SILICON CARBIDE DEVICE WITH SCHOTTKY CONTACT

The silicon carbide device design addresses rapid parameter degradation by integrating Schottky contacts and isolation regions to improve long-term stability and reliability through reduced bipolar degradation and optimized channel density.

DE102019130376B4Active Publication Date: 2024-11-07INFINEON TECHNOLOGIES AG
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
DE102019130376
Authority / Receiving Office
DE · DE
Patent Type
Patents
Current Assignee / Owner
Priority Date
2019-01-04
Filing Date
2019-11-11
Publication Date
2024-11-07
Estimated Expiration
2039-11-11

AI Technical Summary

Technical Problem

Silicon carbide (SiC) devices suffer from rapid degradation of electrical parameters and a need for improved long-term stability.

Method used

A silicon carbide device design incorporating a source region, cathode region, and separation regions with a strip-shaped gate structure, forming Schottky contacts and ohmic contacts, which includes isolation and shielding to reduce electric fields and leakage current, enhancing long-term reliability and short-circuit robustness.

Benefits of technology

The design improves the long-term stability and reduces bipolar degradation by suppressing bipolar current across pn junctions, maintaining high channel density, and optimizing Schottky contact area, thereby enhancing the device's operational reliability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

Silicon carbide device (500), comprising: a silicon carbide body (100) having a source region (110) of a first conductivity type, a cathode region (160) of the first conductivity type and separation regions (140) of a second conductivity type, a strip-shaped gate structure (150) extending along a first direction (291) and directly bordering the source area (110) and the separation areas (140), and a first load electrode (310), wherein, along the first direction (291), the cathode region (160) is located between two of the separation regions (140) and at least one of the separation regions (140) is located between the cathode region (160) and the source region (110), and wherein the source region (110) and the first load electrode (310) form an ohmic contact (OC) and the first load electrode (310) and the cathode region (160) form a Schottky contact (SC), wherein the gate structure (150) extends from a first surface (101) of the silicon carbide body (100) into the silicon carbide body (100), wherein the source region (110) directly borders a first gate sidewall (151) of the gate structure (150) and wherein a shielding region (170) of the second conductivity type directly opposite the source region (110) borders a second gate sidewall (152) of the gate structure (150) opposite the first gate sidewall (151), wherein the silicon carbide body (100) has a body region (120) of the second conductivity type and a drift structure (130), wherein the body region (120) and the drift structure (130) form a first pn junction (pn1) and wherein the first load electrode (310) and the body region (120) form an ohmic contact (OC), wherein a first vertical extent (v1) of the discontinuity regions (140) is greater than a maximum vertical distance (v2) between the first pn transition (pn1) and the first surface (101).
Need to check novelty before this filing date? Find Prior Art

Description

TECHNICAL FIELD

[0001] The present disclosure relates to silicon carbide devices, particularly to silicon carbide devices having transistor cells and Schottky contacts. BACKGROUND

[0002] Power semiconductor devices are typically used as switches and rectifiers in electrical circuits for transforming electrical energy, for example, in DC / AC converters, AC / AC converters, or AC / DC converters, and in electrical circuits driving heavy inductive loads, such as motor drive circuits. The dielectric breakdown field strength of silicon carbide (SiC) is high compared to silicon. Therefore, SiC devices can be significantly thinner than equivalent silicon devices with the same nominal blocking voltage. On the other hand, electrical parameters of silicon carbide devices tend to degrade at a higher rate than is typically the case for equivalent silicon devices.

[0003] The publication US 2005 / 0 285 188 A1 describes the integration of a Schottky diode into an LDMOS (laterally diffused metal-oxide semiconductor) transistor with a central drift zone region. The Schottky diode can be inserted into the lateral center of the drift zone region. Alternatively, the Schottky diode is inserted into a well formed in the periphery of the LDMOS transistor, into which the source region of the LDMOS transistor extends.

[0004] The documents US 2014 / 0 175 508 A1 and JP 2018 098 447 A each show n-channel SiC MOSFETs with a lightly n-doped drift zone and with strip-shaped transistor cells with a transistor channel on both sides. Strip-shaped trench gate electrodes of the transistor cells extend from a front side towards the lightly n-doped drift zone. Between two adjacent transistor cells, a strip-shaped Schottky contact is formed between the front side electrode and the lightly n-doped drift zone. Between a Schottky contact and two adjacent n+ doped source regions of the transistor cells, a p+ doped body contact region extends in a vertical direction to a p-doped body region of the transistor cell and separates the Schottky contact in a lateral direction transverse to the strip direction from an n+ doped source region of the transistor cell.The document JP 2018 098 447 A also shows p+ doped Zener regions between the front side and the n+ doped source regions.

[0005] From JP 2005 108 926 A, a SiC MOSFET with strip-shaped planar gate electrodes and n-doped Schottky regions and n+ doped source regions arranged alternately along the strip direction is known. Along the strip direction, p-well sections extending to the front side separate the n-doped Schottky regions from neighboring n+ doped source regions. The n-doped Schottky regions either end within the p-wells and are only laterally connected, or extend through the p-wells to an n-doped drift region below.

[0006] The publication US 2018 / 0 182 885 A1 describes a SiC MOSFET with strip-shaped trench gate electrodes with a transistor channel on both sides and n-doped regions of Schottky diodes and n+ doped source regions of transistor cells arranged alternately along the strip direction. Along the strip direction, p+ doped contact regions 18 separate the n-doped regions of Schottky diodes from the n+ doped source regions 17.

