Silicon carbide semiconductor component and method for producing the silicon carbide semiconductor component

Abstract

A silicon carbide semiconductor device comprising: a plurality of trenches (7) provided in a silicon carbide substrate (10) of a first conductivity type, the plurality of trenches (7) being provided at a predetermined depth from a front surface of the silicon carbide substrate (10); a first semiconductor region (4) of a second conductivity type provided between adjacent trenches (7) of the plurality of trenches (7); a second semiconductor region (5) of the first conductivity type selectively provided in the first semiconductor region (4); a third semiconductor region (11) of the second conductivity type selectively provided in the silicon carbide substrate (10) to cover a bottom of a trench (7) of the plurality of trenches (7); a fourth semiconductor region (12) of the second conductivity type selectively provided in the silicon carbide substrate (10) to be in contact with the first semiconductor region (4) between adjacent trenches (7) of the plurality of trenches (7); a gate electrode (9) provided on a gate insulating layer (8) in the trench (7) of the plurality of trenches (7); a unit structure arranged in a plurality with a predetermined pitch, each of which has an insulated gate structure formed by the gate electrode (9) in the trench (7) of the plurality of trenches (7); a first electrode (16) connected to the first semiconductor region (4) and the second semiconductor region (5); and a second electrode (18) connected to a rear surface of the silicon carbide substrate (10), characterized in that two or more trenches (7) of the plurality of trenches (7) are arranged between the fourth semiconductor region (12) and an adjacent fourth semiconductor region (12); and a front surface of the second semiconductor region (5) between the two or more trenches (7) is covered by an interlayer insulating film (14).

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention

[0001] Embodiments of the present invention relate to a silicon carbide semiconductor device and a method for manufacturing a silicon carbide semiconductor device. 2. Description of the state of the art

[0002] Silicon carbide (SiC) is expected to replace silicon (Si) as the next-generation semiconductor material. Compared with a conventional semiconductor device using silicon, a semiconductor device using silicon carbide (hereinafter referred to as a silicon carbide semiconductor device) has several advantages, such as reducing the device's on-state resistance to several hundredths of that of a conventional semiconductor device and using it in higher temperature environments (200 degrees C or higher). These differences arise from the properties of the material itself. For example, silicon carbide has a band gap about 3 times that of silicon and a dielectric breakdown field strength nearly 10 times that of silicon.

[0003] Currently, these silicon carbide semiconductor devices, which are now commercial products, include Schottky barrier diodes (SBDs), vertical metal-oxide-semiconductor field-effect transistors (MOSFETs) with a planar gate structure, and vertical MOSFETs (hereinafter referred to as trench-gate SiC MOSFETs) with a trench-gate structure.

[0004] A trench-gate structure is a three-dimensional structure in which an MOS gate is embedded in a trench formed in a semiconductor base containing silicon carbide (hereinafter referred to as a silicon carbide base). In a trench-gate structure, a portion along a sidewall of the trench is used as a channel (inversion layer). Therefore, when comparing elements of the same on-resistance (RonA), elements with a trench-gate structure can have a smaller element area (die area) than elements with a planar gate structure that has a flat-state MOS gate on a silicon carbide base, and therefore, the former can be considered a promising gate structure.

[0005] However, as described above, silicon carbide has a dielectric breakdown field strength nearly 10 times greater than that of silicon. Therefore, the trench bottom in a trench-gate SiC MOSFET is subject to a larger electric field than in a trench-gate MOSFET using silicon, which reduces the breakdown voltage at the trench bottom. The breakdown voltage is the voltage limit that does not cause device failure. Since the breakdown voltage of the device as a whole is a flux-limiting factor in the breakdown voltage at the trench bottom, a cell structure (unit structure) has been proposed to mitigate the electric field at the trench bottom to ensure a predetermined breakdown voltage.

[0006] As a trench-gate SiC MOSFET that ensures a predetermined breakdown voltage, a device was proposed that includes a region of p +-type which is in contact with the entire region of the trench bottom and extends to a position that is in contact with a part of the side wall of the trench (see, for example, Japanese Patent Laid-Open No. JP 2012-099601 A (Section 0021, Fig. 1)).

