third region 31 may further comprise a thermal process in which the implanted dopant atoms are diffused and activated, wherein a common thermal process may be used to form both the first region 11 and the third region 31. The effective doping dose of the completed first region 11 is the integral of the doping concentration of the dopant atoms of the first doping type minus the dopant atoms of the second doping type (which may result from a basic doping of the semiconductor body 100 before the implantation process) in the vertical direction y in the first region 11. Nevertheless, given a basic doping of the semiconductor body 100 and a known ratio of activated versus non-activated dopant atoms in the thermal process, the effective doping dose of the first region 11 may be relatively accurately adjusted by suitably adjusting the implantation dose in the first implantation process, which is part of forming the first region 11. Accordingly, the effective doping dose of the third region 31 may be relatively accurately adjusted by appropriately adjusting the implantation dose in the second implantation process that is part of forming the third region 31.FIG. 1B schematically illustrates the doping dose profile of the first region 11 and the third region 31 in a region comprising a transition between the first region 11 and the third region 31. Referring to FIG. 1B, the effective doping dose of the first region 11 is substantially constant at positions spaced apart from the first region 31, wherein S 11 denotes the effective doping dose at these positions. This doping dose S 11 is defined by the implantation dose in the first implantation process. Accordingly, an effective doping dose S 31 of the third region 31 is defined by the implantation dose in the second implantation process. This effective doping dose may decrease in the lateral direction x, as illustrated in solid lines in FIG. 1B, or may be substantially constant over a certain distance in the lateral direction x, as illustrated in dashed lines. The first region 11 also includes a transition region. Due to the diffusion in the thermal process, this transition region may include dopant atoms from the implantation process being part of forming the first region 11 and the implantation process being part of forming the third region 31. In this transition region, the effective doping dose rapidly decreases from S 11 to S 31.According to an example, the effective doping dose S 11, defined by the implantation dose in the first implantation process, is substantially higher than 1.5 times the breakdown dose S BR, such as between 5 and 10 times S BR. In this example, the beginning x0of the third region 31 may be defined as a position at which the effective doping dose in the transition region has dropped to 1.5 times the doping dose S BR.Referring to FIG. 1B, an amount of a slope of the doping dose profile in the lateral direction x has a first local minimum (which is zero) in the range in which the effective dose S is 11 (this is, for example, a region not covered by the passivation layer 32). In the transition region, the magnitude of the slope increases from the first local minimum to a maximum and then decreases toward the third region, where the slope reaches a second local minimum, which may be zero (compare the dashed line) or may be different from zero (compare the solid line). Thus, according to a further example, the beginning x 0 of the third region 31 is at a position at which an amount of the slope of the doping profile S reaches a second minimum when moving from the first region 11 towards the third region 31 along the lateral direction x.Curve 202 in FIG. 4A illustrates the doping dose of a third region in a conventional edge termination structure comprising an oxide layer instead of an amorphous passivation layer (and without considering surface charges). In this case, the doping dose of the third region also decreases towards the edge surface 103, but has a lower maximum doping dose, i.e. a maximum doping dose S 0_302 of the third region in the conventional edge termination structure is lower than the maximum doping dose S 0_301 of the third region in an edge termination structure of the type illustrated in FIG. 1. ΔS 0 denotes the difference between these two maximum values S 0_301, S 0_302.FIG. 4B illustrates the reverse voltage strength V BR of an edge termination structure of the type illustrated in FIG. 1 and a conventional edge termination structure depending on the maximum doping dose S 0. More specifically, curve 301 shown in FIG. 4B illustrates the blocking voltage capability V BR of the edge termination structure