Silicon carbide semiconductor device

The silicon carbide semiconductor device addresses the issue of low short-circuit resistance and etching damage by employing a controlled impurity distribution and trench structure to manage leakage current and parasitic diode operation, enhancing both on-resistance and short-circuit tolerance.

JP2024158011APending Publication Date: 2024-11-08NATIONAL INSTITUTE OF ADVANCED INDUSTRIAL SCIENCE & TECHNOLOGY +1
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
JP2023072786
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-04-26
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

Conventional silicon carbide semiconductor devices with a built-in Schottky barrier diode (SBD) have lower short-circuit withstand capacity and suffer from variations in Schottky barrier height due to etching damage during trench formation, leading to inconsistent short-circuit resistance.

Method used

A silicon carbide semiconductor device with a specific trench structure and impurity concentration distribution, including a Schottky electrode and a SBD portion with controlled impurity levels, to suppress on-resistance and improve short-circuit resistance by managing leakage current and parasitic diode operation.

Benefits of technology

The device effectively reduces on-resistance and enhances short-circuit tolerance by controlling impurity concentrations and trench configurations, ensuring reliable operation under load short-circuit conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024158011000001_ABST
    Figure 2024158011000001_ABST
Patent Text Reader

Abstract

To provide a silicon carbide semiconductor device capable of improving a short circuit tolerance dose as well as suppressing an increase of an on-resistance.SOLUTION: In a semiconductor substrate 30, a gate trench 7a and a Schottky key trench 7b are alternately and repeatedly provided in a second direction Y. An n-type current diffusion region 3 is provided between trenches 7a and 7b that are adjacent each other, and includes a JFET upper part 3a that is contacted to a gate insulation film 8 of the gate trench 7a; and a SBD part 21 that is contacted to a Schottky key electrode 17 on a side wall of the Schottky key trench 7b. On a bonding surface of the Schottky key electrode 17 and the SBD part 21, a trench SBD 20 is formed. A carrier concentration of the SBD part 21 is 10% or more and less than 100% of a nitrogen concentration of the JFET upper part 3a. A width W1 of the SBD part 21 is narrower than the width projected to a direction where a second high concentration region 11b is separated from the side wall of the Schottky key trench 7b in parallel to the second direction Y, and is 0.1 μm or more and 0.5 μm or less.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] This disclosure relates to silicon carbide semiconductor devices. [Background technology]

[0002] Conventionally, in silicon carbide semiconductor devices using silicon carbide (SiC) as a semiconductor material, a trench-gate MOSFET (Metal Oxide Semiconductor Field Effect Transistor: a MOS type field effect transistor having an insulated gate with a three-layer structure of metal-oxide film-semiconductor) has been known in which a Schottky barrier diode (SBD) is built into the same semiconductor substrate (semiconductor chip) to suppress parasitic pn diode operation and thereby reduce on-resistance (see, for example, Patent Document 1 and Non-Patent Document 1 below). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2020-077664 A [Patent Document 2] International Publication No. 2022 / 244749 [Non-patent literature]

[0004] [Non-Patent Document 1] M. Okawa et al., "First Demonstration of Short-Circuit Capability for a 1.2 kV SiC SWITCH-MOS", IEEE Journal of the Electron Devices Society, Vol.7, (2019), pp.613-620 [Non-Patent Document 2] Krishnaswami et al., "A Study on the Reliability and Stability of High Voltage 4H-SiC MOSFET Devices", Materials Science Forum, Volumes 527-529, (2006), pp.1313-1316 [Non-Patent Document 3] T. Kimoto and JA Cooper, "Fundamentals of Silicon Carbide Technology: Growth, Characterization, Devices and Applications", November 2014, Wiley-IEEE Press, pp.487 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the short-circuit withstand capability of a MOSFET incorporating an SBD (the time from load short circuit to element destruction) is significantly lower than that of a normal MOSFET that does not incorporate an SBD (see Non-Patent Document 1 above). As disclosed in Patent Document 1 above, the short-circuit withstand capability can be improved by increasing the Schottky barrier height of the incorporated SBD, but due to the adverse effects of etching damage during trench formation, Fermi level pinning (a phenomenon in which the Fermi level becomes fixed as if pinned at a specific energy position) occurs at the interface (metal-semiconductor interface) of the Schottky junction formed along the inner wall of the trench in which the SBD is embedded, changing the band arrangement at the metal-semiconductor interface and causing variations in the Schottky barrier height, raising concerns about variations in the short-circuit withstand capability.

[0006] An object of the present disclosure is to provide a silicon carbide semiconductor device capable of suppressing an increase in on-resistance and improving short-circuit resistance in order to solve the problems associated with the conventional techniques described above. [Means for solving the problem]

[0007] A representative embodiment of the present disclosure is as follows. The silicon carbide semiconductor device according to this disclosure has the following features. A first semiconductor region of a first conductivity type is provided inside a semiconductor substrate made of silicon carbide. A second semiconductor region of a second conductivity type is provided between the front surface of the semiconductor substrate and the first semiconductor region. A third semiconductor region of a first conductivity type is selectively provided between the front surface of the semiconductor substrate and the second semiconductor region. A fourth semiconductor region of a first conductivity type is provided between the second semiconductor region and the first semiconductor region, in contact with the second semiconductor region and the first semiconductor region. The fourth semiconductor region is doped with a first conductivity type impurity at a concentration higher than that of the first semiconductor region.

[0008] A plurality of trenches are provided at a predetermined depth from the front surface of the semiconductor substrate. A first trench of the plurality of trenches penetrates the third semiconductor region and the second semiconductor region in the depth direction and terminates inside the fourth semiconductor region. A second trench of the plurality of trenches other than the first trench is disposed away from the third semiconductor region, penetrates the second semiconductor region in the depth direction and terminates inside the fourth semiconductor region. A gate electrode is provided inside the first trench via a gate insulating film. A Schottky electrode contacts the fourth semiconductor region on an inner wall of the second trench.

[0009] A first electrode is electrically connected to the second semiconductor region and the third semiconductor region, and is embedded in the second trench and contacts the Schottky electrode. A second electrode is electrically connected to the back surface of the semiconductor substrate. A first conductivity type SBD portion is part of the fourth semiconductor region and contacts the Schottky electrode at a sidewall of the second trench. A first conductivity type JFET upper portion is part of the fourth semiconductor region and is provided between the sidewall of the first trench and the SBD portion. A Schottky barrier diode is provided that utilizes the rectification of a Schottky barrier formed at a junction between the Schottky electrode and the SBD portion. The carrier concentration of the SBD portion is lower than the concentration of the first conductivity type impurity added to the JFET upper portion.

[0010] In addition, in the silicon carbide semiconductor device disclosed herein, in the above disclosure, the SBD portion is doped with substantially only the first conductivity type impurity. The carrier concentration of the SBD portion is determined by the concentration of the first conductivity type impurity doped in the SBD portion. The concentration of the first conductivity type impurity doped in the SBD portion is 1×10 16 / cm 3 and the concentration of the first conductivity type impurity added to the upper portion of the JFET is within a range of 10% or more and less than 100%.

[0011] In addition, in the silicon carbide semiconductor device disclosed herein, in the above disclosure, the SBD portion is doped with the first conductivity type impurity and the second conductivity type impurity, and the carrier concentration of the SBD portion is determined by subtracting a concentration of the second conductivity type impurity multiplied by an ionization rate from a concentration of the first conductivity type impurity doped in the SBD portion.

[0012] In addition, in the silicon carbide semiconductor device disclosed above, the second conductivity type impurity is aluminum, gallium or boron.

[0013] The silicon carbide semiconductor device according to this disclosure further includes a fifth semiconductor region of a second conductivity type, a sixth semiconductor region of a second conductivity type, and a JFET lower portion of a first conductivity type. The fifth semiconductor region is provided between the second semiconductor region and the first semiconductor region, spaced apart from the second semiconductor region, and faces a bottom surface of the first trench. The sixth semiconductor region of the second conductivity type is provided between the second semiconductor region and the first semiconductor region, spaced apart from the second semiconductor region, and faces a bottom surface of the second trench. The JFET lower portion is a part of the fourth semiconductor region, and is provided between the fifth semiconductor region and the sixth semiconductor region, which are adjacent to each other in a direction parallel to the front surface of the semiconductor substrate, between the JFET upper portion and the first semiconductor region. The SBD portion is not in contact with the JFET lower portion.

[0014] The silicon carbide semiconductor device according to this disclosure, in the above disclosure, further includes a fifth semiconductor region of a second conductivity type and a sixth semiconductor region of the second conductivity type. The fifth semiconductor region is provided between the second semiconductor region and the first semiconductor region, spaced apart from the second semiconductor region, and faces a bottom surface of the first trench. The sixth semiconductor region of the second conductivity type is provided between the second semiconductor region and the first semiconductor region, spaced apart from the second semiconductor region, and faces a bottom surface of the second trench. A width from a junction surface between the Schottky electrode and the SBD portion to an interface between the JFET upper portion and the SBD portion is narrower than a width of the sixth semiconductor region protruding in a direction away from a sidewall of the second trench in parallel to a front surface of the semiconductor substrate.

[0015] Moreover, the silicon carbide semiconductor device according to this disclosure, in the above disclosure, is characterized in that a width from a junction surface between the Schottky electrode and the SBD portion to an interface between the JFET upper portion and the SBD portion is within a range of not less than 0.1 μm and not more than 0.5 μm.

[0016] In addition, in the silicon carbide semiconductor device disclosed above, the first conductivity type impurity is nitrogen.