[0007] Another trench-gate SiC MOSFET was proposed that has a region of p + -type that reaches an n-type drift region from a boundary of a p-type base region and an n-type current spreading layer (see, for example, Japanese Patent Laid-Open No. JP 2015-072 999 A (Section 0023 to 0024, Fig. 9).

[0008] A structure of a conventional trench-gate SiC MOSFET is described. Fig. Figure 20 is a cross-sectional view of a structure of a conventional trench-gate SiC MOSFET. Fig. Figure 20 shows a cross-section of two adjacent cells (element-constituent units). Fig. The conventional trench-gate SiC MOSFET shown in Figure 20 includes a MOS gate of an ordinary trench-gate structure on a front surface side (surface on the side of a p-type base region 104) of a silicon carbide-containing semiconductor base (hereinafter referred to as a silicon carbide base) 110.

[0009] The silicon carbide base 110 is formed by sequentially forming by epitaxial growth on a carrier substrate of the n + -type (hereinafter referred to as silicon carbide substrate of n + -type) 101, which contains silicon carbide, an n-type drift region 102, an n-type region 103, and the p-type base region 104. The n-type region 103 is a so-called current distribution layer (CSL). A first p-type region is selectively formed in the n-type region 103. +-Type 111 intended to cover the entire bottom of a trench 107.

[0010] Furthermore, in the n-type region 103 between adjacent trenches 107 (in a mesa section) there is a second p-type region + -Type 112 is selectively provided. The first and second regions of the p + -Type 111, 112 have the function of attenuating the electric field applied to the bottom of the trench 107. The reference numbers 105, 106, 108, 109, 114 to 117 are each a source region of the n + -type, a contact region of p ++ -type, a gate insulating layer, a gate electrode, an interlayer insulating film, a barrier metal, a source electrode and an exhaust electrode.

[0011] Fig. 21 and Fig. 22 are planar views of the conventional trench-gate SiC MOSFET. The structure at the section line AA-AA' in Fig. 21 and the structure at the section line BB-BB' in Fig. 22 correspond Fig. 20. Fig. 21 shows a planar arrangement of the trench 107 and the contact region of the pp ++ -Type 106. Fig. Figure 22 shows a planar arrangement of the first and second regions of the p + -Type 111, 112 (hatched areas) and a JFET region 113 (white areas).

[0012] As in Fig. 21 shows the contact region of the p ++-type 106 and the trench 107 are arranged in a striped planar configuration, extending along a predetermined direction X (hereinafter referred to as the first direction) and repeatedly alternating along a direction Y (hereinafter referred to as the second direction) orthogonal to the first direction X. Two adjacent trenches 107 each have an end region 107a that is continuous with the trenches, so that they essentially have an annular planar configuration. A connecting region 107b of the trench 107 is connected to a gate runner 119.

[0013] As in Fig. 22 shows the first and second regions of the p + -type 111, 112 and the JFET region 113 are arranged in a striped planar configuration extending along the first direction X. Every second region of the p +-type 112 along the second direction Y is enclosed by the trenches 107 in the substantially annular planar arrangement. Every second pair of the JFET regions 113 along the second direction Y is enclosed by the trenches 107 in the substantially annular planar arrangement.

[0014] In the Fig. In the conventional trench-gate SiC MOSFET shown in Figures 20 to 22, the cell pitch (width of a single cell) w104 is determined by the sum of a width w101 of the first region of the p + -Type 111, (Half the width w102 of the second region from p + -Type 112) × 2 and a width w103×2 between the first and second region of the p + -Type 111, 112 (hereinafter referred to as FET (JFET) connection region 113). The minimum values ​​of the respective widths w101 to w103 of the first and second regions of the p +-Type 111, 112 and the JFET region 113 are determined by process limits (e.g. etching process limits) of the semiconductor manufacturing equipment.

