Wide bandgap semiconductor electronic device with JBS diode having improved electrical characteristics and method for fabricating same

Abstract

To provide a wide band gap power device comprising a JBS diode having improved electrical characteristics, and a manufacturing method thereof.SOLUTION: In a Junction Barrier Schottky diode 50, a body 55 of silicon carbide having a first conductivity type and a surface 55A is formed by: a drift region 59 and a plurality of first portions 59A delimited by the surface 55A; a second portion 59B; a plurality of first implanted regions 62 of a second conductivity type, extending into the drift region 59 from the surface 55A; and a plurality of metal portions (a first portion 65A, a second portion 66A) arranged on the surface 55A. Each metal portion is in Schottky contact with a respective surface portion of the plurality of surface portions so as to form a first-type Schottky diode 51 and a second-type Schottky diode 52. The first-type Schottky diode and the second-type Schottky diode each have, at equilibrium, a Schottky barrier having a height different from each other.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to wide bandgap semiconductor electronic devices having JBS (junction barrier Schottky) diodes with improved electrical characteristics and corresponding fabrication methods, particularly with reference hereinafter to vertical-conduction electronic power devices. [Background technology]

[0002] As is known, semiconductor materials such as silicon carbide (SiC) and gallium nitride (GaN) with a wide bandgap, e.g., greater than 1.1 eV, a low on-state resistance, a high thermal conductivity, a high operating frequency, and a high saturation velocity of charge carriers, make it possible to obtain electronic devices, e.g., diodes and transistors, with better performance than silicon electronic devices, especially in power applications operating at voltages between 600 V and 1300 V or at specific operating conditions, e.g., at high temperatures.

[0003] In particular, it is known to obtain such electronic devices from wafers made of silicon carbide in one of the polytypes, such as 3C-SiC, 4H-SiC, 6H-SiC, etc., distinguished by the properties mentioned above.

[0004] For example, Figure 1 shows a known silicon carbide JBS diode 1. The JBS diode 1 is generally formed from a number of identical elementary cells (only one shown) arranged in parallel on the same die, each elementary cell including a Schottky diode 2 and a pair of PN diodes 3 connected in parallel with each other.

[0005] The JBS diode 1 is formed in a body 5 of silicon carbide (SiC), the body being bounded by a first surface 5A and a second surface 5B opposite each other along a first axis Z of a Cartesian coordinate system XYZ, and having a substrate 7 and a drift region 9 formed on the substrate 7, for example by epitaxial growth. The substrate 7 is of N-type and forms the second surface 5B of the body 5. The drift region 9 is N-type and has a lower doping level than that of the substrate 7 and forms the first surface 5A of the body 5.

[0006] A cathode metallization region 10 of a conductive material, for example nickel or nickel silicide, extends over the second surface 5B of the body 5 and forms the cathode K of the JSB device 1.

[0007] The JBS diode 1 further includes a plurality of barrier regions 12, only two of which are shown in FIG. 1 and which are substantially contained within the drift region 9. The barrier regions 12 are spaced apart from one another along a second axis Y of the Cartesian coordinate system XYZ, and each is formed by a respective implanted region 13 of P type that extends from the first surface 5A of the body 5 into the drift region 9. The barrier regions 12 further extend along a third axis X of the Cartesian coordinate system XYZ.

[0008] Each of the barrier regions 12 further includes a respective ohmic contact region 14 of a conductive material, such as nickel silicide, that extends over and partially into each of the implanted regions 13 and has an extent along the second axis Y that is less than or equal to the extent of each of the implanted regions 13 in the cross-sectional view of Figure 1. Each of the interfaces between the implanted regions 13 and the drift region 9 forms a PN diode 3.

[0009] The JBS diode 1 further includes an anode metallization region 18 made of a metallic material such as titanium, nickel, or molybdenum, which extends over the first surface 5A of the body 5 and forms the anode A of the JBS diode 1. The portions of the anode metallization region 18 that are in direct electrical contact with the drift region 9 near the injection region 13 form Schottky junctions (i.e., semiconductor-metal junctions), each of which constitutes a respective Schottky diode 2.

[0010] Each of the Schottky junctions has a respective barrier with a height φ in equilibrium, which determines the conduction threshold voltage of each Schottky diode 2. In use, a bias voltage can be applied between the cathode K and anode A of the JBS diode 1 to obtain a reverse bias or a forward bias. Specifically, in forward bias (anode A is at a higher potential than cathode K), a forward voltage is applied that lowers the barrier height φ of the Schottky junction, switching on the Schottky diode 2 and allowing an operating current of, for example, 10 A, to flow between the anode A and cathode K of the JBS diode 1.

[0011] The low barrier height φ and therefore the low conduction threshold voltage of the Schottky diode 2 makes it possible to reduce the value of the forward voltage required for the flow of the operating current and therefore to achieve a low power consumption of the JBS diode 1.

[0012] On the other hand, in reverse bias (i.e. when the cathode K is at a higher potential than the anode A), the JBS diode 1 must ensure that the value of the leakage current between the anode A and the cathode K is as low as possible in the presence of a high bias voltage, for example between 600 V and 1300 V.