[0007] There is a need to improve long-term stability parameters of silicon carbide devices. SUMMARY

[0008] This object is achieved with the silicon carbide device according to claim 1. The dependent claims relate to advantageous embodiments. A silicon carbide device comprises a silicon carbide body. The silicon carbide body contains a source region of a first conductivity type, a cathode region of the first conductivity type, and separation regions of a second conductivity type. A strip-shaped gate structure extends along a first direction and directly borders the source region and the separation regions. Along the first direction, the cathode region is located between two of the separation regions, and at least one of the separation regions is located between the cathode region and the source region. A first load electrode and the source region form an ohmic contact. The first load electrode and the cathode region form a Schottky contact. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The accompanying drawings are included to provide a further understanding of the embodiments and are incorporated in and constitute a part of the disclosure of this specification. The drawings illustrate embodiments of a silicon carbide device and, together with the description, serve to explain principles of the embodiments. Additional embodiments are described in the following detailed description and claims. Fig. 1A - 1B illustrate schematic top and vertical cross-sectional views of a portion of a silicon carbide device with a Schottky contact between two transistor cells. Fig. 2A - 2B illustrate schematic top and vertical cross-sectional views of a portion of a silicon carbide device with planar gate structures. Fig. 3A-3B illustrate schematic top and vertical cross-sectional views of a portion of a silicon carbide device having trench gate structures and shallow isolation regions. Fig. 4A-4E illustrate schematic top and vertical cross-sectional views of a portion of a silicon carbide device according to one embodiment having deep shielding and isolation regions. Fig. 5A-5B illustrate schematic top and vertical cross-sectional views of a portion of a silicon carbide device according to another embodiment having deep shielding and isolation regions. Fig. 6A-6B illustrate schematic top and vertical cross-sectional views of a portion of a silicon carbide device according to one embodiment having deep shield regions and shallow separation regions. Fig. 7A-7B illustrate schematic top and vertical cross-sectional views of a portion of a silicon carbide device according to one embodiment having a cathode region with upper and lower cathode regions. Fig. 8A-8B illustrate schematic top and vertical cross-sectional views of a portion of a silicon carbide device according to one embodiment with respect to dividing regions between lateral cathode regions. Fig. 9A-9E illustrate schematic top and vertical cross-sectional views of a portion of a silicon carbide device according to one embodiment having standoff regions between the cathode region and one of the gate structures. Fig. 10A-10E illustrate schematic top and vertical cross-sectional views of a portion of a silicon carbide device according to one embodiment having standoff regions between the cathode region and both adjacent gate structures. Fig. 11A-11E illustrate schematic top and vertical cross-sectional views of a portion of a silicon carbide device with Schottky contacts according to an embodiment having a wide bottom cathode region. Fig. 12A-12E illustrate schematic top and vertical cross-sectional views of a portion of a silicon carbide device having symmetric transistor cells. Fig. 13A-13B illustrate schematic top views of portions of a silicon carbide device according to embodiments having cathode regions adjacent to both adjacent gate structures. Fig. 14A-14B illustrate schematic top views of portions of a silicon carbide device according to embodiments having cathode regions in contact with one of two adjacent gate structures. Fig. 15A-15B illustrate schematic top views of portions of a silicon carbide device according to embodiments having cathode regions separated from both adjacent trench gate structures. DETAILED DESCRIPTION

[0010] In the following detailed description, reference is made to the accompanying drawings, which form a part hereof, and in which is shown, for illustrative purposes, specific embodiments in which a silicon carbide device may be constructed. The drawings are not to scale and are for illustrative purposes only. Corresponding elements are designated by the same reference numerals throughout the various drawings unless otherwise noted.

[0011] As used herein, the terms “having,” “containing,” “comprising,” “having,” and the like are open-ended terms that indicate the presence of the stated structures, elements, or features, but do not preclude the presence of additional elements or features.

[0012] The term "electrically connected" describes a permanent, low-resistance connection between electrically connected elements, for example, a direct contact between the elements in question or a low-resistance connection via a metal and / or a highly doped semiconductor material. The term "electrically coupled" encompasses the possibility that one or more intermediate elements suitable for signal and / or power transmission may be present between the electrically coupled elements, for example, elements that temporarily provide a low-resistance connection in a first state and a high-resistance electrical decoupling in a second state.

[0013] An ohmic contact is a non-rectifying electrical junction with a linear or nearly linear current-voltage characteristic. A Schottky contact is a metal-semiconductor junction with rectifying characteristics, where the work function of the metal and the dopant concentration in the semiconductor material are selected such that, in the absence of an externally applied electric field, a depletion zone forms in the semiconductor material along the metal-semiconductor junction. In the context of a Schottky contact, the term "metal-semiconductor junction" can also refer to a junction between a metal-like semiconductor and a semiconductor, where the junction has the same characteristics as a metal-semiconductor junction. For example, it may be possible to form a Schottky contact between polycrystalline silicon and silicon carbide. If two components (e.g.If two regions each form an ohmic contact or a Schottky contact, this may mean that an ohmic contact or a Schottky contact exists between the two components. In both cases, it is possible that the two regions are directly adjacent to each other. However, it is also possible that another component is positioned between the two regions.

[0014] A safe operating area (SOA) defines the voltage and current conditions under which a semiconductor device can be expected to operate without self-damage. The SOA is specified by published maximum values ​​for device parameters such as maximum continuous load current, maximum gate voltage, and others.

[0015] The figures illustrate relative doping concentrations by indicating “ - " or " + ” next to the doping type “n” or “p”. For example, “n -” a doping concentration that is lower than the doping concentration of an “n” doping region, while an “n + "-doping region has a higher doping concentration than an "n"-doping region. Doping regions of the same relative doping concentration do not necessarily have the same absolute doping concentration. For example, two different "n"-doping regions can have the same or different absolute doping concentrations.

[0016] Two adjacent doping regions of the same conductivity type and with different dopant concentrations form a unipolar junction, e.g., an (n / n+) or a (p / p+) junction, along an interface between the two doping regions. In a unipolar junction, a dopant concentration profile orthogonal to the unipolar junction may exhibit a step or inflection point at which the dopant concentration profile changes from concave to convex or vice versa.

[0017] Ranges specified for physical dimensions include the boundary values. For example, a range for a parameter y from a to b is read as a ≤ y ≤ b. The same applies to ranges with a boundary value such as "at most" and "at least."