[0015] In particular, the minimum value of the width w101 of the first region of the p + -Type 111 1.5 µm and the minimum value of the width w102 of the second region of the p + -Type 112 is 1.0 µm. Fig. 20 shows the second region of the p + -Type 112 for one cell (ie half of the second region from the p + -Type 112). The minimum value of the width w103 of the JFET region 113 is 1.0 µm. When adopting the cell structure of the Fig. For the conventional trench-gate SiC MOSFETs shown in Figure 20, the minimum value of the cell pitch w104 is thus 4.5 µm (= 1.5 µm+(1.0 µm×1 / 2)×2+1.0 µm×2).

[0016] International Patent Application No. WO 2013 / 187019 A1 discloses a method for manufacturing a SiC semiconductor device, comprising: forming a p-type layer within a trench by epitaxial growth; and then forming a p-type SiC layer by leaving the p-type layer only on the bottom portion and the two front end portions of the trench by hydrogen etching. That is, parts of the p-type layer formed on the side surfaces of the trench are removed. In this way, the p-type SiC layer can be formed without relying on oblique ion implantation. SUMMARY OF THE INVENTION

[0017] According to one aspect of the present invention, there is provided a silicon carbide semiconductor device according to claim 1.

[0018] In the silicon carbide semiconductor device, the fourth semiconductor region is provided separately from the third semiconductor region.

[0019] In the silicon carbide semiconductor device, the fourth semiconductor region is partially connected to the third semiconductor region adjacent to it.

[0020] In the silicon carbide semiconductor device, three or more trenches of the plurality of trenches are arranged between the fourth semiconductor region and the adjacent fourth semiconductor region, and the third semiconductor region and an adjacent third semiconductor region are partially connected to each other.

[0021] In the silicon carbide semiconductor device, the plurality of trenches are arranged in a striped configuration extending along a direction parallel to the front surface of the silicon carbide substrate.

[0022] The silicon carbide semiconductor device includes a fifth semiconductor region of the second conductivity type selectively provided in the first semiconductor region; and a sixth semiconductor region of the second conductivity type selectively provided in the first semiconductor region. The fifth semiconductor region is disposed in a position facing the fourth semiconductor region in a depth direction. The sixth semiconductor region is disposed near end portions of two or more trenches of the plurality of trenches arranged between the fourth semiconductor region and the adjacent fourth semiconductor region. The first electrode is connected to the first semiconductor region through the fifth semiconductor region and the sixth semiconductor region.

[0023] In the silicon carbide semiconductor device, the specified step size is 4 µm or less.

[0024] According to a further aspect of the present invention, a method for producing a silicon carbide semiconductor device according to claim 8 is provided.

[0025] In the method, the insulated gate structure comprises: a first semiconductor region of a second conductivity type disposed between adjacent trenches of the plurality of trenches, and a second semiconductor region of the first conductivity type selectively provided in the first semiconductor region. The method includes, prior to forming the plurality of trenches: depositing a first silicon carbide layer of the first conductivity type on a front surface of a starting substrate containing silicon carbide; selectively forming a third semiconductor region of the second conductivity type in the first silicon carbide layer; selectively forming a fourth semiconductor region of the second conductivity type in the first silicon carbide layer such that it is exposed at a surface of the first silicon carbide region;Depositing a second silicon carbide layer of the second conductivity type to cover the third semiconductor region and the fourth semiconductor region, the second silicon carbide layer becoming the first semiconductor region; and forming the silicon carbide substrate obtained by sequentially depositing the starting substrate, the first silicon carbide layer, and the second silicon carbide layer; and selectively forming the second semiconductor region in the second silicon carbide layer. The plurality of trenches are formed to penetrate the second semiconductor region and the second silicon carbide layer and reach the third semiconductor region.

[0026] The method comprises, after etching back the polysilicon: forming a first electrode connected to the second semiconductor region and the second silicon carbide layer; and forming a second electrode connected to a back surface of the silicon carbide substrate.

[0027] Further aspects of the invention are the subject of the subclaims, the drawings and the description of embodiments.