[0013] Such a high bias voltage generates a high electric field in the Schottky junction, which increases the leakage current due to, for example, the tunneling effect. This increase is particularly noticeable when the barrier height φ is small. As a result, a small barrier height φ generates a high leakage current, which in turn causes operational problems for the JBS diode 1.

[0014] Furthermore, under reverse bias, each PN diode 3 forms a respective depletion zone with a low concentration of charge carriers, which extends from the interface between the respective injection region 13 and the drift region 9 and mainly inside the drift region 9. The depletion zone contributes to locally reducing the value of the electric field generated by a high bias voltage in the vicinity of the respective Schottky junction.

[0015] Therefore, it is known to design the JBS diode 1 so that the distance along the second axis Y between two adjacent implant regions 13 is as small as possible. A smaller distance between two adjacent implant regions 13 actually results in a larger electric field drop across the Schottky junction, resulting in a lower leakage current. However, a smaller distance between two adjacent implant regions 13 reduces the area of ​​the Schottky diode 2 available for current flow in forward bias, thus increasing the resistance of the JBS diode 1 and thereby degrading its performance. Summary of the Invention [Problem to be solved by the invention]

[0016] The object of the present invention is to overcome the drawbacks of the prior art. [Means for solving the problem]

[0017] According to the present invention there is provided a wide bandgap semiconductor electronic device comprising a JBS diode and a corresponding method of manufacture, as claimed. In order that the invention may be better understood, an embodiment thereof will now be described, purely by way of non-limiting example, with reference to the accompanying drawings, in which: [Brief explanation of the drawings]

[0018] [Figure 1] 1 is a cross-sectional view of a known wide bandgap electronic semiconductor power device. [Figure 2] 1 is a cross-sectional view of one embodiment of the present wide bandgap electronic semiconductor power device with electrical uniformity highlighted. [Figure 2A] 2, with geometric parameters of the electronic device highlighted. [Figure 3] 3 is a graph showing a simulated forward bias current-voltage curve of the elementary cell of the electronic device of FIG. 2 compared to the elementary cell of the known electronic device of FIG. 1; [Figure 4] 2B is a graph showing the behavior of the electric field along the section line 80 of FIG. 2A. [Figure 5] 1 is a cross-sectional view of the present wide bandgap electronic semiconductor power device according to another embodiment. [Figure 6] 1 is a cross-sectional view of the present wide bandgap electronic semiconductor power device according to another embodiment. [Figure 7] 1 is a cross-sectional view of the present wide bandgap electronic semiconductor power device according to another embodiment. [Figure 8] 3A-3C are cross-sectional views of the electronic power device of FIG. 2 at various steps during fabrication. [Figure 9] 3A-3C are cross-sectional views of the electronic power device of FIG. 2 at various steps during fabrication. [Figure 10] 3A-3C are cross-sectional views of the electronic power device of FIG. 2 at various steps during fabrication. [Figure 11] 3A-3C are cross-sectional views of the electronic power device of FIG. 2 at various steps during fabrication. [Figure 12]6A and 6B are cross-sectional views of the electronic device of FIG. 5 at various steps during manufacture. [Figure 13] 6A and 6B are cross-sectional views of the electronic device of FIG. 5 at various steps during manufacture. [Figure 14] 6A and 6B are cross-sectional views of the electronic device of FIG. 5 at various steps during manufacture. DETAILED DESCRIPTION OF THE INVENTION

[0019] The present invention relates to a vertical conduction electronic power device incorporating a plurality of Schottky barrier diodes of different types, each having a Schottky junction with a different barrier height at equilibrium, and in particular to a silicon carbide wide bandgap semiconductor JBS (junction barrier Schottky) diode.

[0020] In particular, Figures 2 and 2A show a JBS diode 50, which in this embodiment includes a plurality of base cells 54 that are identical to one another and connected in parallel within the same die.

[0021] 2, each basic cell 54 includes one or more first-type Schottky diodes 51, one or more second-type Schottky diodes 52, and one or more parallel-arranged PN diodes 53. In particular, in this embodiment, each basic cell 54 is formed by two first-type Schottky diodes 51 and one second-type Schottky diode 52 arranged between two adjacent PN diodes 53.

[0022] JBS diode 50 is formed in a body 55 of silicon carbide (SiC), defined by a first surface 55A and a second surface 55B opposite each other along a first axis Z of a Cartesian coordinate system XYZ, and including a substrate 57 and a drift region 59 grown thereon, e.g., epitaxially. Substrate 57 is N-type, has a doping level such that it has a low resistivity, e.g., between 2 mΩ·cm and 30 mΩ·cm, has a thickness such that it ... 16 at / cm 3 , has a thickness between 5 μm and 15 μm and forms the first surface 55 A of the body 33 .