[0018] The main components of a layer or structure made of a chemical compound or alloy are those elements whose atoms form the chemical compound or alloy. For example, nickel and silicon are the main components of a nickel silicide layer, and copper and aluminum are the main components of a copper-aluminum alloy.

[0019] According to one embodiment, a silicon carbide device may include a silicon carbide body having a source region of a first conductivity type, a cathode region of the first conductivity type, and isolation regions of a second, complementary conductivity type. A stripe-shaped gate structure may extend along a first direction and may directly border the source region and the isolation regions. The first direction may be a lateral direction of the silicon carbide body.

[0020] Along the first lateral direction, the cathode region is located between two of the separation regions. In some embodiments, exactly one cathode region may be located between two of the separation regions. At least one of (e.g., exactly one of) the separation regions is located between the cathode region and the source region. Each separation region may separate a cathode region and a source region. A first load electrode and the source region form an ohmic contact. The first load electrode and the cathode region form a Schottky contact. In at least one embodiment, the first load electrode may also form an ohmic contact with the separation regions. Furthermore, it may be possible for the first load electrode to form an ohmic contact with a source region and / or a body region and / or a shielding region.

[0021] The source region may form part of a transistor cell, so that a Schottky contact may be formed along the first direction between two transistor cells.

[0022] In at least one operating mode of the semiconductor device, e.g. a reverse conducting mode, a bipolar current comprising holes and electrons flows through the silicon carbide body, wherein the bipolar current passes through at least one pn junction of the transistor cell. In particular, in the vicinity of the pn junctions and further, possibly, in the vicinity of unipolar junctions, holes and electrons can recombine at a more than insignificant rate. The recombination rate can release energy (e.g. thermal energy) and / or momentum that can promote bipolar deterioration. Bipolar deterioration refers to the growth of pre-existing crystallographic defects. For example, BPDs (Basal Plane Dislocations) between neighboring lattice planes can form SSFs (Shockley stacking faults).Shockley stacking faults) that grow along the lattice planes in a direction that is predominantly transverse to a vertical load current flow through the silicon carbide body, so that the lattice defects can increasingly hinder the vertical load current flow through the silicon carbide body.

[0023] A forward voltage drop across a Schottky contact in silicon carbide is typically lower than a forward voltage drop across a pn junction in silicon carbide, so that a reverse current through the silicon carbide body flows predominantly through the Schottky contact. In silicon carbide devices with a high nominal reverse voltage of at least 3.3 kV, the drift region resistance typically dominates the total on-resistance, whereas the channel resistance contributes only marginally to the total on-resistance. In such devices, the area loss due to the Schottky contact increases the on-resistance only to a comparatively small extent. The reduced overall channel width can also contribute to improved short-circuit robustness.

[0024] The ratio between a Schottky contact area of ​​the Schottky contact and the transistor area of ​​the transistor cell can be adjusted, for example, by adjusting the ratio between a length of the cathode region and a length of the source region along the first direction. The cathode region can be defined by means of masking implantations that define the transistor cell, e.g., by means of masking implantations that define a source region, a body region, and / or a shielding region of the transistor cell.

[0025] At least one of the isolation regions can isolate and / or shield the cathode region from the source region. The isolation regions can be shaped and arranged to improve shielding of the transistor channel, to reduce DIBL (drain-induced barrier depression), and to shield the Schottky contact from electric fields to reduce leakage current through the Schottky contact.

[0026] According to one embodiment, the gate structure includes a gate electrode formed on a first surface of the silicon carbide body. In other words, the gate electrode may be a so-called planar gate electrode. By implementing the cathode region and the isolation regions, established designs for silicon carbide transistors can be modified for increased long-term reliability, reduced DIBL, and / or improved short-circuit robustness.

[0027] According to another embodiment, the gate structure may extend from a first surface of the silicon carbide body into the silicon carbide body, for example, along a vertical direction perpendicular to the first direction. In other words, the gate structure may be a trench gate structure. The trench gate structure enables a vertical transistor channel, which may allow for higher area efficiency. Therefore, the cathode region and the isolation regions may be implemented with small structural sizes and / or with high area efficiency. For example, it may be possible to provide the cathode region and the isolation region while still maintaining a high channel density for the MOSFET channel.

[0028] According to one embodiment, the silicon carbide body includes a body region of the second conductivity type and a drift structure. The body region and the drift structure form a first pn junction. Separately or in combination, the body region can directly border the drift structure. The first load electrode and the body region form an ohmic contact. The lower forward voltage drop across the Schottky contact compared to the forward voltage drop across the first pn junction can suppress or at least significantly reduce a bipolar current across the first pn junction and can prevent or at least significantly reduce bipolar degradation.

[0029] According to one embodiment, a first vertical extension of the isolation regions (e.g., each of the isolation regions) is equal to or smaller than a maximum vertical distance between the first pn junction and the first surface. This means that along the vertical direction, the body region may extend at least as deep as, or even deeper than, each of the isolation regions into the silicon carbide body. Comparatively shallow isolation regions with a vertical extension no greater than the maximum vertical distance between the first pn junction and the first surface may enable lateral current spreading directly beneath the isolation regions and thus contribute to a reduction in the ohmic resistance of a JFET (junction field-effect transistor) structure, which may form the isolation regions and / or further p-doped regions around the Schottky contact, e.g., beneath the body region.

[0030] The maximum vertical distance between the first pn junction and the first surface may correspond to a maximum vertical extension of the body region. For example, the body region may contain a highly doped body contact region directly adjacent to the first load electrode, and the first vertical extension of the isolation regions may be equal to a vertical extension of the body contact region. A maximum dopant concentration in the body contact region may be equal to a maximum dopant concentration in the isolation regions. It is possible for the isolation regions to be formed without an additional implantation process in the same process step as the body contact region, for example, by modifying an implantation mask used to define the body contact regions.