of the type shown in FIG. 1 and having a doping dose profile of the third region 31, as illustrated by curve 201 in FIG. 4A, and curve 302 illustrates the blocking voltage capability of a conventional edge termination structure having a doping dose profile of the third region, as illustrated by curve 202 shown in FIG. 4A. The curves 301, 302 each illustrate the reverse voltage strength depending on the respective maximum doping dose. As can be seen from FIG. 4B, there is in each case an optimum maximum doping dose S 0_301_OPT, S 0_302_OPT, wherein the blocking voltage withstand V BR has a maximum when the maximum doping dose S 0_301, S 0_302 has the respective optimum value S 0_301_OPT, S 0_302_OPT. The maximum blocking voltage strengths are substantially the same, wherein the maximum blocking voltage strength of the edge termination structures according to FIG. 1 is slightly lower than the maximum blocking voltage strength of the conventional edge termination structure. This is seen by comparing curve 301 with curve 302', curve 302' resulting from curve 302 by shifting ΔS 0 to the right.According to an example, the maximum doping dose S 0_302_OPT of the conventional edge termination structure in a silicon-based semiconductor device is about 1.2 E12 cm -2 and the maximum doping dose S 0_301_OPT in a silicon-based semiconductor device is about 1.73 E12 cm -2, such that in this example ΔS 0 is about 5.3 E11 cm -2. The maximum blocking voltage strengths in connection with these edge termination structures are about 8200 V, wherein the maximum blocking voltage strength in the inner region 11 is about 8600 V.The curves 301, 302 illustrated in FIG. 4B have substantially the shape of a parabola and can therefore also be referred to below as blocking parabolas. As can be seen from FIG. 4B, the blocking label 301 associated with the edge termination structure of the type shown in FIG. 1 is wider than the blocking label 302 associated with the conventional edge termination structure. In any case, the reverse voltage stability V BR decreases if the doping dose deviates from the respective optimum doping dose S 0_301_OPT, S 0_302_OPT. That is, if the doping dose in the conventional edge termination structure deviates from the optional doping dose S 0_302_OPT by a certain percentage, the blocking voltage strength decreases by a first percentage compared to the maximum blocking voltage strength V BR_302. Accordingly, if the doping dose in the edge termination structure 30 deviates from the optional doping dose S 0_301_OPT by the certain percentage, the blocking voltage withstand V BR decreases by a second percentage compared to the maximum blocking voltage withstand V BR_301. Since the blocking label 301 associated with the edge termination structure according to FIG. 1 is further than the blocking label 302 associated with the conventional edge termination structure, the second percentage is less than the first percentage. This means that the edge termination structure according to FIG. 1 is more robust with respect to parasitic charges in the passivation layer 32, since these parasitic charges have the same effect as a variation of the doping dose in the third semiconductor region 31.According to an example, the amorphous passivation layer has a relatively high state density N F. More specifically, the density of states N F in the passivation layer 32 is given by: where Q BR is the breakdown charge explained above, ε = ε 0 · ε r is the dielectric constant of the semiconductor material of the semiconductor body 100, and E g is the band gap of the semiconductor material of the semiconductor body 100. The band gap depends on the type of semiconductor material and the temperature. In silicon at a temperature of 300K, the band gap E is g for example 1.12 eV, and in 6H-SiC, the band gap E is g at 300K 3.03 eV. The density of states (DOS) in the passivation layer 32 can be measured by optical measurement methods. This is described, for example, in K. Chew et al.: "Gap state distribution in amorphous hydrogenated silicon carbide films reduced from photothermal deflection spectroscopy", JOURNAL OF APPLIED PHYSICS, VOLUME 91, NUMBER 7, 1 April 2002, pages 4319-4325. The amorphous passivation layer 32 is made of one of the following materials according to one example: amorphous hydrogen-containing carbon (aC:H), amorphous silicon (aSi), amorphous silicon carbide (aSiC), or amorphous hydrogen-containing silicon carbide (aSiC:H). A thickness of the amorphous passivation layer 32 is between 200 nanometers (nm) and 1 micrometer (μm), for example.The amorphous passivation layer 32 may be formed by a chemical