[0017] According to the above disclosure, when a MOS-type silicon carbide semiconductor device having an insulated gate with a three-layer structure of metal-oxide-semiconductor is in operation, a current flows through the portion above the JFET with a relatively high impurity concentration, so that an increase in on-resistance can be suppressed. When a load is short-circuited, the leakage current flowing through the SBD (Schottky barrier diode) is suppressed by the SBD portion with a relatively low impurity concentration, so that short-circuit resistance is improved. Effect of the Invention

[0018] Advantageous Effects of Invention The silicon carbide semiconductor device according to the present disclosure exhibits the effect of suppressing an increase in on-resistance and improving short-circuit withstand capability. [Brief description of the drawings]

[0019] [Figure 1] 1 is a cross-sectional view showing a structure of a silicon carbide semiconductor device in accordance with a first embodiment. [Diagram 2] 4 is a characteristic diagram showing a simulation result of an IV waveform when the built-in diode of the silicon carbide semiconductor device according to the first embodiment is operating. FIG. [Diagram 3] 4 is a characteristic diagram showing a result of simulating the relationship between the n-type impurity concentration and the specific on-resistance in the SBD portion of the silicon carbide semiconductor device according to the first embodiment. FIG. [Figure 4] 4 is a characteristics diagram showing a result of a simulation of the relationship between the n-type impurity concentration in the SBD portion of the silicon carbide semiconductor device according to the first embodiment and the SBD leakage current density. FIG. [Diagram 5] 4 is a characteristic diagram showing a result of simulating the relationship between the n-type impurity concentration and the inflection-point current density in the SBD portion of the silicon carbide semiconductor device according to the first embodiment. FIG. [Figure 6] 4 is a characteristics diagram showing a result of a simulation of the relationship between the width of the SBD portion of the silicon carbide semiconductor device according to the first embodiment and the SBD leakage current density. FIG. [Figure 7]4 is a characteristic diagram showing a result of simulating the relationship between the width of the SBD portion of the silicon carbide semiconductor device according to the first embodiment and the inflection point current density. FIG. [Figure 8] FIG. 11 is a cross-sectional view showing a structure of a silicon carbide semiconductor device according to a second embodiment. [Figure 9] FIG. 11 is a characteristic diagram showing the relationship between the ionization rate of aluminum and temperature for the silicon carbide semiconductor device according to the second embodiment. [Figure 10] FIG. 11 is a characteristic diagram showing the relationship between n-type compensation concentration and element temperature in the silicon carbide semiconductor device according to the second embodiment. [Figure 11] FIG. 11 is a characteristic diagram showing a result of simulating the temperature characteristic of an ionization rate of a p-type impurity for the silicon carbide semiconductor device according to the second embodiment. [Figure 12] 1 is a cross-sectional view showing a structure of a silicon carbide semiconductor device of a reference example. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0020] Preferred embodiments of the silicon carbide semiconductor device according to the present disclosure will be described in detail below with reference to the accompanying drawings. In this specification and the accompanying drawings, in a layer or region prefixed with n or p, electrons or holes are the majority carriers, respectively. In addition, + and - attached to n or p respectively mean that the impurity concentration is higher and lower than that of a layer or region not prefixed with that. In the following description of the embodiments and the accompanying drawings, the same reference numerals are attached to similar configurations, and duplicated explanations are omitted.

[0021] (overview) When the MOSFET is used as a switching element (upper and lower arms) of a bridge-type inverter circuit, for example, an SBD (Schottky barrier diode) built into the same semiconductor substrate as the MOSFET suppresses the parasitic pn diode operation of the MOSFET during dead time to prevent the upper and lower arms from turning on simultaneously when the upper and lower arms are repeatedly turned on and off alternately, thereby suppressing deterioration of the forward direction characteristics of the parasitic pn diode and reducing reverse recovery loss. In addition, by suppressing the parasitic pn diode operation of the MOSFET by the built-in SBD, a low on-resistance is achieved.

[0022] First, a reference example of a trench-gate MOSFET having an SBD built in the same semiconductor substrate will be described. Fig. 12 is a cross-sectional view showing the structure of a silicon carbide semiconductor device of the reference example. The silicon carbide semiconductor device 110 of the reference example shown in Fig. 12 is a trench-gate MOSFET in which a gate electrode 109 is embedded in some trenches (hereinafter referred to as gate trenches) 107a of a plurality of trenches 107 provided on the front surface side of a semiconductor substrate 130 via a gate insulating film 108, and an SBD (hereinafter referred to as trench SBD) 120 is embedded in the remaining trenches (hereinafter referred to as Schottky trenches) 107b excluding the gate trenches 107a.

[0023] The semiconductor substrate 130 is + The n-type drain region 101 + On the starting substrate 131 - The silicon carbide layers 132, 133 which become the p-type drift region 102 and the p-type base region 104 are epitaxially grown in sequence. The trenches 107 are provided in a striped pattern in a first direction X parallel to the front surface of the semiconductor substrate 130. The gate trenches 107a and the Schottky trenches 107b are alternately and repeatedly arranged in a second direction Y which is parallel to the front surface of the semiconductor substrate 130 and perpendicular to the first direction X.

[0024] The gate trench 107a is formed in a depth direction Z from the front surface of the semiconductor substrate 130 to a depth +The gate trench 107a is provided with a gate electrode 109 through a gate insulating film 108. The Schottky trench 107b is formed in the gate trench 107a from the front surface of the semiconductor substrate 130 to the p-type source region 105 and the p-type base region 104. ++ The Schottky trench 107b is formed with a source electrode 114. The source electrode 114 penetrates the p-type contact region 106 and the p-type base region 104 to reach the n-type current diffusion region 103. A Schottky electrode 117 is selectively provided along the sidewall of the Schottky trench 107b. A source electrode 114 is embedded on the Schottky electrode 117 inside the Schottky trench 107b.

[0025] Between the front surface (the main surface on the p-type silicon carbide layer 133 side) of the semiconductor substrate 130 and the p-type base region 104, and in contact with the p-type base region 104, + type source region 105 and p ++ The p-type base region 104 and the n-type contact region 106 are selectively provided. - Between the n-type drift region 102, an n-type current diffusion region 103 and a p-type + The n-type current diffusion regions 103 are selectively provided between adjacent p + Between the type regions 111, p + It is provided in contact with the mold region 111 .

[0026] The n-type current diffusion region 103 is + The bottom surface (n + The surface on the side of the n-type drain region 101 - The n-type current diffusion region 103 contacts the n-type drift region 102. The n-type current diffusion region 103 extends on both sides in the second direction Y, reaching the gate trench 107a on one side and the Schottky trench 107b on the other side. The n-type current diffusion region 103 contacts the gate insulating film 108 on the sidewall of the gate trench 107a. The n-type current diffusion region 103 contacts the Schottky electrode 117 on the sidewall of the Schottky trench 107b.

[0027] At least the portion between the adjacent trenches 107 (hereinafter, referred to as the JFET upper portion) 103a of the n-type current diffusion region 103 is formed to have a thickness of 1×10 17 / cm 3 It is doped with nitrogen (N) at an impurity concentration of about p + The n-type region 111 is located closer to the bottom of the trench 107 than the n-type region 111. + The p-type base region 104 is located deep on the p-type drain region 101 side, away from the p-type base region 104, and in contact with the n-type current diffusion region 103. + A plurality of mold regions 111 are provided facing the bottom surface of each trench 107 .

[0028] The interlayer insulating film 112 is provided on the entire front surface of the semiconductor substrate 130 so as to cover the gate electrode 109. The source electrode 114 is provided on the front surface of the semiconductor substrate 130 via an ohmic electrode 113. + type source region 105 and p ++ The source electrode 114 is electrically connected to the p-type contact region 106 on the inner wall of the Schottky trench 107b. ++ The drain electrodes 115 and 116 are in contact with the n-type contact region 106 and are electrically connected to the n-type current diffusion region 103 via a Schottky electrode 117. + The mold is provided on the surface of the starting substrate 131.

[0029] The trench SBD 120 is a diode that utilizes the rectification of a Schottky barrier formed at the junction between the Schottky electrode 117 on the sidewall of the Schottky trench 107b and the JFET upper portion 103a, and is formed along the sidewall of the Schottky trench 107b. + Type regions 111 and n - The pn junction between the n-type drift region 102 and the n-type current diffusion region 103 becomes conductive earlier than the parasitic pn diode (body diode) formed by the pn junction with the n-type drift region 102 and the n-type current diffusion region 103, thereby suppressing the parasitic pn diode operation of the MOSFET.

[0030] However, in the silicon carbide semiconductor device 110 described above (MOSFET incorporating trench SBD 120), the current flowing between the drain and source flows into a load connected to the circuit (a load that is turned on and off by the MOSFET), generating a load voltage. If a load short occurs in this state, a drain current Id flows through the MOSFET while it remains on, and a drain-source voltage Vds is applied from the voltage source of the circuit. This causes n + Type drain region 101 to n - A short-circuit current flows through the n-type drift region 102 and the n-type current diffusion region 103 toward the gate electrode 109, causing a temperature rise along the path of the short-circuit current.

[0031] When the element temperature (temperature of the semiconductor substrate 130) rises due to the short-circuit current, even if the MOSFET is turned off, + Type drain region 101 to n - A leakage current flows through the trench SBD 120 via the n-type drift region 102 and the n-type current diffusion region 103. The leakage current flowing through the trench SBD 120 increases due to thermal field emission, causing a temperature rise along the leakage current path, which leads to thermal destruction of the MOSFET. The above-mentioned Patent Document 1 discloses that the Schottky barrier height between the metal layer and the drift layer is set to be 1.76 eV or more and 3.10 eV or less, thereby reducing the leakage current due to thermal field emission that accompanies a temperature rise during a load short circuit.

[0032] The above Patent Document 2 discloses that for a normal MOSFET without an SBD, the temperature dependency of the ionization rate of aluminum (Al), a p-type impurity co-doped (added) in the upper part of an n-type JFET, is utilized to lower the carrier concentration in the upper part of the JFET at high temperatures caused by a load short circuit, thereby increasing the JFET resistance and suppressing the short circuit current. The above Non-Patent Document 1 discloses that the short circuit withstand capability (the time from load short circuit to element destruction) of a trench-gate MOSFET with an SBD is 6.0 μs, which is as low as 60% of the short circuit withstand capability (=10.0 μs) of a normal MOSFET without an SBD.