[0031] According to another embodiment, the first vertical extension of the isolation regions may be greater than a maximum vertical distance between the first pn junction and the first surface. This means that, along the vertical direction, the isolation regions may extend deeper into the silicon carbide body than the body region. A comparatively large vertical extension of the isolation regions may enable shielding of the transistor channels from more than two lateral sides and may also improve the shielding of the Schottky contact against high electric field strengths. In particular, the cathode region and adjacent isolation region may form an MPS (merged pin Schottky) diode structure.

[0032] The silicon carbide device may include a shield region of the second conductivity type. The shield region may form a pn junction with the drift structure and / or may directly border the drift structure. The shield region and the first load electrode may form an ohmic contact. For example, the first vertical extent may be equal to a maximum vertical distance between the first surface and the pn junction, between the shield region and the drift structure. The vertical extent of the shield region may be greater than a vertical extent of a trench gate structure.

[0033] According to one embodiment, the drift structure may contain a drift zone of the first conductivity type. The drift zone may be lightly doped, i.e., have a lower doping concentration than the source region and / or the cathode region. The cathode region and the drift structure may form a unipolar junction. A maximum dopant concentration in the cathode region may be at least twice as high, e.g., at least ten times as high, as a maximum dopant concentration in the drift zone. A higher dopant concentration in the cathode region may reduce the ohmic resistance between the Schottky contact and the drift zone and may define a suitable forward voltage drop at the Schottky contact.

[0034] The formation of the cathode region can be combined with the formation of a current spreading region between the drift zone and the body region. For example, at least a portion of the cathode region can be formed with the same implantation defining the current spreading region and without additional process steps, e.g., by modifying an implantation mask for the current spreading region. A current spreading region can be a region that prevents the generation of a space charge region via a current-driven voltage drop. For example, the current spreading region can have a higher doping concentration than an adjacent drift zone.

[0035] According to one embodiment, the silicon carbide device may include at least one separating region of the second conductivity type, wherein the at least one separating region may separate the cathode region along the first direction into at least two cathode sections. The at least one separating region may have the same vertical extent as the separation regions, the same maximum dopant concentration as the separation regions, and / or the same vertical dopant profile as the separation regions. The separating regions may further fine-tune the shielding characteristics for the Schottky contact.

[0036] According to one embodiment, a plurality of cathode regions and a plurality of source regions may be arranged along the first direction. Each of the cathode regions may be located between two of the separation regions. At least one (e.g., exactly one) of the separation regions is located between at least one (e.g., exactly one) of the cathode regions and at least one (e.g., exactly one) of the source regions. Along the first direction, the source regions may alternate with the cathode regions, wherein the separation regions may separate each cathode region from the two adjacent source regions along the first direction.

[0037] A desired ratio of the Schottky contact area to the transistor area can be adjusted by selecting a total length of the cathode regions along the first direction. Transistor current and Schottky current can be evenly distributed across the lateral cross-sectional area of ​​the silicon carbide body. The length of the cathode regions along the first direction and the vertical extension of the separation regions can adjust the shielding characteristics for the Schottky contacts and the transistor channel.

[0038] According to one embodiment, the source region directly adjoins a first gate sidewall of the gate structure. A shielding region of the second conductivity type may adjoin (e.g., directly adjoin) a second gate sidewall of the gate structure. The second gate sidewall may be opposite the first gate sidewall. The first and second gate sidewalls may both extend along the first direction (e.g., may be longitudinal sidewalls of the gate structure). The source region may be located directly opposite the shielding region with respect to the gate structure. The transistor cell may thus be an asymmetric transistor cell with a transistor channel formed along only one of the two longitudinal sidewalls of the gate structure. The formation of the isolation regions may be combined with implantation processes that define the shielding regions.For example, the separation regions can be formed without additional process steps by modifying one or more implantation masks used to form the shielding regions.

[0039] According to another embodiment, the silicon carbide device includes a first source region and a second source region. The first source region directly adjoins a first gate sidewall of the gate structure. Directly opposite the first source region, the second source region may directly adjoin a second gate sidewall of the gate structure, which second gate sidewall is opposite the first gate sidewall. In other words, the transistor cell may be a symmetric transistor cell with transistor channels formed along both longitudinal sidewalls of the gate structure. Providing transistor channels on both sides of the gate structures may increase the overall channel width without increasing the chip area.

[0040] According to one embodiment, the silicon carbide device includes a plurality of parallel gate structures, and the cathode region can extend from one gate structure to an adjacent gate structure. The entire width of a SiC mesa between two adjacent gate structures can be used as a Schottky contact, resulting in high area efficiency.

[0041] According to another embodiment, the silicon carbide device includes a plurality of parallel gate structures. A standoff region of the second conductivity type may be formed between one of the gate structures and the cathode region. The standoff region may enhance the shielding of the Schottky contact. The standoff region may also enhance the shielding of the bottom of the trench gate structure in addition to a shielding structure. For example, the standoff region may extend at least as deep, or even deeper (e.g., slightly deeper), than the trench gate structure into the silicon carbide body.

[0042] According to another embodiment, the silicon carbide device includes a plurality of parallel gate structures, and spacer regions of the second conductivity type may be formed between the cathode region and two of the gate structures. The spacer regions may enable shielding of the Schottky contact on all lateral sides.

[0043] According to one embodiment, the cathode region may comprise a lower cathode region and an upper cathode region between the lower cathode region and the first surface. A maximum first width of the upper cathode region orthogonal to the first direction (and orthogonal to the vertical direction) is smaller than a maximum second width of the lower cathode region orthogonal to the first direction (and orthogonal to the vertical direction). The resulting JFET structure with a widened region at a distance from the first surface may further improve the shielding of the Schottky contact.

[0044] In an alternative embodiment, shielding can be improved by means of a narrowed lower cathode region. In such an embodiment, a maximum first width of the upper cathode region orthogonal to the first direction (and orthogonal to the vertical direction) is greater than a maximum second width of the lower cathode region orthogonal to the first direction (and orthogonal to the vertical direction).