vapor deposition (CVD) process, particularly a radio frequency (RF) plasma enhanced chemical vapor deposition (PECVD) process. An RF PECVD process is well known, so that no details of this type of process are necessary. In principle, a PECVD process comprises arranging the semiconductor body 100 in a reactor, introducing process gases into the reactor, activating such gas species (gas species) that are to be deposited on the semiconductor body 100 by plasma, adsorbing the activated gas species on the first surface 101, thereby forming the solid amorphous passivation layer 32. Forming an α-SiC:H layer may include using silane (SiH 4) and methane (CH 4) as process gases, wherein silicon (Si) and carbon (C) from these process gases are activated gas species deposited on the surface 101 to form the amorphous hydrogen-containing silicon carbide (a-SiC:H) layer. Forming an amorphous hydrogen-containing carbon (a-C:H) layer may include using methane (CH 4) as the process gas, wherein carbon (C) and hydrogen (H) are reactive species.The amorphous passivation layer 32 provides back charges to ionized dopant atoms in the third semiconductor region 31 when the pn junction between the third semiconductor region 31 and the second semiconductor region 21 is reverse biased. This is due to a contact potential at a junction between the amorphous passivation layer 32 and the third region 31 and under the influence of a lateral electric field resulting from this reverse poling of the pn junction. The higher the density of states (DOS), the more counter charges can be provided in the amorphous passivation layer 32, because the higher the density of states (DOS), the shorter a shielding length L a, and the shorter the shielding length L a, the higher the ability of the amorphous passivation layer 32 to provide counter charges (mirror charges). In other words, a high state density of the amorphous passivation layer makes it possible to realize the third region 31 with a high doping dose without reducing the blocking voltage strength, wherein the high doping dose increases the robustness of the edge termination structure as explained above.The density of states (DOS) in the amorphous passivation layer 32 may be adjusted by appropriately selecting the process parameters in the deposition process. These process parameters comprise, for example, a flow rate of the at least one process gas, a temperature in the reactor during the deposition process, a frequency value of the HF. Forming an amorphous hydrogen-containing silicon carbide (a-SiC:H) layer with a DOS as given in equation (2) may include an HF PECVD process using silane as a silicon source and methane (CH 4) as a carbon source. A gas flow of these gases may be set to between 50...500 sccm, the RF frequency may be 13.56 MHz, and a temperature may be selected from between 200° C. and 400° C. Optionally, hydrogen or helium may be used as the dilution gas.As stated above, the doping dose S of the third region 31 decreases towards the edge 103, wherein in the middle (at the position x 1) of the edge termination structure 30 in the semiconductor body 100, the doping dose S is higher than the breakdown dose S BR. According to an example, the doping dose S falls below the breakdown dose S BR at a position spaced apart from the first region 11 (i.e. from the beginning x0of the edge termination structure 30 in the semiconductor body 100) by a distance that is between 50% and 60%, in particular between 60% and 70%, of the total width w30of the edge termination structure 30. The maximum of the doping dose S 0 of the third semiconductor region 31 is selected from between 1.5 times and 2 times the breakdown dose S BR. for example. It should be noted that "doping dose" as used herein refers to the electrically active doping dose.As stated above, the doping dose of the third region 31 may be substantially constant between the beginning x 0 and the center x 1 of the edge termination structure 30. However, this is only an example. According to further examples illustrated in FIGS. 5A and 5B, the doping dose S may already decrease between the beginning x 0 and the center x 1 of the edge termination structure 30. Referring to the curve 401 shown in FIG. 5A, this may include that the doping dose S is substantially constant over a certain range starting at the position x 0 and then starts decreasing before the position x 1. However, referring to the curve 402, 403, and 404, a decreasing doping dose between the beginning x 0 and the center x 1 may also include the doping dose decreasing substantially constantly between the beginning x 0 and the center x 1. In any case, an