[0033] Non-Patent Document 2 discloses that, for a normal MOSFET without an SBD, when the difference in concentration (=Nd-Na) between the donor concentration Nd and the acceptor concentration Na in the upper part of the JFET becomes large, the electric field applied to the gate insulating film when the MOSFET is off increases at the contact point between the gate insulating film and the upper part of the JFET. Non-Patent Document 3 discloses that, for an SBD formed at the junction between a metal layer and an n-type silicon carbide layer, the lower the n-type impurity concentration of the n-type silicon carbide layer, the more likely it is that a depletion layer will spread from the junction between the metal layer and the n-type silicon carbide layer when a reverse bias is applied, reducing the electric field strength at the junction and thereby reducing the leakage current flowing through the SBD.

[0034] As a result of extensive research, the present inventors have found that in silicon carbide semiconductor device 110 of the above-mentioned reference example (MOSFET incorporating trench SBD 120: see FIG. 12), the adverse effects of etching damage on the inner wall of Schottky trench 107b cause Fermi level pinning at the interface (metal-semiconductor interface) of the Schottky junction of trench SBD 120, leading to variations in the Schottky barrier height and thus variations in short-circuit resistance. Problems to be solved in this embodiment include suppressing an increase in on-resistance by incorporating an SBD, and improving short-circuit resistance by reducing the leakage current flowing through the SBD.

[0035] (Embodiment 1) The structure of a silicon carbide semiconductor device according to a first embodiment will be described with reference to the cross-sectional view of Fig. 1. The silicon carbide semiconductor device 10 shown in Fig. 1 is a trench-gate MOSFET in which, of a plurality of trenches 7 provided on the front surface side of a semiconductor substrate 30 using silicon carbide (SiC) as a semiconductor material, a gate electrode 9 is embedded via a gate insulating film 8 in some of the trenches (gate trenches: first trenches) 7a, and an SBD (trench SBD) 20 is embedded in the remaining trenches (Schottky trenches: second trenches) 7b excluding the gate trenches 7a.

[0036] FIG. 1 shows two unit cells among a plurality of adjacent unit cells (components of an element) of the same structure (trench gate structure) arranged in an active region. An edge termination region surrounding the active region is omitted in FIG. 1 (as in FIG. 2). The active region is a region through which current flows when the MOSFET is in an on-state. The edge termination region is a region between the active region and the side surface of a semiconductor substrate (semiconductor chip) 30, in which a breakdown voltage structure is arranged to maintain a breakdown voltage by mitigating the electric field on the front surface side of the semiconductor substrate 30. The breakdown voltage is the maximum voltage at which a semiconductor device does not malfunction or break down.

[0037] The semiconductor substrate 30 is a n-type semiconductor substrate using silicon carbide as a semiconductor material. + On the front surface of the starting substrate 31, - The silicon carbide layers 32 and 33 that become the n-type drift region (first semiconductor region) 2 and the p-type base region (second semiconductor region) 4 are epitaxially grown in sequence. The semiconductor substrate 30 has a main surface on the p-type silicon carbide layer 33 side as a front surface, and an n-type + The main surface (n + The back side of the starting substrate 31 is referred to as the back side. + The starting substrate 31 is n + This becomes the type drain region 1. A trench gate structure is provided on the front surface side of the semiconductor substrate 30 in the active region.

[0038] The trench gate structure includes a p-type base region 4, a n + type source region (third semiconductor region) 5, p ++ The semiconductor substrate 30 is made up of a gate contact region 6, a gate trench 7a, a gate insulating film 8, and a gate electrode 9. The trenches 7 are provided, for example, in a stripe pattern in a first direction X parallel to the front surface of the semiconductor substrate 30. The gate trenches 7a and Schottky trenches 7b are alternately and repeatedly arranged in a second direction Y parallel to the front surface of the semiconductor substrate 30 and perpendicular to the first direction X.

[0039] The gate trench 7a is formed in a depth direction Z from the front surface of the semiconductor substrate 30 to n +The Schottky trench 7b penetrates the p-type source region 5 and the p-type base region 4 and terminates inside an n-type current diffusion region (fourth semiconductor region) 3 described below. The n-type current diffusion region 3 includes a JFET upper portion 3a, a JFET lower portion 3b, and an SBD portion 21 described below. A gate electrode 9 is provided inside the gate trench 7a with a gate insulating film 8 interposed therebetween. The Schottky trench 7b extends in the depth direction Z from the front surface of the semiconductor substrate 30 to the p-type source region 5 and the p-type base region 4. ++ The Schottky trench 7b is formed with a source electrode (first electrode) 14. The Schottky trench 7b is formed with a Schottky electrode 17. The Schottky electrode 17 passes through the p-type contact region 6 and the p-type base region 4 and terminates inside the n-type current diffusion region 3. A Schottky electrode 17 is selectively provided along the sidewall of the Schottky trench 7b. A source electrode (first electrode) 14 is embedded on the Schottky electrode 17 inside the Schottky trench 7b.

[0040] n + Type source region 5 and p ++ The n-type contact region 6 is selectively provided between the front surface of the semiconductor substrate 30 and the p-type base region 4 and in contact with the p-type base region 4. + Type source region 5 and p ++ The n-type contact region 6 is a diffusion region formed by ion implantation of a p-type impurity (acceptor) such as aluminum (Al) into the p-type silicon carbide layer 33. + Type source region 5 and p ++ The mold contact region 6 is exposed on the front surface of the semiconductor substrate 30. Exposing on the front surface of the semiconductor substrate 30 means that the front surface of the semiconductor substrate 30 is in contact with the ohmic electrode 13.

[0041] n + The n-type source region 5 contacts the gate insulating film 8 on the side wall of the gate trench 7a. + The source region 5 is provided away from the Schottky trench 7b. ++ The contact region 6 is provided away from the gate trench 7a. ++ The contact region 6 contacts the source electrode 14 on the side wall of the Schottky trench 7b. ++ The p-type contact region 6 may not be provided. ++Instead of the n-type contact region 6, the p-type base region 4 is exposed on the front surface of the semiconductor substrate 30. + Type source region 5 and p ++ The portion excluding the contact region 6 is the p-type base region 4 .

[0042] p-type base region 4 and n - Between the n-type drift region 2, an n-type current diffusion region 3 and a p-type + The n-type current diffusion region 3 is a so-called current spreading layer (CSL) that reduces the spreading resistance of carriers. - The n-type current diffusion region 3 is a diffusion region formed by ion implantation into the p-type silicon carbide layer 32. + Type region 11 than n + A portion 3a on the side of the source region 5 (hereinafter referred to as the upper portion of the JFET) and a portion 3b on the side of the source region 5 in the second direction Y + The JFET upper portion 3a, the JFET lower portion 3b, and the SBD portion 21 have a lower n-type impurity concentration than the above portions. - The n-type impurity concentration in the n-type drift region 2 is higher than that in the n-type drift region 3.

[0043] The JFET upper part 3a has a top surface (n + The bottom surface (n + The JFET upper portion 3a contacts the JFET lower portion 3b and a first high concentration region 11a and a second high concentration region 11b described below at its lower surface (the surface facing the n-type drain region 1). The JFET upper portion 3a extends on both sides in the second direction Y, reaching the gate trench 7a on one side and the SBD portion 21 on the other side. The JFET upper portion 3a contacts the gate insulating film 8 at the side walls of the gate trench 7a. The JFET lower portion 3b is a portion between the first high concentration region (fifth semiconductor region) 11a and the second high concentration region (sixth semiconductor region) 11b adjacent to each other, and contacts these regions in the second direction Y. -The JFET upper portion 3a and the JFET lower portion 3b are in contact with the first-type drift region 2. The JFET upper portion 3a and the JFET lower portion 3b are doped with nitrogen (N), which is an n-type impurity (donor: first conductivity type impurity), as a dopant (impurity added to a semiconductor).

[0044] The n-type impurity concentration (nitrogen concentration) of the JFET upper portion 3a and the JFET lower portion 3b is, for example, 1×10 17 / cm 3 The n-type impurity concentration of the JFET upper part 3a and the JFET lower part 3b is, for example, 7×10 16 / cm 3 More than 7×10 17 / cm 3 It is desirable to keep it below the lower limit (=7×10 16 / cm 3 ) or more, an increase in on-resistance can be prevented. 17 / cm 3 ) or less, the electric field applied to the gate insulating film 8 when the MOSFET is off can be set to 3 MV / cm or less in consideration of long-term reliability (see Non-Patent Document 2 above).

[0045] The SBD portion 21 is provided between the JFET upper portion 3a and the Schottky trench 7b. The SBD portion 21 contacts the p-type base region 4 on the upper surface and contacts only the second high concentration region 11b on the lower surface. The SBD portion 21 does not contact the JFET lower portion 3b. The SBD portion 21 contacts the JFET upper portion 3a on one side in the second direction Y and reaches the Schottky trench 7b on the other side in the second direction Y. The SBD portion 21 is a SiC semiconductor region (hatched portion) that contacts the Schottky electrode 17 on the side wall of the Schottky trench 7b to form a trench SBD 20 described later. The SBD portion 21 has a function of extending a depletion layer from the junction surface with the Schottky electrode 17 into the semiconductor substrate 30 when the load is short-circuited, thereby suppressing a leakage current flowing into the trench SBD 20.

[0046] The SBD portion 21 is doped with nitrogen, which is an n-type impurity (donor: first conductivity type impurity), as a dopant. The n-type impurity concentration (nitrogen concentration) of the SBD portion 21 is in the range of 10% or more and less than 100% of the n-type impurity concentration of the JFET upper portion 3a. The carrier concentration of the SBD portion 21 is lower than the n-type impurity concentration of the JFET upper portion 3a. The carrier concentration is the concentration of charge carriers (charged particles) that are activated and responsible for electrical conduction among the impurities (dopants) added to a semiconductor. Nitrogen as an n-type impurity is activated almost 100% in the assumed temperature range in a silicon carbide semiconductor and emits electrons that are charge carriers of the n-type semiconductor. Therefore, the carrier concentration of the n-type SBD portion 21 is the concentration of activated electrons, but can be approximated by the concentration of nitrogen, which is an n-type impurity.