[0045] According to one embodiment, a plurality of cathode regions and a plurality of source regions are arranged between the gate structures, in particular between two adjacent gate structures. The cathode regions may be arranged along the first direction and / or may extend along it, and the source regions may be arranged along a second direction orthogonal to the first direction (and orthogonal to the vertical direction) and / or may extend along it. In this way, the cathode regions and the isolation regions can be formed by means of comparatively simple modifications of implantation masks.

[0046] According to another embodiment, the cathode regions and the source regions may alternate along the second direction (ie along lines extending orthogonally to the first direction).

[0047] The Schottky contacts and the transistor cells can thus be arranged in a checkerboard pattern, with the transistor cells corresponding to the "white squares" and the Schottky contacts corresponding to the "black squares" of a checkerboard. Both the unipolar Schottky current and the bipolar transistor current can thus be distributed more evenly across the lateral cross-sectional area of ​​the silicon carbide body.

[0048] Fig. 1A and Fig. 1B show a portion of a semiconductor device 500 having a silicon carbide body 100. The semiconductor device 500 may be or may include, for example, an IGFET (Insulated Gate Field Effect Transistor), for example, a MOSFET (Metal Oxide Semiconductor FET), an MCD (MOS Controlled Diode), or an IGBT (Insulated Gate Bipolar Transistor).

[0049] The silicon carbide body 100 may contain or consist of single-crystal silicon carbide, e.g., a silicon carbide crystal containing the main components silicon and carbon. The silicon carbide crystal may contain undesired impurities such as hydrogen and / or oxygen and / or intended impurities, e.g., dopant atoms. The polytype of the silicon carbide crystal may be 15R or may be a hexagonal polytype, e.g., 2H, 6H, or 4H. The silicon carbide body 100 may contain or consist of an epitaxially grown silicon carbide layer.

[0050] The silicon carbide semiconductor body may have an off-direction, for example, in a <11-20> or a <1-100> direction. A first surface 101 on a front side of the silicon carbide body 100 may be planar or ribbed. A surface normal 104 orthogonal to a planar first surface 101 or orthogonal to a mid-plane of a ribbed first surface 101 defines a vertical direction. Directions orthogonal to the surface normal 104 are lateral directions (e.g., the first and second directions). The surface normal 104 may coincide with a principal lattice direction or may be inclined to a principal lattice direction by an angle to the axis, wherein the angle to the axis may be in a range from 2° to 8°, in particular 4°.At the back of the silicon carbide body 100, a second surface 102 may extend parallel to a planar first surface 101 or parallel to a central plane of a ribbed first surface 101.

[0051] At least two transistor cells TC and one Schottky contact SC are formed on the front side of the silicon carbide body 100, wherein each Schottky contact SC is formed between two transistor cells TC along a lateral first direction 291. A drift structure 130 extends laterally through the silicon carbide body 100 between the transistor cells TC and the second surface 102.

[0052] The drift structure 130 includes a drift zone 131 of a first conductivity type. The drift zone 131 forms a voltage-holding structure, wherein a vertical extension and a dopant concentration in the drift zone 131 can be selected such that the semiconductor device 500 provides a nominal blocking voltage capability in an off-state of the semiconductor device 500. The drift zone 131 can be formed in an epitaxially grown layer. An average net dopant concentration in the drift zone 131 can, for example, be in a range of 1E15 cm -3 up to 5E16 cm -3 lay.

[0053] A highly doped contact region 139 may be formed between the drift structure 130 and a second load electrode 320 directly adjacent to the second surface 102, wherein the highly doped contact region 139 forms a low-resistance ohmic contact with the second load electrode 320. The contact region 139 may have the same conductivity type as the drift zone 131, the opposite conductivity type, or may contain zones of both conductivity types.

[0054] Each transistor cell TC includes a source region 110 and a body region 120, wherein the body region 120 and the drift structure 130 form a first pn junction pn1, and wherein the body region 120 and the source region 110 form a second pn junction pn2. A gate structure 150 extends along the first direction 291 and is in contact with the source regions 110 and the body regions 120 of the transistor cells TC.

[0055] Cathode regions 160 are located between the source regions 110 along the lateral first direction 291. The cathode regions 160 and the drift structure 130 may form unipolar junctions. A length l0 of the cathode region 160 along the first direction 291 may range from a few 100 nm to almost the entire length of the trench gate structure along a longitudinal extension of the trench gate structure. For example, a length of the cathode region 160 is at least 0.5 µm, e.g., at least 1 µm, or at least 2 µm, and / or at most 5 µm, e.g., at most 2 µm.

[0056] Separation regions 140 of the second conductivity type separate the source regions 110 and the cathode regions 160 along the first direction 291. A first vertical extent v1 of the separation regions 140 can be in a range from a few tens of nanometers, in particular at least 100 nm, to a few micrometers. The first vertical extent v1 can be equal to a maximum vertical distance v2 between the first pn junction pn1 and the first surface 101, smaller than the maximum vertical distance v2, or greater than the maximum vertical distance v2.

[0057] The source regions 110 and the body regions 120 of the transistor cells TC may be electrically connected or coupled to a first load electrode 310, which may form a first load terminal L1 or which may be electrically connected or coupled to such a first load terminal L1. The cathode regions 160 and the first load electrode 310 form Schottky contacts SC.

[0058] The second load electrode 320 may form a second load terminal L2 or may be electrically connected or coupled to such a second load terminal. Gate electrodes 155 of the transistor cell TC may be electrically connected or coupled to a gate terminal G. The first load terminal L1 may be an anode terminal of an MCD, a source terminal of an IGFET, or an emitter terminal of an IGBT. The second load terminal L2 may be, for example, a cathode terminal of an MCD, a drain terminal of an IGFET, or a collector terminal of an IGBT.

[0059] The transistor cells TC may be n-channel FET cells with p-doped body regions 120, n-doped source regions 110, and an n-doped drift zone 131. Alternatively, the transistor cells TC are p-channel FET cells with n-doped body regions 120, p-doped source regions 110, and a p-doped drift zone 130. As a mere example, the following description refers to n-channel FET cells.