[0023] The thickness of drift region 59 can be selected during design based on the particular application of JBS diode 50, for example, according to the maximum operating voltage to be applied to JBS diode 50. A cathode metallization region 54 of a conductive material, such as nickel or nickel silicide, extends over second surface 55B of body 55 and forms cathode K of JBS diode 50. JBS diode 50 further includes a plurality of implanted regions 62 contained within drift region 59.

[0024] Implanted regions 62 are P-type and have a doping level higher than that of drift region 59, each extending a depth from first surface 55A of body 55 into drift region 59 along first axis Z, and each having a width along second axis Y of Cartesian coordinate system XYZ, for example between 1 μm and 6 μm.

[0025] Furthermore, the implanted regions 62 are spaced apart from one another along the second axis Y at a distance, for example, between 2 μm and 5 μm, S YIn a plan view (not illustrated here), the implanted regions 62 may have the shape of strips extending longitudinally along a third axis X of a Cartesian coordinate system XYZ, or may have any other shape; for example, they may form regular or irregular geometric figures such as squares, rectangles, hexagons, or circles.

[0026] 2 may be part of a single region having a more complex shape that can be seen as separate regions in the illustrated cross section. The interfaces between implant region 62 and drift region 59 form PN junctions, each forming a respective PN diode 53.

[0027] JBS diode 50 may further include a plurality of ohmic contact regions 63 of a conductive material, each of which extends over and has a shape generally corresponding to that of a respective implant region 62. Each ohmic contact region 63 is in direct electrical contact with a respective implant region 62 and extends along second axis Y across a width equal to or less than the width of a respective implant region 62.

[0028] Specifically, in this embodiment, the ohmic contact regions 63 are made of nickel silicide, each of which extends partially within a respective implant region 62 and has a width that is smaller than that of the respective implant region 62.

[0029] JBS diode 50 further includes a plurality of anode metallization regions, one for each type of Schottky diode (as described in more detail below), that extend across first surface 55A of body 55. In this embodiment, JBS diode 50 includes first anode metallization region 65 and second anode metallization region 66.

[0030] The first anode metallization regions 65 also have shapes generally corresponding to those of the respective implanted regions 62, but as will be explained in more detail below, they have a greater width, and furthermore, as described above for the implanted regions 62, they may form part of a first single anode metallization region 65 having a more complex shape. They are in any case electrically connected in parallel, and therefore they can be considered as a single region.

[0031] The first anode metallization regions 65 are made of a first metallic material, for example molybdenum, and each has at least one first portion 65A (FIG. 2A) that extends over and is in direct electrical contact with a respective first portion 59A of the drift region 59 on its side adjacent to a respective implantation region 62. In this embodiment, each of the first anode metallization regions 65 further has a respective second region 65B that extends over and is in direct electrical contact with a respective implantation region 62 and a respective ohmic contact 63.

[0032] The first portions 65A of the first anode metallization regions 65 each have a width W1 and are spaced a distance W2 from the first portions 65A of the adjacent first anode metallization regions 65.

[0033] The first portions 65A of the first anode metallization regions 65 each form a Schottky junction with a respective first portion 59A of the drift region 59 to form a respective first-type Schottky diode 51 (FIG. 2). The junctions of the first-type Schottky diodes 51 each have a respective Schottky barrier, which in equilibrium has a height φ1, e.g., about 0.5 eV, that determines the first conduction threshold voltage of each first-type Schottky diode 51.

[0034] The second anode metallization region 66 is made of a second metal material, such as titanium or nickel, and extends over the first surface 55A of the body 55 and over the first anode metallization region 65. As a result (FIG. 2A), a portion 66A of the second metallization region 66 is in direct electrical contact (Schottky contact) with the second portion 59B of the drift region 59 over a width equal to the distance W2 between two first portions 65A of adjacent first anode metallization regions 65.

[0035] In other words, the second portion 59B is located at a greater distance from the respective same implanted region 62 relative to the first portion 59A.

[0036] Thus, the portion 66A of the second metallization region 66 forms, together with the second portion 59B of the drift region 59, a Schottky junction constituting each of the second-type Schottky diodes 52 (FIG. 2). The junctions of the second-type Schottky diodes 52 each have a respective Schottky barrier, which, at equilibrium, has a height φ2, e.g., 1.2 eV, greater than the barrier height φ1 that determines the second conduction threshold voltage of each of the second-type Schottky diodes 52, which is higher than the first conduction threshold voltage.

[0037] The first and second anode metallization regions 65 , 66 further form the anode A of the JBS diode 50 .

[0038] In use, in forward bias (higher voltage applied to anode A than to cathode K of JBS diode 50), a low voltage allows a high operating current, e.g., 10 A, to flow from anode A to cathode K of JBS diode 50. Indeed, first-type Schottky diode 51 allows current flow at a lower forward voltage than second-type Schottky diode 52 because the first conduction threshold voltage is lower than the second conduction threshold voltage.

[0039] As a result, overall, the JBS diode 50 has a lower conduction threshold than the JBS diode 1 illustrated in FIG. 1, as shown in simulations performed by the inventors and illustrated in FIG. 3, where the solid curve represents a plot of current I as a function of voltage V for the JBS diode 50 and the dotted line represents a similar plot for the known JBS diode 1 in terms of a basic cell. As can be seen, with the JBS diode 50, it is possible to obtain the same current flow with a lower forward bias voltage applied between the anode A and the cathode K. Therefore, the JBS diode 50 has lower power consumption in forward bias.