[0047] The width W1 of the SBD portion 21 is within a range in which the entire lower surface of the SBD portion 21 is in contact with the second high-concentration region 11b and the SBD portion 21 is not in contact with the JFET lower portion 3b. The width W1 of the SBD portion 21 is the width in the second direction Y from the junction surface between the Schottky electrode 17 and the SBD portion 21 to the interface between the JFET upper portion 3a and the SBD portion 21. Since the SBD portion 21 is not in contact with the JFET lower portion 3b, the SBD portion 21 does not become a current path for the main current that flows through the channel during MOSFET operation (when the MOSFET is on). This makes it possible to suppress an increase in the on-resistance of the MOSFET (see FIG. 3 described later).

[0048] On the other hand, when a load (a load that is turned on and off by the MOSFET) is short-circuited, the SBD portion 21 becomes a current path for a leakage current flowing through the trench SBD 20. For example, in the MOSFET of the reference example (silicon carbide semiconductor device 110: see FIG. 12), when the element temperature (for example, about 250° C. to 500° C.) rises due to a load short-circuit, the leakage current flowing through the trench SBD 120 increases after the MOSFET is turned off, but in the first embodiment, the n-type impurity concentration of the SBD portion 21 in the n-type current diffusion region 3 is relatively low as described above, thereby increasing the conduction resistance of the trench SBD 20 and reducing the leakage current flowing through the trench SBD 20 when the load is short-circuited (see FIG. 4 described later).

[0049] The lower the n-type impurity concentration of the SBD portion 21, the greater the effect of reducing the leakage current flowing through the trench SBD 20 when the load is short-circuited. This suppresses heat generation (temperature rise in the semiconductor substrate 30) after the gate is turned off (after the voltage applied to the gate electrode 9 becomes less than the gate threshold voltage) in the MOSFET (silicon carbide semiconductor device 10), improving the short-circuit resistance. On the other hand, the lower the n-type impurity concentration of the SBD portion 21, the greater the conduction resistance of the trench SBD 20. As a result, the parasitic pn diode (body diode) of the MOSFET begins to operate at a relatively low current density before the trench SBD 20 becomes conductive, causing degradation of the forward direction characteristics of the parasitic pn diode and an increase in reverse recovery loss.

[0050] The conduction resistance of the trench SBD 20 is the resistance (differential resistance) at the conduction (start of operation) point of the IV (current-voltage) characteristic of the trench SBD 20. The parasitic pn diode of the MOSFET is the p-type base region 4 and the p + Type region 11 (first and second high concentration regions 11a and 11b) and n - The parasitic pn diode is a built-in diode formed in the semiconductor substrate 30 by a pn junction with the n-type drift region 2 and the n-type current diffusion region 3 (JFET upper portion 3a, JFET lower portion 3b, SBD portion 21). The forward current density at which the parasitic pn diode of the MOSFET starts to operate (conducts) can be obtained based on the IV characteristics of the built-in diode of the semiconductor substrate 30. The result of simulating the IV waveform of the parasitic pn diode of the semiconductor substrate 30 is shown in FIG. 2.

[0051] 2 is a characteristic diagram showing a simulation result of an IV waveform during operation of the built-in diode (parasitic pn diode) of the silicon carbide semiconductor device according to the first embodiment. The horizontal axis of FIG. 2 is a positive voltage (source-drain voltage Vsd) applied to source electrode 14 with respect to drain electrode 15, which corresponds to the forward voltage of the parasitic pn diode. The vertical axis of FIG. 2 corresponds to the forward current density of the parasitic pn diode. FIG. 2 shows the n-type impurity concentration of SBD portion 21 (shown as the concentration of the SBD portion in FIG. 2) of 1.1×10 17 / cm 3The IV waveform 41 (solid line) of the parasitic pn diode when the n-type impurity concentration of the SBD portion 21 is set to 1.8×10 16 / cm 3 12 shows an IV waveform 42 (dashed line) of the parasitic pn diode when the n-type impurity concentrations of the JFET upper portion 3a and the SBD portion 21 are equal (i.e., the silicon carbide semiconductor device 110 of the reference example in FIG. 12) are the characteristics (comparative example).

[0052] As shown in FIG. 2, the n-type impurity concentration of the SBD portion 21 is set to 1.1×10 17 / cm 3 In this case, the IV waveform 41 of the parasitic pn diode shows that the forward current density increases with an increase in the forward voltage. 16 / cm 3 When it is lowered to 750A / cm 2 At a forward current density of about 1000 mA, the IV waveform 42 of the parasitic pn diode is bent, and the parasitic pn diode begins to operate, forming the upper portion 3a of the JFET, the lower portion 3b of the JFET, and the n - It can be seen that minority carriers (holes) are injected into the type drift region 2, increasing the forward current density of the parasitic pn diode, and the conduction resistance of the parasitic pn diode decreases at the inflection point 42a of the forward current density (conductivity modulation).

[0053] Hereinafter, the forward current density at the inflection point of the IV waveform of the parasitic pn diode of a MOSFET is called the inflection point current density, and is an index showing the ease of operation of the parasitic pn diode. The inflection point current density of the IV waveform of the parasitic pn diode of a MOSFET is at least the rated current density of the MOSFET (for example, 500 A / cm 2 Due to the above constraints, the n-type impurity concentration of the SBD portion 21 is preferably, for example, 1×10 16 / cm 3 More than 1×10 17 / cm 3(See FIG. 5, which will be described later.) Here, the element temperature (temperature of the semiconductor substrate 30) was set to 175° C., but the relationship between the n-type impurity concentration in the SBD portion 21 and the forward current density of the parasitic pn diode shows the same tendency regardless of the element temperature (See FIG. 4, which will be described later).

[0054] p + The type region 11 is n - The p-type impurity diffusion region is formed by ion implantation of p-type impurities such as aluminum into the p-type silicon carbide layer 32. + The n-type region 11 is electrically connected to the source electrode 14 at a portion not shown in the figure, and has the function of depleting the n-type region 11 (or depleting the n-type current diffusion region 3, or both) when the MOSFET is turned off, thereby reducing the electric field applied to the bottom surface of the gate trench 7a. + The n-type region 11 is located closer to the bottom of the trench 7 than the n-type region 11. + The p-type drain region 1 is provided at a deep position away from the p-type base region 4. + A plurality of mold regions 11 are provided facing the bottom surface of each trench 7 .

[0055] Specifically, p + The mold region 11 is composed of a first high concentration region 11a below the gate trench 7a and a second high concentration region 11b below the Schottky trench 7b. The width (width in the second direction Y) W21 of the first high concentration region 11a is wider than the width (width in the second direction Y) W11 of the gate trench 7a, and the first high concentration region 11a protrudes in a direction away from both side walls of the gate trench 7a in parallel to the second direction Y. The width (width in the second direction Y) W22 of the second high concentration region 11b is wider than the width (width in the second direction Y) W12 of the Schottky trench 7b, and the second high concentration region 11b protrudes in a direction away from both side walls of the Schottky trench 7b in parallel to the second direction Y.

[0056] The width of the second high concentration region 11b protruding in the direction parallel to the second direction Y away from the sidewall of the Schottky trench 7b is wider than the width W1 of the SBD portion 21. That is, the second high concentration region 11b contacts the entire lower surface of the SBD portion 21 and contacts the JFET upper portion 3a. The second high concentration region 11b contacts the entire lower surface of the SBD portion 21, so that the SBD portion 21 and the JFET lower portion 3b are spaced apart from each other. The width W22 of the second high concentration region 11b may be wider than the width W21 of the first high concentration region 11a or may be the same as the width W21 of the first high concentration region 11a. In FIG. 1, the relationship between the width W22 of the second high concentration region 11b, the width W1 of the SBD portion 21, and the width W12 of the Schottky trench 7b is W22>2·W1+W12.

[0057] n - The n-type current diffusion region 3 and the p-type silicon carbide layer 32 + The part excluding the mold region 11 is n - The interlayer insulating film 12 is disposed on the entire front surface of the semiconductor substrate 30 so as to cover the gate electrode 9. + Type source region 5, p ++ The type contact region 6 and the Schottky trench 7b are exposed. The ohmic electrode 13 is in ohmic contact with the front surface of the semiconductor substrate 30 through a contact hole in the interlayer insulating film 12. The ohmic electrode 13 is, for example, a nickel silicide (NixSiy, where x and y are positive numbers) electrode.

[0058] The source electrode 14 is provided over substantially the entire front surface of the semiconductor substrate 30 in the active region so as to fill the contact hole and the Schottky trench 7b in the interlayer insulating film 12. The source electrode 14 is connected to the front surface of the semiconductor substrate 30 via the ohmic electrode 13. + Type source region 5 and p ++ The source electrode 14 is electrically connected to the p-type contact region 6 on the side wall of the Schottky trench 7b. ++The Schottky electrode 17 is in contact with the n-type contact region 6 and is electrically connected to the n-type current diffusion region 3 (the JFET upper portion 3 a, the JFET lower portion 3 b, and the SBD portion 21 ).

[0059] The Schottky electrode 17 is provided along the sidewall of the Schottky trench 7b between the sidewall of the Schottky trench 7b and the source electrode 14. The Schottky electrode 17 contacts the SBD portion 21 on the sidewall of the Schottky trench 7b. The Schottky electrode 17 may extend from the sidewall to the bottom surface of the Schottky trench 7b. The Schottky electrode 17 is provided along the inner wall of the Schottky trench 7b between the second highly doped region 11b, the p-type base region 4 and the p-type base region 5. ++ The Schottky electrode 17 may be in contact with the type contact region 6. The material of the Schottky electrode 17 is, for example, titanium (Ti).