amount of an average normalized slope |m1| of the doping dose between the beginning x0and the center x1is given by where S(x0) denotes the doping dose at the beginning x0and S(x1) denotes the doping dose at the center x1of the edge termination structure 30, and x1-x0(=0.5·w30) is the distance between the beginning x0and the center x1. In any case, the doping dose in the middle x 1 is higher than the breakdown dose S BR.FIG. 5B illustrates blocking labels associated with the doping dose profiles shown in FIG. 5A, where curve 501 is the blocking label associated with the doping dose profile 401 shown in FIG. 5A, curve 502 is the blocking label associated with the doping dose profile 402 shown in FIG. 5A, curve 503 is the blocking label associated with the doping dose profile 403 shown in FIG. 5A, and curve 504 is the blocking label associated with the doping dose profile 404 shown in FIG. 5A. Referring to FIG. 5A, curve 401 has the lowest average slope |m|, and curve 404 has the highest average slope |m| of the four curves illustrated in this figure. The slope of curve 402 is higher than the slope of curve 401 and lower than the slope of curve 403, and the slope of curve 403 is lower than the slope of curve 404. As can be seen from FIG. 5B, the higher the average slope between the beginning x0and the center x1, the wider the associated blocking parabolic curve. That is, in the example illustrated in FIG. 5B, the blocking label 504 associated with the doping dose profile 404 is wider than each of the other blocking labels, and the blocking label 501 associated with the doping dose profile 401 is narrower than each of the other blocking labels. FIG. 5B illustrates the maximum blocking voltage capabilities V BR depending on the respective maximum doping dose, which in this example is the doping dose at the position x 0. As can be seen from FIG. 5B, the maximum reverse withstand voltage V BR also decreases as the average slope increases. That is, there is a correlation between the maximum reverse withstand voltage and the robustness. According to one example, the third region 31 is realized such that the magnitude of the average normalized slope |m| is higher than 0.1 and lower than 0.6, i.e., 0.1<|m1|<0.6.FIG. 6A illustrates doping profiles 601, 602according to two further examples, and FIG. 6B illustrates the associated blocking labels, wherein the blocking label 701 shown in FIG. 6B is associated with the doping dose profile 601 shown in FIG. 6A and the blocking label 702 shown in FIG. 6B is associated with the doping dose profile 602 shown in FIG. 6A. The doping dose profiles 601, 602shown in FIG. 6A differ in the slope of the doping dose profile between the center x 1 and the end x 3 of the edge termination structure. The "end x 3" is the position where the surface portion 33 abuts on either the channel stopper 22 or the edge surface 103. Referring to FIG. 6A, the doping dose profile 601 has a larger amount of slope than the doping dose profile 602. As can be seen from FIG. 6B, the greater slope of the profile 601 leads to a further (broader) blocking parabolic. According to an example, an amount of an average normalized slope |m2| in a region of the edge termination structure 30 that is from the end x 3 to a position x 31 spaced apart from the end x 3 by 0.4·w 30 is between 1.5 and 10, 1.5<|m2|< 10, in particular between 2 and 10, wherein S(x 31) is the (effective) doping dose at the position x 31 and S(x 3) is the (effective) doping dose at the end x 3 of the edge termination structure, which may be zero. In addition, x3-x31 (=0.4·w30) is the distance between the end x3 and the position x31.FIG. 7A illustrates two further examples of doping profiles 801. 802. In these examples, the doping dose decreases abruptly from a starting level S 1 to a respective ending level S 2 1, S 2 2 at a position arranged between the center x 1 and the end x 3 of the edge termination structure. FIG. 7B illustrates the magnitude of a lateral component |E x| of the electric field in the third region 31 when the pn junction between the second region 21 and the third region 31 is reverse biased, wherein the curve 901 shown in FIG. 7B is associated with the profile 801 shown in FIG. 7A and the curve 902 shown in FIG. 7B is associated with the profile 802 shown in FIG. 7A. As can be seen from FIG. 7B, a maximum of the electric field occurs at the position x 4 where the doping dose decreases abruptly, the higher the difference between the start level S 1 and the respective end level S 2 1, S 2 2 is.According to an example, the doping dose profile of the third region 31 has a profile as shown in FIG. 7A, i.e. the