[0040] In reverse bias, a high voltage, for example between 600 V and 1300 V, can be applied to JBS diode 50 between anode A and cathode K. In reverse bias, the interfaces between injection region 62 and drift region 59 each form a respective depletion region having a low concentration of charge carriers, which locally reduces the electric field generated by the high voltage, particularly near first surface 55A of body 55. Indeed, in this embodiment, the depletion region extends primarily into drift region 59 because the doping level of drift region 59 is lower than that of injection region 62.

[0041] In particular, first portions 59A of drift regions 59 each extend a smaller distance from their respective PN junctions than second portions 59B of drift regions 59. Therefore, first portions 59A are more susceptible to the influence of their respective depletion zones. As a result, first portions 59A are locally exposed to a lower average electric field near first surface 55A of body 55 than the average electric field in second portions 59B of drift regions 59.

[0042] In other words, the electric field in the drift region 59 has a minimum value near the first surface 55A of the body 55, at the interface between the first portion 59A of the drift region 59 and the implanted region 62, and increases with increasing distance from the implanted region 62 along the second axis Y until it reaches a maximum value at the center of the second portion 59B of the drift region 59, i.e., at the maximum distance from the implanted region 62.

[0043] The above electrical behavior is illustrated in our simulation of FIG. 4, which shows the behavior of the electric field E along line 80 in FIG. 2A, which extends in a direction parallel to first surface 55A from the interface between implant region 62 and first portion 59A of drift region 59 to the midpoint of second portion 59B of drift region 59.

[0044] The first type Schottky diode 51 extends into the first portion 59A of the drift region 59 and is therefore exposed to a low average electric field, so that the first type Schottky diode 51 can ensure a low leakage current despite a high voltage applied in reverse bias.

[0045] The second-type Schottky diode 52 extends into the second portion 59B of the drift region 59 and is therefore exposed to a higher average electric field. However, the second-type Schottky diode 52 has a barrier height φ2 that is greater than the barrier height φ1, thereby ensuring a low leakage current even in the presence of a higher electric field. Therefore, overall, the JBS diode 50 can achieve a low leakage current under reverse bias.

[0046] At the design stage, the distance S between two adjacent implanted regions 62 is YBy keeping W1 constant, the width W1 and distance W2 can be adjusted based on the characteristics of the JBS diode 50 required by a particular application. For example, if a low forward bias voltage is desired, the width W1 can be increased and the distance W2 can be decreased. On the other hand, if a low leakage current in reverse bias is desired, the width W1 can be decreased and the distance W2 can be increased.

[0047] Furthermore, the distance S between two adjacent implanted regions 62 is increased, for example, to the point where the depletion regions of two adjacent PN diodes 53 have almost no overlap. Y It is possible to configure the JBS diode 50 so that .DELTA..times ...

[0048] The fact that two adjacent depletion regions 62 have almost no overlap means that the distance S Y However, the larger barrier height of the second type Schottky diode 52 allows the larger average electric field to be compensated for, ensuring a low leakage current. Y allows the JBS diode 50 to have a lower resistance to current flow between the anode and cathode, thus lowering power consumption in forward bias.

[0049] distance S Y If W is actually large, the area of ​​the first type Schottky diode 51 and the second type Schottky diode 52 available for current flow also depends on the width W and the distance W, respectively, and is smaller than the area of ​​the first type Schottky diode 51 and the second type Schottky diode 52. Y Therefore, overall, the JBS diode 50 can have a larger area available for current passage and therefore a lower resistance under forward bias.

[0050] In other words, the JBS diode 50 is capable of providing both low power consumption in forward bias and low leakage current in reverse bias.

[0051] FIG. 5 shows a JBS diode 100 according to another embodiment.

[0052] JBS diode 100 has a similar structure to JBS diode 50, and therefore common elements are numbered identically.

[0053] Specifically, JBS diode 100 includes a first type Schottky diode 101, a second type Schottky diode 102, and a PN diode 53, and is formed within a body 55. Body 55 also includes a substrate 57, a drift region 59, an implantation region 62, and an ohmic contact region 63. Additionally, cathode metallization region 54 extends onto a second surface 55B of body 55.

[0054] JBS diode 100 further includes a plurality of N-type implanted anode regions 105 extending from first surface 55A of body 55 into drift region 59 beside each implanted region 62. Implanted anode regions 105 have a doping level higher than that of drift region 59, e.g., 1×10 17 at / cm 3 It has a higher concentration of dopant atoms than the first one and has a small depth, for example between 10 and 100 nm.

[0055] Specifically, the implanted anode regions 105 each extend into the drift region 59 adjacent to a respective implanted region 62, each having a width W1 and spaced a distance W2 from a respective adjacent implanted anode region 105, as described above with respect to the JBS diode 50 for the first portion 65A of the first anode metallization region 65.