[0060] The front surface of the semiconductor substrate 30 is protected by a passivation film. The portion of the source electrode 14 exposed through the opening of the passivation film is a source electrode pad. + A drain electrode (second electrode) 15 is provided on the entire surface of the back surface of the starting substrate 31. The drain electrode 15 is + Type drain region 1(n + ohmic contact with the starting substrate 31) + The drain electrode 15 is connected to the drain region 1. A drain electrode pad 16 is provided on the entire surface of the drain electrode 15.

[0061] The trench SBD 20 is a diode that utilizes the rectification of a Schottky barrier formed at the junction between the Schottky electrode 17 on the sidewall of the Schottky trench 7b and the SBD portion 21, and is formed along the sidewall of the Schottky trench 7b. + Type region 11 (first and second high concentration regions 11a, 11b), n-type current diffusion region 3 (JFET upper part 3a, JFET lower part 3b, SBD part 21) and n -It becomes conductive earlier than the parasitic pn diode (body diode) formed at the pn junction with the type drift region 2, and thus has the function of suppressing the parasitic pn diode operation of the MOSFET.

[0062] <Operation of Silicon Carbide Semiconductor Device According to First Embodiment> The operation of the silicon carbide semiconductor device 10 will now be described. When a voltage (gate voltage) equal to or greater than the gate threshold voltage is applied to the gate electrode 9 (during steady-state operation) with a positive voltage (drain-source voltage Vds: forward voltage of the MOSFET) applied to the drain electrode 15 relative to the source electrode 14, a channel (n-type inversion layer) is formed in the p-type base region 4 along the sidewall of the gate trench 7a. + From the drain region 1 through the channel + A main current (drift current) flows toward the type source region 5, and the MOSFET (silicon carbide semiconductor device 10) turns on.

[0063] On the other hand, when the gate voltage becomes less than the gate threshold voltage with the drain-source voltage Vds applied, the p-type base region 4 and the p + The n-type region 11 and the n-type current diffusion region 3 and - The pn junction (the main junction of the MOSFET) with the type drift region 2 is reverse biased, so that the main current does not flow and the MOSFET remains in the off state. Also, the depletion layer that spreads from the main junction of the MOSFET + As a result, the n-type region 11 (or the n-type current diffusion region 3, or both) is depleted, and the electric field applied to the gate insulating film 8 on the inner wall of the gate trench 7a is relaxed.

[0064] When the load is short-circuited, the reduced impurity concentration of SBD portion 21 makes it easier for a depletion layer to spread from the junction (Schottky junction) between Schottky electrode 17 and SBD portion 21 into n-type current diffusion region 3, reducing the electric field strength at the Schottky junction. This reduces the electric field applied to trench SBD 20, reducing the leakage current flowing through trench SBD 20. The n-type impurity concentration of SBD portion 21 can be controlled with good reproducibility by the dose of n-type impurity ion implantation, providing high controllability for the conduction resistance of trench SBD 20.

[0065] In addition, the p-type base region 4 and the p + Type region 11 and n - When a parasitic pn diode formed by a pn junction between the n-type drift region 2 and the n-type current diffusion region 3 is forward biased, the trench SBD 20 conducts at a lower voltage than the parasitic pn diode and earlier than the parasitic pn diode. Therefore, the parasitic pn diode of the MOSFET does not operate. As a result, there is no degradation of forward characteristics or reverse recovery loss due to the operation of the parasitic pn diode of the MOSFET.

[0066] <Method of Manufacturing Silicon Carbide Semiconductor Device According to First Embodiment> A method for manufacturing silicon carbide semiconductor device 10 will be described. + n-type drain region 1 + On the front surface of the starting substrate (semiconductor wafer) 31, - For example, a nitrogen-doped n-type drift region 2 is formed. - The n-type silicon carbide layer 32 is epitaxially grown (deposited). At this stage, - The n-type silicon carbide layer 32 (i.e., the n-type silicon carbide layer 32 of a predetermined thickness constituting the semiconductor substrate 30 of the product thickness) - The n-type silicon carbide layer 32) corresponds to the region where the JFET upper portion 3a and the SBD portion 21 are formed. - No silicon carbide layer is formed.

[0067] Next, by photolithography and selective ion implantation of p-type impurities such as aluminum (Al), -In the surface region of the silicon carbide layer 32, the first high concentration region 11a and the second high concentration region 11b are alternately and repeatedly formed in the second direction Y and spaced apart from each other. The n-type impurity regions 11a and 11b are formed by photolithography and selective ion implantation of an n-type impurity such as nitrogen. - In the surface region of the silicon carbide layer 32, an n-type JFET lower portion 3b is formed between the first high concentration region 11a and the second high concentration region 11b adjacent to each other. The n-type impurity concentration of the JFET lower portion 3b is, for example, 1×10 17 / cm 3 It is said to be about that level.

[0068] Further epitaxial growth leads to n - The additional epitaxially grown n-type silicon carbide layer 32 is then grown to a predetermined thickness. - The n-type impurity concentration of the thickened portion of the silicon carbide layer 32 is set to a nitrogen concentration (e.g., 1.8×10 16 / cm 3 This n - The thickened portion (n - The n-type silicon carbide layer in FIG. - This corresponds to the region of the n-type silicon carbide layer 32 where the JFET upper portion 3a and the SBD portion 21 are formed. - The thickness of the thickened portion of the silicon carbide layer 32 is set to the same thickness as the JFET upper portion 3a and the SBD portion 21. At this stage, the n-type silicon carbide layer 32 shown in FIG. - An entire silicon carbide layer 32 is formed.

[0069] Next, by photolithography, - An ion implantation mask having an opening corresponding to a region for forming the JFET upper portion 3a is formed on the surface of the silicon carbide layer 32. Next, the n-type impurity such as nitrogen is ion-implanted using the ion implantation mask to form an n-type silicon carbide layer 32. - The JFET upper portion 3a is selectively formed in the thickened portion of the silicon carbide layer 32 to a depth reaching the JFET lower portion 3b. The n-type impurity concentration of the JFET upper portion 3a is, for example, 1×10 17 / cm 3After that, the ion implantation mask is removed.

[0070] n - Of the thickened portion of the silicon carbide layer 32, the portion that is not ion-implanted and remains at the n-type impurity concentration at the time of epitaxial growth due to being covered by the ion implantation mask becomes the SBD portion 21. The n-type impurity concentration of the SBD portion 21 may be adjusted by additionally implanting ions of an n-type impurity such as nitrogen to form the SBD portion 21. In this case, - An ion implantation mask having an opening corresponding to the region where the SBD portion 21 is to be formed is formed on the surface of the silicon carbide layer 32, and ions of n-type impurities are implanted using the mask. The order of forming the JFET upper portion 3a and the SBD portion 21 may be reversed.

[0071] Next, n - On the n-type silicon carbide layer 32, for example, an aluminum-doped p-type silicon carbide layer 33 that will become the p-type base region 4 is epitaxially grown (deposited). - The n-type silicon carbide layer 32 contacts the JFET upper portion 3a and the SBD portion 21 formed in the surface region of the n-type silicon carbide layer 32. + A semiconductor base (semiconductor wafer) 30 is produced (manufactured) by laminating silicon carbide layers 32 and 33 in this order on a mold starting substrate 31.

[0072] Next, n-type silicon carbide is formed in the p-type silicon carbide layer 33 in the surface region of the front surface of the semiconductor substrate 30 by photolithography and ion implantation. + Type source region 5 and p ++ Then, a contact region 6 is selectively formed in each of the p-type silicon carbide layers 32 and 33. A portion of the p-type silicon carbide layer 33 that is not ion-implanted and that remains at the same impurity concentration as during epitaxial growth becomes the p-type base region 4. Next, a heat treatment is performed to activate the impurity ions implanted into the silicon carbide layers 32 and 33.

[0073] Next, the trenches 7 (gate trench 7a and Schottky trench 7b) are formed by, for example, dry etching. The gate trench 7a is formed on the front surface of the semiconductor substrate 30.+ The gate trench 7a penetrates the p-type source region 5 and the p-type base region 4 to reach the JFET upper portion 3a and faces the first high concentration region 11a in the depth direction Z. The JFET upper portion 3a may be interposed between the bottom surface of the gate trench 7a and the first high concentration region 11a, or the gate trench 7a may penetrate the JFET upper portion 3a in the depth direction Z and terminate inside the first high concentration region 11a.

[0074] The Schottky trench 7b is formed on the front surface of the semiconductor substrate 30. ++ The Schottky trench 7b penetrates the p-type contact region 6 and the p-type base region 4 to reach the SBD portion 21 and faces the second high concentration region 11b in the depth direction Z. The SBD portion 21 may be interposed between the bottom surface of the Schottky trench 7b and the second high concentration region 11b, or the Schottky trench 7b may penetrate the SBD portion 21 in the depth direction Z and terminate inside the second high concentration region 11b.

[0075] Next, a gate electrode 9 is formed inside the gate trench 7a via a gate insulating film 8 by a general method. Next, an interlayer insulating film 12 is formed on the entire front surface of the semiconductor substrate 30. Next, a contact hole is formed in the interlayer insulating film 12 in the depth direction Z to reach the front surface of the semiconductor substrate 30. The contact hole in the interlayer insulating film 12 is filled with n + Type source region 5, p ++ The mold contact region 6 and the Schottky trench 7b are exposed.

[0076] Next, a typical method is used to form an n-type ion implantation electrode in the contact hole of the interlayer insulating film 12. + Type source region 5 and p ++ An ohmic electrode 13 is formed in ohmic contact with the type contact region 6. A Schottky electrode 17 is also formed on the side wall of the Schottky trench 7b in contact with the SBD portion 21. This forms a trench SBD 20 that utilizes the rectification of a Schottky barrier formed at the junction between the Schottky electrode 17 and the SBD portion 21.

[0077] Next, a source electrode 14 is formed on the front surface of the semiconductor substrate 30 so as to fill the contact hole of the interlayer insulating film 12 and the Schottky trench 7b. As a result, the source electrode 14 is connected to the Schottky electrode 17 inside the Schottky trench 7b. In addition, a drain electrode 15 is formed on the back surface of the semiconductor substrate 30, and a drain electrode pad 16 is formed on the drain electrode 15. Thereafter, the semiconductor wafer is diced (cut) into individual chips, thereby completing the silicon carbide semiconductor device 10 shown in FIG.