[0028] The purposes, features and advantages of the present invention are particularly set forth in the following detailed description of the invention, or will become apparent when read in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a cross-sectional view of a structure of a silicon carbide semiconductor device according to a first embodiment; Fig. 2 is a characteristic diagram for the number of distances of a second region from p + -Type; Fig. 3A, Fig. 3B, Fig. 4A, Fig. 4B, Fig. 5A, Fig. 5B, Fig. 6A, Fig. 6B, Fig. 7A, Fig. 7B and Fig. 8 are cross-sectional views of the silicon carbide semiconductor device according to the first embodiment during manufacture; Fig. 9 is a plan view of a planar arrangement of the silicon carbide semiconductor device according to a second embodiment; Fig. 10, Fig. 11 and Fig. 12 are plan views of other examples of the planar arrangement of the silicon carbide semiconductor device according to the second embodiment; Fig. 13 is a cross-sectional view of the structure of the silicon carbide semiconductor device according to a third embodiment; Fig. 14 is a plan view of a planar arrangement of the silicon carbide semiconductor device according to the third embodiment; Fig. 15 is a cross-sectional view of the structure of the silicon carbide semiconductor device according to a fourth embodiment; Fig. 16 is a plan view of a planar arrangement of the silicon carbide semiconductor device according to the fourth embodiment; Fig. 17 is a plan view of a planar arrangement of the silicon carbide semiconductor device according to a fifth embodiment; Fig. 18 is a plan view of another example of the planar arrangement of the silicon carbide semiconductor device according to the fifth embodiment; Fig. 19 is a cross-sectional view of the structure of the silicon carbide semiconductor device according to a sixth embodiment; Fig. 20 is a cross-sectional view of a conventional trench-gate SiC MOSFET structure; and Fig. 21 and Fig. 22 are planar views of the conventional trench-gate SiC MOSFET. DETAILED DESCRIPTION OF THE INVENTION

[0029] Embodiments of a silicon carbide semiconductor device and a method for manufacturing a silicon carbide semiconductor device according to the present invention will be described in detail with reference to the accompanying drawings. In the present description and the accompanying drawings, layers and regions with an n- or p-prefix mean that the majority carriers are electrons or holes. In addition, a + or - appended to n or p means that the impurity concentration is higher or lower, respectively, than layers and regions without + or -, and it represents a single example. Depending on the optimal structure, the concentrations may be reversed. In the description of the following embodiments and the accompanying drawings, identical main parts are given the same reference numerals and will not be described repeatedly.

[0030] A structure of the silicon carbide semiconductor device according to a first embodiment will be described. Fig. 1 is a cross-sectional view of the structure of the silicon carbide semiconductor device according to the first embodiment. Although in Fig. 1 two adjacent cells (constituent units of the element) in the plurality of cells arranged in an active region 41 are shown, all cells have the same structure (unit structure) (similarly applies to Fig. 3 to 7). The Fig. The silicon carbide semiconductor device according to the first embodiment shown in FIG. 1 is a trench-gate SiC MOSFET having a MOS gate with a trench-gate structure on a front surface side (the surface of the p-type base region 4 side) of a semiconductor base (silicon carbide base (semiconductor chip)) 10 containing silicon carbide.

[0031] The silicon carbide base 10 is formed by sequential formation by means of epitaxial growth on a carrier substrate 1 containing silicon carbide from the n + -type (silicon carbide substrate of n + -type), wherein the silicon carbide layers (first and second silicon carbide layers) 21, 22 become an n-type drift region 2 and a p-type base region (first semiconductor region) 4. The MOS gate is formed by the p-type base region 4, a source region (second semiconductor region) of n + -Type 5, a contact region (fifth semiconductor region) of p ++-type 6, a trench 7, a gate insulating layer 8, and a gate electrode 9 are formed. Specifically, on a surface layer on a source side (a source electrode 16 side) of the n-type silicon carbide layer 21, which becomes the n-type drift region 2, an n-type region 3 (hereinafter referred to as an n-type CSL region) is provided so as to be in contact with the p-type base region 4.