doping dose abruptly decreases from a start level S 1 to a finish level S 2 at a position x 4 between the center x 1 and the end of the edge termination structure x 3, wherein a ratio S 2 / S 1 between the finish level S 2 and the start level S 1 is lower than 0.9 and higher than 0.5, i.e. 0.5<S 2 / S 1<0.9.An edge termination structure 30 of the type explained above may be realized in various types of semiconductor devices. Two examples are illustrated below.Referring to FIG. 8, the semiconductor device may be realized as a diode. In this case, the first region 11 forms an emitter region and the second region 21 forms a base region of the diode. The first emitter region 11 is connected to a first electrode 41. The diode also comprises a second emitter region 42 of the same doping type as the second region 21, but which is more highly doped than the second region 21. According to an example, the doping concentration of the second emitter region 42 is in the same range as the doping concentration of the first region 11. According to an example, the first region 11 is p-doped to form a p-emitter of the diode and the second emitter region 42 is n-doped to form an n-emitter of the diode. In this case, the first electrode 41 constitutes an anode, and the second electrode 43 constitutes a cathode of the diode.According to a further example, which is illustrated in FIG. 9A, the semiconductor device is realized as a transistor device. The transistor device comprises a plurality of transistor cells 53, wherein one transistor cell 53 is illustrated in detail in FIG. 9B. Referring to FIG. 9B, each transistor cell 53 includes a portion of the first region 11, wherein this portion of the first region 11 forms a body region of the transistor cell 53. Moreover, each transistor cell 53 comprises a source region 55, wherein the source region 55 is separated from the second region 21 by the body region 11. The second region 21 forms a drift region of the transistor device. The source region 53 and the body region 11 are connected to a first electrode 54 forming a source electrode of the transistor device. Referring to FIG. 9B, the body region 11 may be adjacent to the first surface 101. However, this is only an example. According to a further example (not shown), the source region 55 separates the body region 11 from the first surface 101. In this case, contact plugs (not shown) extend from the source electrode 54 through the source region 55 into the body region 11.Referring to FIGS. 9A and 9B, each transistor cell further comprises a gate electrode 56 dielectrically insulated from the body region 11 by a gate dielectric 57. The gate electrodes 56 of the individual transistor cells 53 are connected to a gate node G, which is schematically illustrated in FIG. 9A.Referring to FIG. 9A, the transistor device further comprises a drain region 51. the drain region 51 abuts the second surface 102 of the semiconductor body 100.The transistor device may be realized as a MOSFET. In this case, the drain region 51 has the same doping type as the drift region 21, but is doped higher. According to a further example, the transistor device is realized as an IGBT. In this case, the drain region 51 (which in this case may also be referred to as collector region) has a doping type complementary to the doping type of the drift region 21. The source region 55 has the same doping type as the drift region 21, but is doped higher. Doping concentrations of the source region 55 and the drain region 51 are, for example, in the range between 1E19cm -3 and 1E21cm -3.FIG. 10 illustrates an example of a method for producing a third region 31 of the type explained above. Referring to FIG. 10, forming the third region 31 may include implanting dopant atoms via the first surface 101 into the semiconductor body 100 using an implantation mask. According to an example, this implantation mask has a first portion 61 with a varying thickness increasing towards the edge surface 103. The thicker the first mask portion 61 at a certain position, the less dopant atoms pass through the first mask portion 61 during the implantation process, so that the doping dose below the first mask portion 61 decreases towards the edge surface 103. The thickness of the first mask portion 61 may continuously increase toward the edge surface 103 or may increase in steps (as represented by 61 1 in FIG. 10 ). A second mask portion 62 above the first region 11 is thick enough to prevent dopant atoms from passing through this second mask portion 62 during the implantation process. The first region 11 may be formed before or after forming the third region 31.