[0078] <Simulation Verification of Silicon Carbide Semiconductor Device According to First Embodiment> The relationship between the n-type impurity concentration of the SBD portion 21 (see FIG. 1) described above and the element characteristics of the MOSFET was examined. The results of the examination by simulation are shown in FIGS. 3 to 7. In FIGS. 3 to 7, the simulation was performed using a MOSFET having the structure of the silicon carbide semiconductor device 10 according to the first embodiment shown in FIG. 1. In the MOSFET, the JFET upper portion 3a and the JFET lower portion 3b were doped with the same n-type impurity concentration (1×10 17 / cm 3 ) was kept constant.

[0079] 3 to 5, various characteristics were simulated by changing the n-type impurity concentration of the SBD portion 21 with the width W1 of the SBD portion 21 set to 0.5 μm. The horizontal axis of FIGS. 3 to 5 represents the n-type impurity concentration of the SBD portion 21. The vertical axis of FIG. 3 represents the characteristic on-resistance of the MOSFET (on-resistance that does not depend on the surface area of ​​the active region). The vertical axis of FIG. 4 represents the current density of the leakage current flowing through the trench SBD 20 (hereinafter referred to as the SBD leakage current density). The vertical axis of FIG. 5 represents the inflection point current density of the IV waveform of the parasitic pn diode of the MOSFET.

[0080] In FIG. 6 and FIG. 7, the n-type impurity concentration of the SBD portion 21 is set to 1.8×10 16 / cm 3Assuming that the width W1 of the SBD portion 21 is 0.05 V, the width W1 of the SBD portion 21 is varied to simulate various characteristics. The horizontal axis in Figures 6 and 7 is the width W1 of the SBD portion 21. The vertical axis in Figures 6 and 7 is the SBD leakage current density and the inflection point current density of the IV waveform of the MOSFET's parasitic pn diode, respectively. In Figure 3, the gate voltage Vg of the MOSFET is set to 20 V. In Figures 4 and 6, the gate voltage Vg is set to 0 V and the drain-source voltage Vds is set to 600 V. Figures 4 and 6 assume that a high voltage is applied to the MOSFET when the load is short-circuited.

[0081] 3, 5, and 7, the element temperature was set to 175°C. FIG. 4 shows simulation results when the element temperatures were 175°C, 250°C, and 500°C. In FIG. 6, the element temperature was set to 500°C. The element temperature is the junction temperature Tj at the main junction (pn junction) of the MOSFET, and corresponds to the temperature of the semiconductor substrate 30. FIGS. 3 to 7 show simulation results for a nitrogen-doped silicon carbide semiconductor, in which the n-type impurity concentration (nitrogen concentration) in SBD portion 21 is approximately equal to the carrier concentration in the semiconductor, and the temperature dependence of the forward characteristics of trench SBD 20 is also negligible.

[0082] 3 is a characteristics diagram showing the results of a simulation of the relationship between the n-type impurity concentration and the characteristic on-resistance of the SBD portion of silicon carbide semiconductor device 10. It was confirmed from the results shown in FIG. 3 that the characteristic on-resistance of the MOSFET remains almost unchanged even when the impurity concentration of SBD portion 21 is reduced. Therefore, a predetermined characteristic on-resistance of the MOSFET can be set regardless of the n-type impurity concentration of SBD portion 21.

[0083] 4 is a characteristic diagram showing the results of a simulation of the relationship between the n-type impurity concentration in the SBD portion of silicon carbide semiconductor device 10 and the SBD leakage current density. The results shown in FIG. 4 confirm that the SBD leakage current density when the MOSFET is off can be reduced by lowering the n-type impurity concentration of SBD portion 21. Furthermore, although the SBD leakage current density increases as the element temperature increases, the tendency of the SBD leakage current density reduction caused by lowering the impurity concentration of SBD portion 21 remains the same regardless of the element temperature. Therefore, in order to reduce the SBD leakage current density, it is advisable to lower the impurity concentration of SBD portion 21.

[0084] On the other hand, as described above, when the n-type impurity concentration of the SBD portion 21 is low, the conduction resistance of the trench SBD 20 increases, and the parasitic pn diode of the MOSFET tends to exceed the threshold voltage and become conductive (see FIG. 2). When the parasitic pn diode becomes conductive, the conduction resistance of the parasitic pn diode decreases due to conductivity modulation, and the slope of the IV waveform of the parasitic pn diode changes significantly. To obtain the effect of providing the trench SBD 20 (suppressing the operation of the parasitic pn diode of the MOSFET), the n-type impurity concentration of the SBD portion 21 should be set so that the trench SBD 20 becomes conductive at a forward voltage lower than the forward voltage at the inflection point current density of the IV waveform of the parasitic pn diode.

[0085] 5 is a characteristics diagram showing the results of a simulation of the relationship between the n-type impurity concentration and the inflection-point current density in the SBD portion of silicon carbide semiconductor device 10. From the results shown in FIG. 5, it was confirmed that lowering the n-type impurity concentration in SBD portion 21 increases the conduction resistance of trench SBD 20 and reduces the inflection-point current density of the IV waveform of the parasitic pn diode. To prevent the parasitic pn diode of the MOSFET from conducting before trench SBD 20 and adversely affecting the operation of the MOSFET during steady-state operation, the inflection-point current density of the IV waveform of the parasitic pn diode must be at least as high as the rated current density of the MOSFET (for example, 500 A / cm in the first embodiment). 2 It is desirable that the degree of cracking is equal to or higher than that shown in the dashed line in Figure 5.

[0086] Therefore, from the results shown in FIG. 5, the n-type impurity concentration of the SBD portion 21 is 1×10 16 / cm 3 For example, when the n-type impurity concentration of the JFET upper portion 3a is 1×10 17 / cm 3 (the same n-type impurity concentration as the rightmost simulation point in FIG. 5), the n-type impurity concentration of SBD portion 21 will be approximately 10% or more of the n-type impurity concentration of upper JFET portion 3a. That is, in order to reduce the SBD leakage current density and keep the inflection point current density of the IV waveform of the parasitic pn diode of the MOSFET equal to or higher than the rated current density of the MOSFET, the n-type impurity concentration of SBD portion 21 should be 10% or more and less than 100% of the n-type impurity concentration of upper JFET portion 3a.

[0087] Furthermore, the width W1 of the SBD portion 21 was examined. Fig. 6 is a characteristics diagram showing the results of a simulation of the relationship between the width of the SBD portion of the silicon carbide semiconductor device according to the first embodiment and the SBD leakage current density. The results of the simulation of the relationship between the width W1 of the SBD portion 21 and the SBD leakage current density when the MOSFET is off are shown by a solid line in Fig. 6. Also, Fig. 6 shows by a dashed line the SBD leakage current density when the silicon carbide semiconductor device 110 (see Fig. 12) of the reference example not including an SBD portion is off. It was confirmed from the results shown in Fig. 6 that the SBD leakage current density is reduced as the width W1 of the SBD portion 21 is wider.

[0088] In order to avoid adverse effects on the characteristic on-resistance of the MOSFET, it is desirable that the width W1 of the SBD portion is set to a range where the SBD portion 21 does not contact the JFET lower portion 3b at most (for example, about 0.5 μm or less in the first embodiment). The reason is that if the SBD portion 21 extends in the second direction Y away from the sidewall of the Schottky trench 7b to a range where the SBD portion 21 contacts the upper surface of the JFET lower portion 3b, the path of the main current of the MOSFET becomes narrower during steady operation, leading to an increase in the characteristic on-resistance of the MOSFET. In other words, the fact that the SBD portion 21 does not contact the JFET lower portion 3b means that the interface between the SBD portion 21 and the JFET upper portion 3a is located on the second high concentration region 11b side with respect to the interface between the JFET lower portion 3b and the second high concentration region 11b.

[0089] 7 is a characteristic diagram showing the result of simulating the relationship between the width of the SBD portion of the silicon carbide semiconductor device 10 and the inflection point current density. FIG. 7 shows the result of simulating the relationship between the width W1 of the SBD portion 21 and the inflection point current density of the IV waveform of the parasitic pn diode (see FIG. 2). From the result shown in FIG. 7, it was confirmed that the narrower the width W1 of the SBD portion 21, the higher the inflection point current density of the IV waveform of the parasitic pn diode. If the width W1 of the SBD portion 21 is too narrow, it becomes difficult to control the width W1, so the width W1 of the SBD portion 21 is desirably about 0.1 μm or more. Therefore, the width W1 of the SBD portion 21 is desirably about 0.1 μm or more and 0.5 μm or less (a range that does not contact the JFET lower portion 3b).

[0090] As described above, according to the first embodiment, the n-type current diffusion region has a JFET upper portion that contacts the gate insulating film at the sidewall of the gate trench, and an SBD portion that contacts the Schottky electrode at the sidewall of the Schottky trench. A trench SBD is formed along the sidewall of the Schottky trench, utilizing the rectification of the Schottky barrier formed at the junction between the Schottky electrode and the SBD portion. As a result, during steady-state operation of the MOSFET, the main current flows through the upper portion of the JFET with a relatively high impurity concentration, and the rise in specific on-resistance can be suppressed. During load short-circuiting, the leakage current flowing through the trench SBD is suppressed by the SBD portion with a relatively low impurity concentration. Therefore, the rise in specific on-resistance can be suppressed and the short-circuit resistance can be improved. The n-type impurity concentration of the SBD portion can be controlled with good reproducibility by the dose of ion implantation of the n-type impurity, so that the negative impact on the manufacturing yield caused by making the n-type impurity concentration of the SBD portion lower than the n-type impurity concentration of the upper portion of the JFET is small.