[0060] Fig. 2A and Fig. 2B illustrate planar gate structures 150 including conductive gate electrodes 155 formed on the first surface 101. A gate dielectric 159 separates the gate electrode 150 from the silicon carbide body 100. The transistor cells TC may be arranged in pairs, wherein the transistor cells TC of each pair are arranged with respect to a plane corresponding to the cross-sectional planes of the Fig. 2A and Fig. 2B orthogonal vertical plane of symmetry are mirrored.

[0061] An interlayer dielectric 210 may separate the gate electrode 155 and the first load electrode 310. The interlayer dielectric 210 may further separate the first load electrode 310 and a drain region 132 of the drift structure 130, wherein the drain region 132 is formed between the drift zone 131 and the first surface 101. In the on-state of the transistor cells TC, inversion channels form in the silicon carbide body 100 along the first surface opposite the gate electrode 155, between the source region 110 and the drain region 132. Between the source regions 110 of adjacent pairs of transistor cells TC, the body regions 120 may contain highly doped body contact regions 125 that directly border the first surface.

[0062] The separation regions 140 may be formed as lateral extensions of the body regions 120 along the first direction 291. The separation regions 140 may include highly doped separation contact regions 141 formed as lateral extensions of the body contact regions 125.

[0063] Fig. 3A and Fig. 3B shows trench gate structures 150 extending from the first surface 101 into the silicon carbide body 100. The source regions 110 and the body regions 120 are in contact with first gate sidewalls 151 of the trench gate structures 150.

[0064] Shield regions 170 may be in contact with opposite second gate sidewalls 152 of the trench gate structures 150 and may also be formed along a portion of a bottom surface of the trench gate structures 150. In the shield regions 170, a maximum dopant concentration may be higher than in the body regions 120. The dopant concentration in the shield region 170 may be sufficiently high such that no inversion layer is formed along the second trench sidewalls 152, at least as long as the semiconductor device 500 operates within the SOA. The shield regions 170 are separated from the first trench sidewalls 151. For example, the source regions 110 and the body regions 120 may be located between the shield regions 170 and the first trench sidewalls 151 of the gate structures 150.The shielding regions 170 may include highly doped shielding contact regions 171 along the first surface 101, wherein the first load electrode 310 and the highly doped shielding contact regions 171 may form ohmic contacts. A vertical extension v1 of the isolation region 140 may be equal to a vertical extension v3 of the shielding contact regions 171.

[0065] In Fig. 4A-4E, each shield region 170 comprises a shield contact region 171, a shield connection region 172, and a shield bottom region 173. The shield bottom region 173 may directly border the drift structure 130, and a maximum dopant concentration in the shield bottom region 173 may have a greater distance from the first surface 101 than the bottom of the trench gate structure 150. The shield connection regions 172 connect the shield bottom regions 173 to the highly doped shield contact regions 171.

[0066] A maximum dopant concentration in the shielding connection regions 172 is at least ten times higher than an average dopant concentration in the body regions 120. An interlayer dielectric 210 separates the first load electrode 310 and the gate electrode 155 in the trench gate structures 150.

[0067] The drift structure 130 includes current spreading regions 137 between the body regions 120 and the drift zone 131. The current spreading regions 137 can separate the body regions 120 and the drift zone 131. The current spreading regions 137 and the drift zone 131 can form one or more first unipolar junctions jn1. The current spreading regions 137 can also be formed between the cathode region 160 and the drift zone 131. A maximum dopant concentration in the current spreading regions 137 is at least twice as high, for example, at least ten times as high, as a maximum dopant concentration in the drift zone 131. A maximum dopant concentration in the cathode regions 160 can be equal to, higher than, or lower than the maximum dopant concentration in the current spreading regions 137.

[0068] The separation regions 140 may be formed from the same implantations that define the shield contact regions 171, the shield connection regions 172, and the shield bottom regions 173 of the shield regions 170. For example, the separation regions 140 may include separation bottom regions 143 having the same vertical extent, the same distance from the first surface 101, and / or the same dopant concentration as the shield bottom regions 173 of the shield regions 170. The separation connection regions 142 may have the same vertical extent, the same distance from the first surface 101, and / or the same dopant concentration as the shield connection regions 172 of the shield regions 170.

[0069] Additionally or alternatively, isolation contact regions 141 may have the same vertical extent and / or the same dopant concentration as the shielding contact regions 171 of the shielding regions 170.

[0070] The separation regions or at least areas of the separation regions 140 can form lateral extensions or extensions of the shielding regions 170.

[0071] The shielding regions 170 and the separation regions 140 may result entirely from the same implantations. Alternatively, the separation regions 140 and the shielding regions 170 may differ in at least one implantation.

[0072] For example, included in Fig. 5A and Fig. 5B, the separation regions 140 do not include a region corresponding to the shielding ground region 173. Instead, the separation region 140 includes only a separation contact region 141 corresponding to the shielding contact region 171 and a separation connection region 142 corresponding to the shielding connection region 172.

[0073] In Fig. 6A and Fig. 6B, the separation regions 140 comprise a separation contact region 141, which may correspond to the shielding contact region 171, and a further region formed by a lateral extension or extension of the body region 120.

[0074] In Fig. 7A and Fig. 7B, the cathode region 160 comprises a lower cathode region 162 and an upper cathode region 161 between the lower cathode region 162 and the first surface 101. A maximum first length l1 along the first direction 291 in the upper cathode region 161 may be greater than a maximum second length l2 of the lower cathode region 162. The lower cathode region 162 may be formed simultaneously with the current spreading regions 137 by using the same implantation processes and / or may have the same vertical dopant profile as the current spreading regions. A maximum dopant concentration in the upper cathode region 161 may be equal to or greater than a maximum dopant concentration in the lower cathode region 162.

[0075] In Fig. 8A and Fig. 8B, a dividing region 180 of the conductivity type of the separation regions 140 divides the cathode region 160 laterally along the first direction 291 into two lateral cathode regions 163. The dividing regions 180 may have the same vertical configuration as the separation regions 140 and / or the shielding regions 170 or a different configuration.