[0056] Thus, implanted anode region 105 constitutes a surface portion of drift region 59 similar to first portion 59A of JBS diode 50 and has a higher doping level than first portion 59A.

[0057] An anode metallization region 115 of, for example, molybdenum, titanium, or nickel extends over the first surface 55A of the body 55 and over the ohmic contact region 63.

[0058] The first portions 115A of the anode metallization regions 115, which are in direct electrical contact (Schottky contact) with the implanted anode regions 105, each form a respective first-type Schottky diode 101, each of which has a Schottky barrier of height φ1 in the equilibrium state.

[0059] The second portions 115B of the anode metallization regions 115, which are in direct electrical contact with the second portions 59B of the drift region 59, each having a width equal to the distance W2 between two adjacent implanted anode regions 105, each form a respective second-type Schottky diode 102, each having a Schottky barrier with a height φ2 greater than that of the first-type Schottky diode 101 at equilibrium.

[0060] In this embodiment, the difference in barrier height between Schottky diode 101 and second type Schottky diode 102 depends on the difference between the doping level of implanted anode region 105 and the doping level of drift region 59. This difference can be adjusted according to the specific application at the design stage.

[0061] As discussed above with respect to JBS diode 50, the Schottky junction of first-type Schottky diode 101 has a smaller barrier height and is exposed to a lower electric field than the Schottky junction of second-type Schottky diode 102. Therefore, overall, JBS diode 100 has low power consumption, low leakage current, and is highly adaptable to different operating conditions.

[0062] 6 shows a further embodiment of the present JBS diode, JBS diode 200. JBS diode 200 has a similar general structure to JBS diode 50, and therefore common elements are numbered the same.

[0063] Specifically, JBS diode 200 includes a first type Schottky diode 201, a second type Schottky diode 202, and a PN diode, and is formed in a body 55 having a substrate 57. The substrate again includes a drift region 59, an implanted region 62, an ohmic contact region 63, and an implanted anode region 105. Additionally, cathode metallization region 54 extends onto a second surface 55B of body 55.

[0064] Additionally, JBS diode 200 includes first anode metallization region 215 and second anode metallization region 216, which extend on first surface 55A of body 55. First anode metallization region 215 has a shape similar to that of first anode metallization region 65 described with reference to JBS diode 50 of FIG. 2 and is made of a first metal material, such as molybdenum.

[0065] In particular, each first anode metallization region 215 has a plurality of portions, including at least one first portion 215A having a width W1, and is disposed over and in direct electrical contact (Schottky contact) with a respective injector anode region 105. In this embodiment, each first anode metallization region 215 also has a second portion 215B, which extends over and is in direct electrical contact with a respective injector region 62 and a respective ohmic contact region 63.

[0066] A first portion 215A of the first anode metallization region 215 in Schottky contact with each injecting anode region 105 forms a first type of Schottky diode 201, each of which has a Schottky barrier height φ1 in the equilibrium state. A second anode metallization region 216 is made of a second metallic material, such as titanium or nickel, and extends over the first surface 55A of the body 55 and over the first anode metallization region 215.

[0067] Thus, a portion 216A of the second anode metallization region 216 is in direct electrical contact (Schottky contact) with a second portion 59B of the drift region 59 that is disposed between two adjacent injector anode regions 105, the second portion 59B having a width equal to the distance W2. The portion 216A of the second anode metallization region 216, together with the second portion 59B of the drift region 59, form second-type Schottky diodes 202, each of which has a Schottky junction that, in equilibrium, has a barrier with a height φ2 that is greater than the barrier with height φ1 of the first-type Schottky diode 201.

[0068] In this embodiment, the barrier heights of the first type Schottky diode 201 and the second type Schottky diode 202 can be adjusted both by modifying the difference between the doping level of the injecting anode region 105 and the doping level of the drift region 59, and by modifying the difference in barrier height induced by the first and second metal materials forming the first and second anode metallization regions 215, respectively.

[0069] Thus, JBS diode 200 has the same advantages as those previously described for JBS diodes 50 and 100, but has greater adaptability to different operating conditions, which can be adjusted at the design stage.

[0070] 7 shows a further embodiment of the present JBS diode, JBS diode 250. JBS diode 250 has a structure similar to that of JBS diode 200, and therefore common elements are numbered the same.

[0071] JBS diode 250 includes a first type Schottky diode 251, a second type Schottky diode 252, and a third type Schottky diode 253, which are arranged in parallel with PN diode 53. Specifically, in this embodiment, portions 265A of first anode metallization regions 265 (similar to first portions 215A of first anode metallization regions 215 of JBS diode 200) each extend over first portions 105A of respective injector anode regions 105 adjacent to respective injector regions 62, over a width smaller than width W1 of injector anode regions 105.

[0072] A second anode metallization region 266 (which is similar to the second anode metallization region 216 of the JBS diode 200) extends over the first surface 55A of the body 55. Specifically, a first portion 266A of the first anode metallization region 266 is in direct electrical contact (Schottky contact) with a second portion 59B of the drift region 59, and a second portion 266B of the second anode metallization region 266 is in direct electrical contact (Schottky contact) with a second portion 105B of the injecting anode region 105 adjacent to the first portion 105A.