[0091] (Embodiment 2) The structure of a silicon carbide semiconductor device according to the second embodiment will be described with reference to the cross-sectional view of Fig. 8. The second embodiment differs from the first embodiment (see Fig. 1) in that SBD portion 51 (hatched portion) is doped with nitrogen, which is an n-type impurity (first conductivity type impurity), and is co-doped (added) with aluminum (Al), which is a p-type impurity (second conductivity type impurity). The configuration of SBD portion 51, other than the fact that it is co-doped with aluminum and the nitrogen concentration range described below, is the same as that of SBD portion 21 of the first embodiment, which is doped with only nitrogen.

[0092] In the second embodiment, the actual n-type impurity concentration of nitrogen acting as majority carriers (electrons) in the SBD portion 51 is compensated for by aluminum acting as minority carriers (holes), and the SBD portion 51 is an n-type region with a lower carrier concentration (lower compensation concentration) than the JFET upper portion 3a. The ionization rate of aluminum depends on temperature, and when the element temperature (for example, about 500°C) rises due to a short circuit of the load (load that is turned on and off by the MOSFET), the aluminum in the SBD portion 51 is ionized at a rate of nearly 100%, and the SBD portion 51 has a lower compensation concentration according to the concentration of holes released from the aluminum.

[0093] Trench SBD 52 is formed by utilizing the rectification of a Schottky barrier formed at the junction between Schottky electrode 17 and SBD portion 51. The configuration of trench SBD 52 other than SBD portion 51 is the same as trench SBD 20 of the first embodiment. That is, when a load is short-circuited, the n-type carrier concentration of SBD portion 51 is a concentration obtained by subtracting the aluminum concentration multiplied by the ionization rate from the nitrogen concentration of SBD portion 51. By lowering the compensation concentration of SBD portion 51 forming trench SBD 52, it is possible to reduce the leakage current flowing through trench SBD 52 when a load is short-circuited, as in the first embodiment (see Non-Patent Document 3 mentioned above).

[0094] The carrier concentration of the SBD portion 51 is a compensation concentration obtained by compensating the concentration of n-type impurities doped and activated in the SBD portion 51 with the concentration of p-type impurities co-doped and activated in the SBD portion 51. The carrier concentration of the SBD portion 51 is lower than the n-type impurity concentration of the JFET upper portion 3a, as in the SBD portion 21 in the first embodiment. Therefore, the second embodiment also has characteristics similar to those of the simulation results verified in the above-mentioned FIGS. 2 to 7. It is only necessary that the carrier concentration of the SBD portion 51 is lower than the n-type impurity concentration of the JFET upper portion 3a during load short circuit, and the SBD portion 51 and the JFET upper portion 3a may have substantially the same n-type impurity concentration.

[0095] The portion of the SBD portion 51 adjacent to the Schottky trench 7b has the function of reducing the electric field applied to the trench SBD 52. The nitrogen concentration of the SBD portion 51 is, for example, 1×10 16 / cm 3 More than 7×10 17 / cm 3 or less. If the nitrogen concentration in SBD portion 51 is less than the lower limit, the conduction resistance of trench SBD 52 increases and the inflection point current density of the IV waveform of the parasitic pn diode of MOSFET (silicon carbide semiconductor device 50) decreases, which is undesirable. The aluminum concentration in SBD portion 51 is, for example, in the range of about 10% or more (preferably 50% or more) but less than 100% of the nitrogen concentration in SBD portion 51.

[0096] The temperature variation of the ionization rate of the p-type impurity (acceptor) depends greatly on the depth of the acceptor level (see FIG. 9 described later). For this reason, the p-type impurity to be co-doped into the SBD portion 51 is a p-type impurity whose ionization rate at the element temperature (e.g., about 500° C.) during load short-circuit is greater than the ionization rate at the element temperature (e.g., about 175° C.) during steady-state operation of the MOSFET. Specifically, instead of using aluminum as the p-type impurity to be co-doped into the SBD portion 51, gallium (Ga) or boron (B (shallow boron)), which has a deeper acceptor level than aluminum, may be used.

[0097] <Method of Manufacturing Semiconductor Device According to Second Preferred Embodiment> A method for manufacturing silicon carbide semiconductor device 50 will be described. The second embodiment differs from the first embodiment in that n - Nitrogen ion implantation and aluminum ion implantation are performed in the thickened portion of the silicon carbide layer 32 in a region for forming the SBD portion 51 so that the nitrogen concentration range and the aluminum concentration range are both set as described above. The order of the nitrogen ion implantation and aluminum ion implantation for forming the SBD portion 51 can be appropriately set.

[0098] Specifically, in the second embodiment, n -After increasing the thickness of the n-type silicon carbide layer 32, and before epitaxially growing the p-type silicon carbide layer 33, - An ion implantation mask having an opening corresponding to the region where the SBD portion 51 is to be formed is formed on the surface of the silicon carbide layer 32, and nitrogen ions and aluminum ions are implanted using the mask. - The order in which the JFET upper portion 3a and the SBD portion 21 are formed in the thickened portion of the silicon carbide layer 32 can be set as appropriate. The nitrogen ion implantation for forming the SBD portion 51 may be performed simultaneously with the nitrogen ion implantation for forming the JFET upper portion 3a.

[0099] <Operation of Silicon Carbide Semiconductor Device According to Second Embodiment> As described above, only the SBD portion 51 of the n-type current spreading region 3 is co-doped with aluminum, which is a p-type impurity, at a concentration lower than the concentration of nitrogen. The SBD portion 51 has a low compensation concentration because the ionization rate of aluminum increases when the device temperature rises. In other words, the carrier concentration of the SBD portion 51 decreases as the device temperature increases. For this reason, the SBD portion 51 has a low compensation concentration when the load is short-circuited, when the device temperature is higher than during steady-state operation of the MOSFET.

[0100] The low compensation concentration of the SBD portion 51 during a load short circuit reduces the short circuit current, and as a result, similar to the first embodiment, the temperature rise due to the short circuit current is reduced, and the increase in the leakage current flowing through the trench SBD 52 can be reduced. Furthermore, the low compensation concentration of the SBD portion 51 during a load short circuit makes it easier for the depletion layer width to expand from the junction between the Schottky electrode 17 and the SBD portion 51, reducing the electric field strength at the junction. This reduces the electric field applied to the trench SBD 52, and reduces the leakage current flowing through the trench SBD 52.

[0101] Therefore, a temperature rise due to leakage current is suppressed, and thermal destruction of the MOSFET (silicon carbide semiconductor device 50) can be suppressed, thereby improving the short circuit resistance. Furthermore, by co-doping the SBD portion 51 with aluminum, the leakage current flowing through the trench SBD 52 during a load short circuit is reduced, so there is no need to relatively lower the nitrogen concentration in the SBD portion 51 of the n-type current diffusion region 3 in order to reduce the leakage current flowing through the trench SBD 52. Therefore, an increase in the characteristic on-resistance can be suppressed, and the short circuit resistance can be improved.

[0102] The operation of silicon carbide semiconductor device 50 according to the second embodiment during steady operation and when off is similar to that of silicon carbide semiconductor device 10 according to the first embodiment.

[0103] <Verification of Silicon Carbide Semiconductor Device According to Second Embodiment> The relationship between the n-type impurity concentration and element temperature in SBD portion 51 of silicon carbide semiconductor device 50 was examined. Fig. 9 is a characteristics diagram showing the relationship between the ionization rate of aluminum and temperature for the silicon carbide semiconductor device according to the second embodiment. Fig. 10 is a characteristics diagram showing the relationship between the n-type compensation concentration and element temperature for the silicon carbide semiconductor device according to the second embodiment.

[0104] FIG. 9 shows the temperature characteristics of the ionization rate of aluminum doped in a silicon carbide semiconductor (Al ionization rate), cited from FIG. 4 of Patent Document 2. It can be seen that in all of the samples with different aluminum concentrations shown in FIG. 9, the ionization rate of aluminum increases with increasing temperature, saturating at a temperature of 800 K or higher and reaching approximately 100%. In the temperature on the horizontal axis of FIG. 9, 800 K corresponds to approximately 500° C., which is the element temperature when the load is short-circuited in the second embodiment. It can also be seen from the temperature characteristics shown in FIG. 9 that the temperature fluctuation of the ionization rate of aluminum shows approximately the same tendency regardless of the aluminum concentration.

[0105] Figure 10 shows the temperature characteristics of the n-type compensation concentration of a silicon carbide semiconductor doped with nitrogen and aluminum, which is cited from FIG. 5 of the above Patent Document 2. Since the n-type compensation concentration is the concentration of carriers that effectively contribute to electrical conduction, it can also be referred to as the carrier concentration. The aluminum concentration of each sample in FIG. 10 corresponds to the aluminum concentration of the sample in FIG. 9. The n-type compensation concentration in FIG. 10 is calculated by subtracting the aluminum concentration (acceptor concentration Na) multiplied by the ionization rate (K) from the nitrogen concentration (donor concentration Nd) (= Nd - K·Na). The ionization rate (K) is a positive constant less than or equal to 1 and has temperature dependence.

[0106] The n-type silicon carbide semiconductor doped with nitrogen and aluminum shown in FIG. 10 corresponds to the SBD portion 51 in Embodiment 2. From the temperature characteristics shown in FIG. 10, it is presumed that the n-type compensation concentration of the n-type silicon carbide semiconductor doped with nitrogen and aluminum (i.e., the SBD portion 51) decreases as the element temperature (the temperature of the semiconductor substrate 30) increases. Therefore, from the temperature characteristics shown in FIGS. 9 and 10, it is presumed that by codoping aluminum into the SBD portion 51, the low compensation concentration of the SBD portion 51 can be achieved in response to an increase in the element temperature.

[0107] <Verification of Simulation of Ionization Rate of p-Type Impurities> The ionization rate of p-type impurities was verified. FIG. 11 is a characteristic diagram showing the results of simulating the temperature characteristics of the ionization rate of p-type impurities. For a plurality of p-type impurities, the impurity concentration (acceptor concentration Na) was set to 1.4×10 17 / cm 3 and the results of simulating the temperature characteristics of the ionization rate at 175°C (the element temperature during the steady operation of the MOSFET) and 500°C (the element temperature during load short circuit) are shown in FIG. 11.