[0032] A part of the n-type silicon carbide layer 21, excluding the n-type 3 CSL region, is the n-type 2 drift region. The n-type 3 CSL region is a so-called current spreading layer (CSL), which reduces the carrier bulk resistance. The n-type 3 CSL region is provided, for example, uniformly along a direction (hereinafter referred to as the horizontal direction) parallel to a front base surface (the front surface of the silicon carbide base 10). The n-type 3 CSL region may be an n-type silicon carbide layer formed by epitaxial growth on the n-type silicon carbide layer 21, which becomes the n-type 2 drift region. In the n-type 3 CSL region, first and second regions of the p + -type (third and fourth semiconductor regions) 11, 12 are selectively provided.

[0033] The first region of the p +-Type 11 is provided so as to cover a bottom portion and a lower corner portion of the trench 7. The lower corner portion of the trench 7 is a boundary region of the bottom and a side wall of the trench 7. The first region of the p + -Type 11 is provided from a deep position further on an exhaust side than an interface of the p-type 4 base region and the n-type 3 CSL region, to a depth that does not reach an interface of the n-type 3 CSL region and the n-type 2 drift region. The arrangement of the first p-type 4 region + -Type 11 allows the formation of a pn junction between the first region of the p + -type 11 and the n-type CSL region 3 near the bottom of the trench 7. Therefore, the high electric field in a portion of the gate insulating layer 8 along the bottom of the trench 7 can be suppressed.

[0034] The second region of the p +-Type 12 (12a, 12b) is provided between adjacent trenches 7 (mesa section) 30(31) so that it is separated from the first region of the p + -type 11 and is in contact with the base region of p-type 4. Furthermore, the second region is separated from the p + -Type 12 from the interface of the p-type 4 base region and the n-type 3 CSL region to a depth that does not reach the interface of the n-type 3 CSL region and the n-type 2 drift region. The arrangement of the second region of the p + -Type 12 enables the formation of a pn junction between the second region of the p + -type 12 and the n-type CSL region 3 in a mesa portion 30(31) at a deep position farther away on the suction side than the bottom of the trench 7. Consequently, the high electric field in a portion of the gate insulating layer 8 along the bottom of the trench 7 can be suppressed.

[0035] Furthermore, the second region of the p + -type 12 at certain intervals, forming a mesa section (hereinafter referred to as the first mesa section) 31 in which the second region of the p + -type 12 is provided, and a mesa section (hereinafter referred to as second mesa section) 32 in which the second region of the p + -Type 12 is not provided. The removal of the second region of the p + -Type 12 is the provision of the mesa section 30 (32), in which the second region of the p + -type 12 is not provided. In other words, between adjacent second regions of the p + -Type 12, which face each other along the second direction Y, there are at least two trenches 7. Fig. 1 shows a case where two trenches are located between adjacent second regions of the p +-type 12, which face each other along the second direction Y (number of distances = 1). The second region of the p + -Type 12 acts as the base region of the p-type.

[0036] A JFET region 13a of the first mesa section 31 is a region between adjacent first and second regions of the p + -type 11, 12. A JFET region 13b of the second mesa section 32 is a region between adjacent first regions of the p + -Type 11. In other words, the cell structure of the present invention is compared with a conventional cell structure (see Fig. 20) a structure in which some of the mesa sections (the second mesa sections 32), the second region of the p +-type and a JFET region are not present, making the cell pitch (the width of a cell) w5 smaller than that of the conventional structure. In particular, the cell pitch w5 is determined by the sum of a width w1 of the first region of the p + -Type 11, half the width w2 of the second region of the p + -type 12 of the first mesa section 31, a width w3 of the JFET region 13a of the first mesa section 31 and half a width w4 of the JFET region 13b of the second mesa section 32.