[0076] For example, the dividing regions 180 may be formed from the same implantations that define the isolation contact regions 141, the isolation connection regions 142, and the isolation bottom regions 143 of the isolation regions 140. For example, the dividing regions 180 may comprise dividing bottom regions 183 having the same vertical extent, the same vertical distance from the first surface 101, and / or the same dopant concentration as the isolation bottom regions 143. Dividing connection regions 182 may have the same vertical extent, the same distance from the first surface 101, and / or the same dopant concentration as the isolation connection regions 142. Additionally or alternatively, dividing contact regions 181 may have the same vertical extent and / or the same dopant concentration as the isolation contact regions 141.Alternatively, the dividing regions 180 may differ from the separation regions 140 in at least one implantation.

[0077] Fig. 8A and Fig. 8B show only a single dividing region 180 per cathode region 160. Other embodiments may provide two or more dividing regions 180 that separate a cathode region 160 into three or more lateral cathode regions 163.

[0078] In Fig. 9A-9E, the cathode regions 160 are in contact with a gate structure 150, and a spacer region 145 separates the cathode region 160 from a second adjacent gate structure 150. The cathode regions 160 may have the same width orthogonal to the first direction 291 as the source regions 110.

[0079] As in Fig. 9A and Fig. As illustrated in Figure 9E, the standoff regions 145 may be lateral extensions of the shielding regions 170 and may include the subregions corresponding to the shielding contact region 171, the shielding connection region 172, and the shielding bottom region 173. Alternatively, the standoff regions 145 may differ from the shielding regions 170 by at least one implantation.

[0080] In Fig. 10A-10E, spacing regions 145 separate the cathode region 160 from both adjacent gate structures 150. As illustrated in the figures, the spacing regions 145 and the separation regions 140 may be formed as lateral extensions of the shielding regions 170.

[0081] The cathode region 160 may include a lower cathode region 162 and an upper cathode region 161 between the lower cathode region 162 and the first surface 101. A maximum first width w1 orthogonal to the first direction 291 in the upper cathode region 161 may be smaller than a maximum second width w2 of the lower cathode region 162. The lower cathode region 162 may be formed simultaneously with the current spreading regions 137 using the same implantation processes and / or may have the same vertical dopant profile as the current spreading regions 137. A maximum dopant concentration in the upper cathode region 161 may be equal to or greater than a maximum dopant concentration in the lower cathode region 162.Decoupling the position of the edges between the spacer regions 145 and the upper cathode regions 161 from the position of the edges between the spacer regions 145 and the lower cathode regions 162 can fine-tune a trade-off between conductivity and shielding. Additionally, shallow n-type implants in the upper cathode region 161 can also overcompensate for (p+) extensions of an implant extending to the first surface 101.

[0082] Fig. 11A-11E show another silicon carbide device 500 including a silicon carbide body 100. Transistor cells TC having gate structures 150 extending along the lateral first direction 291 are formed along a first surface 101 of the silicon carbide body 100. A drift structure having a lightly doped drift zone 131 is formed between the transistor cells TC and a second surface 102 of the silicon carbide body 100. The silicon carbide body 100 further includes shield regions 170 and cathode regions 160, as described above, and the semiconductor device 500 further includes a first load electrode 310 on the backside of the silicon carbide body and a second load electrode 320 on the backside of the silicon carbide body, as described above.

[0083] Each cathode region 160 may extend from the first surface 101 into the silicon carbide body 100. The cathode regions 160 and the drift zone 131 have the same conductivity type. An average dopant concentration in the cathode regions 160 may be the same as or higher than the average dopant concentration in the drift zone 131. For example, an average dopant concentration in the cathode region 160 may be at least twice as high as an average dopant concentration in the drift zone 131. The cathode region 160 and the drift structure, e.g., the drift zone 131 or a barrier layer of the conductivity type of the drift zone 131, may form a unipolar junction. A length of the cathode region 160 along the first direction 291 may be at least 100 nm.For example, the length of the cathode regions 160 may be at least 500 nm, for example at least 1 µm, or at least 2 µm and / or at most 5 µm, for example at most 2 µm.

[0084] The cathode regions 160 and the first load electrode 310 form Schottky contacts SC. Along the lateral first direction 291, each Schottky contact SC is formed between two adjacent transistor cells TC.

[0085] Each shield region 170 includes a shield connection region 172 and a shield bottom region 173. The shield connection region 172 is located between the first surface 101 and the shield bottom region 173. The shield connection region 172 may further include a shield contact region formed along the first surface 101 as described above.

[0086] The shielding regions 170 include lateral spacer regions 174. Each spacer region 174 is in contact with an active sidewall of a gate structure 150 between adjacent source regions 110. The shielding region 170 laterally completely surrounds each cathode region 160. Each cathode region 160 may be spaced apart from both adjacent gate structures 150.

[0087] Each cathode region 160 comprises a lower cathode region 162 and an upper cathode region 161 between the first surface 101 and the lower cathode region 162. An average dopant concentration in the lower cathode region 162 and an average dopant concentration in the upper cathode region 161 may be the same or may be different. The upper cathode region 161 has a first average width wm1 along the first direction 291. The lower cathode region 162 has a second average width wm2 along the first direction 291. The second average width wm2 is at least 120%, for example at least 150% or at least 200%, of the first average width wm1. A vertical extension v4 of the lower cathode region 162 may be at least 50 nm, for example at least 100 nm.

[0088] The upper cathode region 161 is laterally opposite the shield connection region 172. The upper cathode region 161 and the shield connection region 172 can form a vertical or nearly vertical pn junction. The lower cathode region 162 is laterally opposite the shield bottom region 173. The lower cathode region 162 and the shield bottom region 173 can form a vertical or nearly vertical pn junction. A higher unipolar reverse current can flow through the cathode region 160 without the body diode becoming active.