[0073] Thus, each second portion 105B of the injector anode regions 105 is disposed between a respective first portion 105A of the injector anode regions 105 and a respective second portion 59B of the drift region 59. Furthermore, each second portion 105B of the injector anode regions 105 is at a greater distance from the same respective injector region 62 than each first portion 105A, and at a lesser distance from the same respective injector region 62 than each second portion 59B.

[0074] That is, the first portion 265A of the first anode metallization region 265, which is in direct electrical contact with the first portion 105A of the injection anode region 105, forms a Schottky junction that has a barrier of height φ1 in the equilibrium state and forms a first type Schottky diode 251.

[0075] A first portion 266A of the second anode metallization region 266, which is in direct electrical contact with a second portion 59B of the drift region 59, forms a Schottky junction that, in equilibrium, has a barrier height φ2 greater than height φ1 and forms a second type of Schottky diode 252.

[0076] A second portion 266B of the second anode metallization region 266, which is in direct electrical contact with the second portion 105B of the injecting anode region 105, forms a Schottky junction that, in an equilibrium state, has a barrier height φ3 that is greater than height φ1 and less than height φ2 and forms a third type Schottky diode 253.

[0077] Therefore, the third-type Schottky diode 253 has a lower conduction threshold voltage than that of the second-type Schottky diode 252. Overall, therefore, the JBS diode 250 has a low conduction threshold and, consequently, low power consumption.

[0078] Furthermore, the third type Schottky diode 253 is located in the second portion 105B of the injector anode region 105, which is located at a distance from an adjacent injector region 62 that is less than the distance of the second portion 59B from the same adjacent injector region 62, and therefore is exposed to a lower average electric field in reverse bias than the second portion 59B of the drift region 59.

[0079] As a result, the first type Schottky diode 251, the second type Schottky diode 252, and the third type Schottky diode 253 can also guarantee a low leakage current for the JBS diode 250 as a whole.

[0080] Hereinafter, the steps for manufacturing the JBS diodes 50, 100, 200, and 250 will be described.

[0081] JBS diode 50 can be fabricated from a silicon carbide (SiC) wafer 300, illustrated in cross section in FIG. 8, having a first surface 300A and a second surface 300B and having undergone a first known processing step.

[0082] In particular, in FIG. 8, wafer 300 has already been processed to form body 55 (whose first and second surfaces 55A, 55B correspond to first surface 300A and second surface 300B, respectively) having substrate 57, drift region 59, implantation region 62, and ohmic contact region 63.

[0083] Next, as shown in FIG. 9, a first metallization layer 305 of a first metal material, for example molybdenum, is deposited on the first surface 300A of the wafer 300 and patterned (FIG. 10) via known lithography steps to form a first anode metallization region 65.

[0084] A second metallization layer 306 of a second metallic material, for example titanium or nickel, is then deposited on the first surface 300A of the wafer 300 to form a second anode metallization region 66, as shown in FIG.

[0085] Known processing steps, such as thinning the substrate 57, form cathode metallization regions on the second surface 300B of the wafer 300, and the wafer 300 is diced and packaged to obtain the JBS diodes 50.

[0086] The JBS diode 100 can be fabricated from a silicon carbide wafer 400, illustrated in Figure 12. The wafer 400 is similar to the wafer 300 illustrated in Figure 8, and therefore, like elements are numbered the same.

[0087] Specifically, wafer 400 has a first surface 400A and a second surface 400B, and has been processed to form substrate 57, drift layer 59, and implanted regions 62. A hard mask 401 having a plurality of windows 402 is then provided on first surface 400A of wafer 400, as shown in Figure 13. Windows 402 are positioned over implanted regions 62 and over portions of the drift region adjacent to each implanted region 62 where implanted anode regions 105 are intended to be formed.

[0088] The window 402 is used for ion implantation, as schematically represented by arrow 405, and to form a doped region 410, here of N type, intended to form the implanted ion region 105. For example, 11 and 10 13 at / cm 2 Especially between 10 and 12 at / cm 2 This dose is sufficient to increase the doping level of each portion of the drift region 59 without jeopardizing the doping of the implanted region 62.

[0089] Alternatively, the hard mask 401 can have a portion (not shown) that also extends over the implantation region 62 so that the window 402 is located only over the portion of the drift region 59 where the implantation anode region 105 is intended to be formed.

[0090] In a known manner, not illustrated, hard mask 401 is then removed and wafer 400 is exposed to annealing, e.g., at a temperature greater than 1500° C., to activate doped regions 410 to form implanted anode regions 105 (FIG. 14).

[0091] Thereafter, as shown in FIG. 14, ohmic contact regions 63 are formed in a known manner and a first metallization layer 406 of a second metal material, for example titanium or nickel, is deposited on the first surface 400A of the wafer 400 to form a second anode metallization region 115.