Claims

A semiconductor device, comprising: a semiconductor body (100) having a first main surface (101), an edge surface (103), an inner region (110) and an edge region (120) arranged between the inner region (110) and the edge surface (103); a first semiconductor region (11) of a first doping type arranged in the inner region (110) and a second semiconductor region (21) of a second doping type arranged in the inner region (110) and the edge region (120), wherein a pn junction is formed between the first semiconductor region (11) and the second semiconductor region (21); An edge termination structure (30) comprising: a third semiconductor region (31) of the first doping type arranged in the edge region (120) and adjoining the first semiconductor region (11), wherein a doping dose of the third semiconductor region (31) decreases towards the edge surface (103); a surface portion (33) of the second semiconductor region (21) adjoining the first main surface (101); and an amorphous passivation layer (32) having a resistivity higher than 10 9 Ωcm formed on the first main surface (101) and adjoining the third semiconductor region (31) and the surface portion (33) of the second semiconductor region (21), wherein the edge termination structure (30) in the semiconductor body has a width (w30) (100) in the lateral direction, wherein an electrically active doping dose of the third semiconductor region (31) at a lateral position (x1) spaced apart from the first semiconductor region (11) by 50% of the width (w30) of the edge termination structure (30) is at least Q BR / q, wherein Q BR is the breakdown charge of the semiconductor material of the semiconductor body (100) and q is the elementary charge, and wherein the amorphous passivation layer (32) is realized such that a state density N F in the passivation layer (31) is given by: N F ≥ 1 ε ⋅ ( Q B R E g ) 2, wherein Q BR is the breakdown charge of the semiconductor material of the semiconductor body (100), E g is the band gap of the semiconductor material of the semiconductor body (100), and ε=ε 0 ·ε r is the dielectric constant of the semiconductor material of the semiconductor body (100).The semiconductor device according to claim 1, wherein a width (w33) of the surface portion (33) of the second semiconductor region (21) is less than 1 / 3 of the width (w30) of the edge termination structure (30) in the semiconductor body (100).Semiconductor device according to claim 1 or 2, wherein a maximum electrically active doping dose (S 0) of the third semiconductor region (31) is selected from between 1.05 and 2 times Q BR / q.The semiconductor device according to any of the preceding claims, wherein the width (w30) of the edge termination structure is between 2 times and 3 times, in particular between 2.2 times and 2.8 times, a dimension of the second semiconductor region (21) in a vertical direction y of the semiconductor body (100) in the inner region (110).The semiconductor device according to any one of the preceding claims, further comprising: a channel stopper (22) of the second doping type arranged in the second semiconductor region (21) in the edge region (120), wherein the third semiconductor region (31) and the surface portion (33) of the second semiconductor region (21) are arranged between the first semiconductor region (11) and the channel stopper (22).The semiconductor device according to any of the preceding claims, wherein a doping dose of the first semiconductor region (11) is higher than 10 times a maximum doping dose of the third semiconductor region (31).The semiconductor device according to any one of the preceding claims, wherein an average of an amount of a normalized slope (m) of a doping dose profile of the third semiconductor region (31) in the lateral direction in a portion between the first semiconductor region (11) and the first position (x1) is higher than 0.1 and lower than 0.6.The semiconductor device according to any of the preceding claims, wherein a doping dose of the third semiconductor region (31) abruptly decreases from a first doping dose level (S1) to a second doping dose level (S2) lower than the first doping dose level (S1), wherein a ratio between the second doping dose level (S2) and the first doping dose level (S1) is lower than 0.9 and higher than 0.5.Semiconductor component according to any one of claims 1 to 8, wherein the semiconductor body (100) comprises monocrystalline silicon (Si).Semiconductor device according to any one of claims 1 to 8, wherein the semiconductor body (100) comprises monocrystalline silicon carbide (SiC).The semiconductor device according to any one of the preceding claims, wherein a material of the amorphous passivation layer (31) is selected from the group consisting of: amorphous hydrogen-containing carbon (a-C:H); amorphous silicon (a-Si); amorphous silicon carbide (a-SiC); or amorphous hydrogen-containing silicon carbide (aSiC:H).The semiconductor device according to any one of claims 1 to 11, wherein the semiconductor device is realized as a vertical diode in which the first semiconductor region (11) forms a first emitter region and the second semiconductor region (21) forms a base region, and wherein the semiconductor device further comprises: a second emitter region (42) of the second doping type, wherein the base region (21) is arranged between the first emitter region and the second emitter region (42).The semiconductor device according to any of claims 1 to 11, wherein the semiconductor device is realized as a vertical transistor in which the first semiconductor region (11) forms a body region and the second semiconductor region (21) forms a drift region, the semiconductor device further comprising: at least one source region (55) of the second doping type, the body region being arranged between the source region (55) and the drift region, at least one gate electrode (56) arranged adjacent to the body region and dielectrically insulated from the body region by a gate dielectric (57); and a drain region (51), the drift region being arranged between the drain region (51) and the body region.A semiconductor device according to claim 13, wherein the transistor device is an IGBT and the drain region (51) is of the first doping type.A semiconductor device according to claim 13, wherein the transistor device is a MOSFET and the drain region (51) is of the second doping type.The semiconductor device according to any of the preceding claims, wherein the resistivity of the amorphous passivation layer is lower than 1E14 Ωcm.

Images