[0108] As the p-type impurity to be co-doped into the SBD portion 51 described above, a p-type impurity whose ionization rate is greater at the element temperature during load short circuit than at the element temperature during steady-state operation of the MOSFET is preferable. From the results shown in FIG. 11, it has been found that, in addition to aluminum, gallium or boron, which has a deeper acceptor level than aluminum, can also be used.

[0109] As described above, according to the second embodiment, by co-doping the SBD portion with aluminum, the carrier concentration in the SBD portion is reduced as a compensation concentration during load short circuit when the element temperature increases, thereby providing the same effect as in the first embodiment. Furthermore, according to the second embodiment, the n-type impurity concentration in the SBD portion can be increased up to the amount by which the carrier concentration in the SBD portion is reduced as a compensation concentration during load short circuit, and the specific on-resistance can be increased by the amount by which the n-type impurity concentration in the SBD portion is increased.

[0110] The first and second embodiments are common in that the carrier concentration in the SBD portion is lower than the concentration of the n-type impurity added to the upper portion of the JFET. The first and second embodiments are also common in that the lower limit of the carrier concentration in the SBD portion is 10% or more of the concentration of the n-type impurity added to the upper portion of the JFET. The first and second embodiments are different in the following points.

[0111] In the first embodiment, the carrier concentration of the SBD portion is substantially equal to the concentration of the n-type impurity added thereto. The reason is that the n-type impurity is activated almost 100% in the assumed operating temperature range of the MOSFET and emits carriers (electrons). That is, the carrier concentration of the SBD portion is determined by the concentration of the n-type impurity added thereto. Therefore, in the first embodiment, it is assumed that only the n-type impurity is substantially added to the portion above the JFET and the SBD portion. "Substantially" means that other impurities are allowed to be mixed in to an extent that does not impair the significance of the embodiment. More specifically, the amount of other impurities that is allowed is about 1 / 100 or less, or about 1 / 1000 or less, of the impurity intentionally added.

[0112] On the other hand, in the second embodiment, both n-type and p-type impurities are added to the SBD portion. Therefore, the carrier concentration in the SBD portion is determined by the concentration (=Nd-K·Na) obtained by subtracting the concentration (Na) of the p-type impurity multiplied by the ionization rate (K) from the concentration (Nd) of the added n-type impurity. This is because, while the n-type impurity is activated almost 100% in the expected operating temperature range of the MOSFET and emits carriers (electrons), the ionization rate (K) of the p-type impurity varies with temperature, and the concentration of the emitted carriers (holes) changes. Therefore, in the second embodiment, the carrier concentration in the SBD portion is determined by the compensation concentration (=Nd-K·Na).

[0113] The impurity elements and impurity concentrations in semiconductor elements can be measured by SIMS (secondary ion mass spectrometry). Depending on the shape of the semiconductor element, measurements by AES (Auger electron spectroscopy) may be used in addition as auxiliary data. The carrier concentration in the semiconductor element can be determined based on the impurity concentration measured by the above-mentioned method, with reference to the ionization rate calculated from the depth and density of the dopant level as shown in Figure 11. Depending on the shape of the semiconductor element, measurements by sMIM (scanning microwave impedance microscope) may be used in addition as auxiliary data.

[0114] The present disclosure is not limited to the above-described embodiments, and various modifications can be made without departing from the spirit and scope of the present disclosure. For example, in each of the above-described embodiments, the n-type current diffusion region may be doped with n-type impurities (donors) other than nitrogen (e.g., phosphorus (P)). [Industrial Applicability]

[0115] INDUSTRIAL APPLICABILITY As described above, silicon carbide semiconductor devices according to the present disclosure are useful for power semiconductor devices used in power conversion devices, power supply devices for various industrial machines, and the like. [Explanation of symbols]

[0116] 1n + Type drain region 2n - Type drift region (first semiconductor region) 3. n-type current diffusion region (fourth semiconductor region) 3a JFET upper part 3b JFET lower part 4 p-type base region (second semiconductor region) 5n + Type source region (third semiconductor region) 6 p ++ Mold Contact Area 7. Trench 7a Gate trench 7b Schottky Trench 8 Gate insulating film 9 Gate electrode 10,50 Silicon carbide semiconductor device 11 p + type area 11a first high concentration region (fifth semiconductor region) 11b Second high concentration region (sixth semiconductor region) 12 Interlayer insulating film 13 Ohmic electrode 14 Source electrode 15 Drain electrode 16 Drain electrode pad 17 Schottky electrode 20,52 Trench type SBD 21,51 SBD part 30 Semiconductor substrate 31 n + Starting substrate 32n - mold silicon carbide layer 33 p-type silicon carbide layer W1 Width of SBD part W11 Gate trench width W12 Schottky trench width W21 Width of first high concentration region W22 Width of the second high concentration region X: a first direction parallel to the front surface of the semiconductor substrate Y: A second direction that is parallel to the front surface of the semiconductor substrate and perpendicular to the first direction Z depth direction

Claims

1. A semiconductor substrate made of silicon carbide; a first semiconductor region of a first conductivity type provided within the semiconductor substrate; a second semiconductor region of a second conductivity type provided between the front surface of the semiconductor substrate and the first semiconductor region; a third semiconductor region of a first conductivity type selectively provided between the front surface of the semiconductor substrate and the second semiconductor region; a fourth semiconductor region of a first conductivity type provided between the second semiconductor region and the first semiconductor region and in contact with the second semiconductor region and the first semiconductor region, the fourth semiconductor region being doped with a first conductivity type impurity at a concentration higher than that of the first semiconductor region; a plurality of trenches having a predetermined depth from a front surface of the semiconductor substrate; Among the plurality of trenches, a first trench penetrates the third semiconductor region and the second semiconductor region in a depth direction and terminates inside the fourth semiconductor region; a second trench, which is disposed apart from the third semiconductor region and is one of the plurality of trenches other than the first trench, penetrates the second semiconductor region in a depth direction, and terminates inside the fourth semiconductor region; a gate electrode provided inside the first trench via a gate insulating film; a Schottky electrode in contact with the fourth semiconductor region on an inner wall of the second trench; a first electrode electrically connected to the second semiconductor region and the third semiconductor region, embedded in the second trench, and in contact with the Schottky electrode; A second electrode electrically connected to a rear surface of the semiconductor substrate; a first conductivity type SBD portion which is a part of the fourth semiconductor region and is in contact with the Schottky electrode on a sidewall of the second trench; a JFET upper portion of a first conductivity type that is a part of the fourth semiconductor region and is provided between a sidewall of the first trench and the SBD portion; a Schottky barrier diode utilizing the rectification of a Schottky barrier formed at a junction between the Schottky electrode and the SBD portion; Equipped with a carrier concentration of the SBD portion is lower than a concentration of the first conductivity type impurity added to the JFET upper portion.

2. The SBD portion is doped with substantially only the first conductivity type impurity, a carrier concentration of the SBD portion is determined by a concentration of the first conductivity type impurity added to the SBD portion; The concentration of the first conductive type impurity added to the SBD portion is 1×10 16 / cm 3 2. The silicon carbide semiconductor device according to claim 1, wherein the concentration of the first conductivity type impurity added to the JFET upper portion is within a range of 10% or more and less than 100%.

3. the SBD portion is doped with the first conductivity type impurity and the second conductivity type impurity; 2. The silicon carbide semiconductor device according to claim 1, wherein a carrier concentration of the SBD portion is determined by a concentration obtained by subtracting a concentration of the second conductivity type impurity multiplied by an ionization rate from a concentration of the first conductivity type impurity added to the SBD portion.

4. 4. The silicon carbide semiconductor device according to claim 3, wherein the second conductivity type impurity is any one of aluminum, gallium, and boron.

5. a fifth semiconductor region of a second conductivity type provided between the second semiconductor region and the first semiconductor region and spaced apart from the second semiconductor region, the fifth semiconductor region facing a bottom surface of the first trench; a sixth semiconductor region of a second conductivity type provided between the second semiconductor region and the first semiconductor region and spaced apart from the second semiconductor region, the sixth semiconductor region facing a bottom surface of the second trench; a first conductivity type JFET lower portion that is a part of the fourth semiconductor region and is provided between the fifth semiconductor region and the sixth semiconductor region that are adjacent to each other in a direction parallel to the front surface of the semiconductor substrate, between the JFET upper portion and the first semiconductor region, 5. The silicon carbide semiconductor device according to claim 1, wherein the SBD portion is not in contact with a lower portion of the JFET.

6. a fifth semiconductor region of a second conductivity type provided between the second semiconductor region and the first semiconductor region and spaced apart from the second semiconductor region, the fifth semiconductor region facing a bottom surface of the first trench; a sixth semiconductor region of a second conductivity type provided between the second semiconductor region and the first semiconductor region and spaced apart from the second semiconductor region, the sixth semiconductor region facing a bottom surface of the second trench, 5. The silicon carbide semiconductor device according to claim 1, wherein a width from a junction surface between the Schottky electrode and the SBD portion to an interface between the JFET upper portion and the SBD portion is narrower than a width by which the sixth semiconductor region protrudes in a direction away from a sidewall of the second trench parallel to a front surface of the semiconductor substrate.

7. 7. The silicon carbide semiconductor device according to claim 6, wherein a width from a junction surface between said Schottky electrode and said SBD portion to an interface between said JFET upper portion and said SBD portion is within a range of 0.1 μm or more and 0.5 μm or less.

8. 5. The silicon carbide semiconductor device according to claim 1, wherein the first conductivity type impurity is nitrogen.

Citation Information

Patent Citations

  • Semiconductor device

    JP2018060984A

  • Semiconductor device and method of manufacturing the same

    JP2018107167A

  • Silicon carbide semiconductor device

    WO2022244749A1

  • Silicon carbide semiconductor device

    JP2020077664A