[0092] Known processing steps such as thinning the substrate 57 follow to form cathode metallization regions on the second surface 400B of the substrate 400, and the wafer 400 is diced and packaged to obtain the JBS 100.

[0093] Those skilled in the art will appreciate that JBS diodes 200 and 250 can be obtained from wafer 400 of FIG. 12 through steps similar to those already described with reference to FIGS. 9-14 for JBS diodes 50 and 100, and a detailed description thereof will be omitted.

[0094] Finally, it will be apparent that various modifications and variations can be made to the wide bandgap semiconductor electronic device and the method of manufacture described and illustrated above without departing from the scope of the appended claims. For example, the conductivity types P and N can be reversed and the above can be applied to MPS (merged PiN Schottky) diodes.

Claims

1. In vertical conduction electronic power devices (50; 100; 200; 250), a body (55) of a wide bandgap semiconductor having a first conductivity type and a surface (55A), the body (55) having a drift region (59) and a plurality of surface portions (59A, 59B; 105, 105A; 105B) defined by the surface; a plurality of first implanted regions (62) of a second conductivity type extending from the surface into the drift region; a plurality of metal portions (65A, 66A; 115A, 115B; 215A, 216A; 265A, 266A, 266B) disposed on the surface (55A), each metal portion making Schottky contact with a respective one of the plurality of surface portions to form a plurality of Schottky diodes (51, 52; 101, 102; 201, 201; 251, 252, 253) including a first Schottky diode and a second Schottky diode; the first Schottky diode has a Schottky barrier having a different height than the second Schottky diode at equilibrium; a first surface portion of the plurality of surface portions (59A; 105; 105A) of the first Schottky diode (51; 101; 201; 251) having the first conductivity type and a first doping level, and a second surface portion (59B) of the plurality of surface portions of the second Schottky diode (52; 102; 202; 252) having the first conductivity type and a second doping level different from the first doping level.

2. The first Schottky diode (51; 101; 201; 251) has a first metal portion (65A; 115A; 215A; 265A) of the plurality of metal portions and the first surface portion of the plurality of surface portions (59A; 105; 105A), the first metal portion being in Schottky contact with the first surface portion, and having a first height (φ １ 2. The power device of claim 1, wherein the first Schottky diode has a first metal portion (66A; 115B; 216A; 266A) of the plurality of metal portions and a second surface portion (59B) of the plurality of surface portions, the second metal portion being in Schottky contact with the second surface portion and forming, in an equilibrium state, a Schottky barrier having a second height (φ2), the first height being smaller than the second height.

3. 3. A power device according to claim 2, wherein the first surface portions (59A; 105; 105A) are each located at a first distance from a respective first implanted region (62), and the second surface portions (59B) are each located at a second distance from the same respective first implanted region between two adjacent first surface portions, the first distance being smaller than the second distance.

4. 4. A power device according to claim 2 or 3, wherein the first metal portion (65A; 115A; 215A; 265A) is made of a first metal material and the second metal portion (66A; 115B; 216A; 266A) is made of a second metal material, the drift region and the first and second surface portions have the first conductivity type, the first metal material being different from the second metal material and / or the first and second surface portions have different doping levels.

5. 5. A power device according to claim 1, wherein the drift region and the first and second surface portions have the same doping level, and the first metal material is different from the second metal material.

6. 5. The power device of claim 4, wherein the first surface portion has a first doping level and the second surface portion has a second doping level lower than the first doping level, and the first metallic material is different from the second metallic material.

7. 7. A power device according to claim 5 or 6, wherein each first surface portion is adjacent to a respective first implant region and is located between the respective first implant region and a respective second implant region, the device further comprising a plurality of first metal regions (65; 215) and second metal regions (66, 216), each first metal region having an ohmic contact portion (65B; 215B) above a respective first implant region and a Schottky contact portion (65A; 215A) forming a respective first metal portion in contact with a respective first surface portion (59A), and the second metal region extending over the first metal region and forming a second metal portion in contact with the second surface portion (59B).

8. The semiconductor device further includes a third Schottky diode (253) including a third metal portion (266B) of the plurality of metal portions and a third surface portion (105B) of the plurality of surface portions, the third metal portion being in Schottky contact with the third surface portion (105B) and having a third height (φ ３ ), the third surface portions have the first doping level and are each disposed between a respective first surface portion (105A) and a respective second surface portion (59B), and the third height (φ ３ ) is the first height (φ １ ) and the second height (φ ２ 7. The power device of claim 6, wherein the .lambda.

9. 10. The power device of claim 8, wherein each first surface portion is adjacent to a respective first implanted region, and the power device further comprises a plurality of first metal regions (265) and second metal regions (266), each first metal region having an ohmic contact portion (265B) over a respective first implanted region and a Schottky contact portion (265A) forming a respective first metal portion in contact with a respective first surface portion, and the second metal region extending over the first metal region and forming the second metal portion (266A) in contact with the second surface portion (59B) and a third metal portion (266B) in contact with the third surface portion (105B).