[0089] Fig. 12A-12E show a silicon carbide device 500 with symmetric transistor cells. First source regions 111 directly adjoin gate sidewalls 151 of each trench gate structure 150, and second source regions 112 directly adjoin a second gate sidewall 152, with the first and second source regions 111, 112 being formed directly opposite each other. Shield regions 170 may extend from the first surface 101 into the silicon carbide body 100 in the center of the SiC mesas 190 between the first and second source regions 111, 112.

[0090] Fig. 13A-15B show top views of semiconductor devices 500 with stripe-shaped trench gate structures 150 extending along the first direction 291. SiC mesas 190 between adjacent trench gate structures 150 comprise the semiconducting regions of the transistor cells TC and Schottky contacts SC with source regions 110, cathode regions 160, and separation regions 140 separating adjacent cathode regions 160 and source regions 110.

[0091] Fig. 13A and Fig. 13B show cathode regions 160 extending across the entire width of the semiconductor mesas 190.

[0092] In Fig. 13A, the cathode regions 160 of adjacent SiC mesas 190 extend along a lateral second direction 292 orthogonal to the first direction 291. The cathode regions 160 may thus extend along lines extending along the second direction 292.

[0093] In Fig. 13B, only the cathode regions 160 in every second SiC mesa 190 are arranged along the same line along the second direction 292. In other words, cathode regions 160 alternate with source regions 110 along the second direction 292.

[0094] The Fig. 14A and Fig. The embodiments shown in Figure 14B combine the arrangement of Schottky contacts SC and transistor cells TC as shown in Fig. 13A and Fig. 13B, with the layout of a Schottky contact SC with a spacer region 145 separating the cathode region 160 from one of the adjacent gate structures 150.

[0095] The Fig. 15A and Fig. The embodiments shown in Figure 15B combine the arrangement of Schottky contacts SC and transistor cells TC as shown in Fig. 13A and Fig.13B, with the layout of a Schottky contact with spacer regions 145 separating the cathode region 160 from two adjacent gate structures 150.

Claims

[1] Silicon carbide device (500), comprising: a silicon carbide body (100) having a source region (110) of a first conductivity type, a cathode region (160) of the first conductivity type, and separation regions (140) of a second conductivity type, a strip-shaped gate structure (150) extending along a first direction (291) and directly bordering the source region (110) and the isolation regions (140), and a first load electrode (310), wherein, along the first direction (291), the cathode region (160) is located between two of the separation regions (140) and at least one of the separation regions (140) is located between the cathode region (160) and the source region (110), and wherein the source region (110) and the first load electrode (310) form an ohmic contact (OC) and the first load electrode (310) and the cathode region (160) form a Schottky contact (SC), wherein the gate structure (150) extends from a first surface (101) of the silicon carbide body (100) into the silicon carbide body (100), wherein the source region (110) directly borders a first gate sidewall (151) of the gate structure (150), and wherein a shielding region (170) of the second conductivity type directly opposite the source region (110) borders a second gate sidewall (152) of the gate structure (150) opposite the first gate sidewall (151), wherein the silicon carbide body (100) has a body region (120) of the second conductivity type and a drift structure (130), wherein the body region (120) and the drift structure (130) form a first pn junction (pn1) and wherein the first load electrode (310) and the body region (120) form an ohmic contact (OC), wherein a first vertical extent (v1) of the separation regions (140) is greater than a maximum vertical distance (v2) between the first pn junction (pn1) and the first surface (101). [2] Silicon carbide device according to the preceding claim, wherein the drift structure (130) has a drift zone (131) of the first conductivity type, the cathode region (160) and the drift zone (131) form a unipolar junction (jn), and a maximum dopant concentration in the cathode region (160) is at least twice as high as a maximum dopant concentration in the drift zone (131). [3] Silicon carbide device according to one of the preceding claims, further comprising: at least one dividing region (180) of the second conductivity type, wherein the at least one dividing region (180) separates the cathode region (160) into at least two cathode regions (163) along the first direction (291). [4] Silicon carbide device according to one of the preceding claims, wherein a plurality of cathode regions (160) and a plurality of source regions (110) are arranged along the first direction (291), wherein each of the cathode regions (160) is located between two of the separation regions (140), and wherein at least one of the separation regions (140) is located between at least one of the cathode regions (160) and at least one of the source regions (110). [5] Silicon carbide device according to one of the preceding claims, comprising: a plurality of parallel gate structures (150), wherein the cathode region (160) extends from one gate structure (150) to an adjacent gate structure (150). [6] Silicon carbide device according to one of claims 1 to 4, comprising: a plurality of parallel gate structures (150), wherein a spacer region (145) of the second conductivity type is formed between one of the gate structures (150) and the cathode region (160). [7] Silicon carbide device according to one of claims 1 to 4, comprising: a plurality of parallel gate structures (150), wherein spacing regions (145) of the second conductivity type are formed between the cathode region (160) and two gate structures (150). [8] Silicon carbide device according to the preceding claim, wherein the cathode region (160) has a lower cathode region (162) and an upper cathode region (161) between the lower cathode region (162) and the first surface (101), wherein a maximum first width (w1) of the upper cathode region (161) orthogonal to the first direction (291) is smaller than a maximum second width (w2) of the lower cathode region (162) orthogonal to the first direction (291). [9] Silicon carbide device according to one of the four preceding claims, wherein a plurality of cathode regions (160) and a plurality of source regions (110) are arranged between the gate structures (150), wherein the cathode regions (160) are arranged along the first direction (291) and the source regions (110) are arranged along a second direction (292) orthogonal to the first direction (291). [10] Silicon carbide device according to one of claims 5 to 8, wherein a plurality of cathode regions (160) and a plurality of source regions (110) are arranged between the gate structures (150), wherein the cathode regions (160) and the source regions (110) alternate along a second direction (292) orthogonal to the first direction (291).

Citation Information

Patent Citations

  • Semiconductor device

    JP2005108926A

  • Mosfet

    JP2018098447A

  • LDMOS transistor

    US20050285188A1

  • Semiconductor device with schottky barrier diode

    US20140175508A1

  • Semiconductor device and method of manufacturing thereof

    US20180182885A1