10. 5. The power device of claim 4, wherein the first surface portion (105) has a first doping level and the second surface portion (59B) has a second doping level lower than the first doping level, and the first metal material is the same as the second metal material.

11. 11. The power device of claim 10, wherein each first surface portion is adjacent to a respective first implanted region (62) and is disposed between the respective first implanted region and a respective second surface portion, the first surface portion being formed by a second implanted region (105), and the first and second metal portions (115A, 115B) form a single metal region (115) extending over and in direct electrical contact with the second implanted region and the second surface portion.

12. 12. A power device according to any one of claims 1 to 11, further comprising a plurality of conductive regions (63), each of which is in ohmic contact with a respective implanted region (62) and with a respective metal portion.

13. 1. A method of fabricating a vertical conduction electronic power device from a wide bandgap semiconductor wafer (300; 400) having a first conductivity type and a surface (300A; 400A) including a drift region (59) and a plurality of first implanted regions (62), the plurality of first implanted regions having a second conductivity type and extending from the surface into the drift region, the method comprising: forming a plurality of metal portions (65A, 66A; 115A, 115B; 215A, 216Z, 266A, 266B) on the surface, each metal portion being in Schottky contact with the drift region at a plurality of surface portions (59A, 59B; 105) of the drift region, each surface portion being partitioned by the surface to form a plurality of Schottky diodes (51, 52; 101, 102; 201, 202; 251, 252, 253) including a first Schottky diode and a second Schottky diode; The first Schottky diode has a Schottky barrier having a different height than the second Schottky diode in an equilibrium state. a first surface portion of the plurality of surface portions (59A; 105; 105A) of the first Schottky diode (51; 101; 201; 251) having the first conductivity type and a first doping level, and a second surface portion (59B) of the plurality of surface portions of the second Schottky diode (52; 102; 202; 252) having the first conductivity type and a second doping level different from the first doping level.

14. A method for manufacturing a vertical conduction electronic power device from a wafer (300; 400) of wide bandgap semiconductor having a first conductivity type and a surface (300A; 400A), the wafer including a drift region (59) and a plurality of first implanted regions (62), the plurality of first implanted regions having a second conductivity type and extending from the surface into the drift region, comprising: forming a plurality of metal portions (65A, 66A; 115A, 115B; 215A, 216Z, 266A, 266B) on the surface, each metal portion being in Schottky contact with the drift region at a plurality of surface portions (59A, 59B; 105) of the drift region, each surface portion being partitioned by the surface to form a plurality of Schottky diodes (51, 52; 101, 102; 201, 202; 251, 252, 253) including a first Schottky diode and a second Schottky diode; the first Schottky diode has a Schottky barrier having a different height than the second Schottky diode in an equilibrium state; The method, wherein forming the plurality of metal portions includes depositing a layer of metal material (406), the method further including introducing a dopant species of a first conductivity type into the surface portions.

15. A method for manufacturing a vertical conduction electronic power device from a wafer (300; 400) of wide bandgap semiconductor having a first conductivity type and a surface (300A; 400A), the wafer including a drift region (59) and a plurality of first implanted regions (62), the plurality of first implanted regions having a second conductivity type and extending from the surface into the drift region, comprising: forming a plurality of metal portions (65A, 66A; 115A, 115B; 215A, 216Z, 266A, 266B) on the surface, each metal portion being in Schottky contact with the drift region at a plurality of surface portions (59A, 59B; 105) of the drift region, each surface portion being partitioned by the surface to form a plurality of Schottky diodes (51, 52; 101, 102; 201, 202; 251, 252, 253) including a first Schottky diode and a second Schottky diode; the first Schottky diode has a Schottky barrier having a different height than the second Schottky diode in an equilibrium state; forming a plurality of metal portions; depositing a first layer (305) of a first metallic material; patterning the first layer to form a first metal portion (65A; 115A; 215A; 265A) in Schottky contact with a first surface portion (59A; 105; 105A) of the plurality of surface portions; and depositing a second layer (306) of a second metallic material; A method that includes:

16. A method for fabricating a vertical conduction electronic power device from a wafer (300; 400) of wide bandgap semiconductor having a first conductivity type and a surface (300A; 400A), the wafer including a drift region (59) and a plurality of first implanted regions (62), the plurality of first implanted regions having a second conductivity type and extending from the surface into the drift region, comprising: forming a plurality of metal portions (65A, 66A; 115A, 115B; 215A, 216Z, 266A, 266B) on the surface, each metal portion being in Schottky contact with the drift region at a plurality of surface portions (59A, 59B; 105) of the drift region, each surface portion being partitioned by the surface to form a plurality of Schottky diodes (51, 52; 101, 102; 201, 202; 251, 252, 253) including a first Schottky diode and a second Schottky diode; the first Schottky diode has a Schottky barrier having a different height than the second Schottky diode in an equilibrium state; the drift region having a first conductivity type and a first doping level, and the method further comprising selectively implanting ions of the first conductivity type into the drift region at a surface area around the implanted region to form doped regions (410), each having a second doping level higher than the first doping level. A method that includes:

Images