[0037] The minimum values ​​of a width w6 and the widths w1 to w4 of Trench 7, the first and second regions of the p + -type 11, 12 and the JFET regions 13a, 13b are determined by process limits (e.g., etching process limits) of the semiconductor manufacturing equipment. In particular, the minimum value of the width w6 of the trench 7 is approximately 0.7 µm to 0.8 µm. The first region of the p +-Type 11 is arranged in such a way that the entire bottom of the trench 7 is covered and thus - assuming a deviation of the positioning of the first region of the p + -Type 11 along the second direction Y with respect to the trench 7 - the minimum value of the width w1 of the first region of the p + -Type 11 approximately 1.5 µm. The minimum value of the width w2 of the second region of the p + -Type 12 is about 1.0 µm. Fig. 1 shows the second region of the p + -Type 12 for one cell (ie half of the second region from the p + -Type 12). The minimum value of the widths w3, w4 of the JFET region 13a, 13b is approximately 1.0 µm. Provided that it is ensured that the width w4 of the JFET region 13b of the second mesa section 32 is approximately 1.0 µm, it is achieved that the adjacent first regions of the p + -Type 11 sandwich the second mesa section 32.

[0038] Based on the above dimensions, the minimum value of the cell pitch w5 of the cell structure, including the second region of the p + -Type 12 (ie, the cell structure formed by adjacent first and second mesa sections 31, 32) is 3.5 µm (=1.5 µm+(1.0 µm×1 / 2)+1.0 µm+(1.0 µm×1 / 2)), which is about 1.0 µm less than that of the conventional structure (the cell pitch w104 = 4.5 µm). Consequently, the element area (chip area) can be reduced as a whole. Furthermore, the cell pitch w5 of the cell structure, including the second region of the p + -Type 12, about 4 µm or less. In this case, for example, the width w2 of the second region of the p + -Type 12 can be increased by the difference to the minimum value of the cell pitch w5 (4.0 µm-3.5 µm=0.5 µm). The following effects are achieved in this case.

[0039] The larger the width w2 of the second region of the p +-type 12, the larger the width of a contact opening for contacting a source electrode (first electrode) 16 and a semiconductor portion (the silicon carbide base 10). Furthermore, the larger the width w2 of the second region of the p + -Type 12, the greater the assured safety margin when assuming a patterning deviation, e.g. for an ion implantation mask to form the contact region of the p ++ -Type 6 and an etching mask for creating the contact opening for the contact between the source electrode 16 and the semiconductor portion. Therefore, the contact between the source electrode 16 and the semiconductor portion is facilitated, which enables a reduction in the defect rate of the contact between the source electrode 16 and the semiconductor portion.

[0040] In a surface region (surface layer) of the front base surface of the p-type silicon carbide layer 22, which becomes the p-type base region 4, the source region of the n + -Type 5 and the contact region of the p ++ -Type 6. The source region of the n + -type 5 faces the gate electrode 9 via a part of the gate insulating layer 8 along the sidewall of the trench 7. Furthermore, the source region of the n + -Type 5 is also provided in the first and second mesa sections 31, 32. The contact region of the p ++ -Type 6 is arranged in the first mesa section 31 and is, for example, the second region of the p + -Type 12 in a depth direction. For example, the depth of the contact region of the p ++ -Type 6 deeper than that of the source region of the n + -Type 5.

[0041] The contact region of the p ++ -Type 6 has contact with the source region of the n +-Type 5 of the first mesa section 31. Furthermore, a contact region of the p ++ -type that make contact with the source region of the n + -Type 5 in the second mesa section 32, arranged near the end regions of the adjacent trenches 7, which sandwich the second mesa section 32 (reference numbers 6b and 6c in Fig. 9 to 12, which are described below). By contacting the source electrode 16 with these contact regions of the p ++ -Type, the source regions of the n + -type 5 of the first and second mesa sections 31, 32 are fixed at the same potential (source potential). A part of the n-type silicon carbide layer 22, excluding the source region of the n + -Type 5 and the contact region of the p ++ -type 6, the base region 4 is of p-type 4.

[0042] Trench 7 penetrates the source region of the n +-type 5 and the p-type base region 4 from the front base surface and reaches the n-type CSL region 3. In the trench 7, the gate insulating film 8 is arranged along the sidewall of the trench 7. The gate electrode 9 is arranged on the gate insulating film 8 so as to be embedded in the trench 7. A source-side terminal of the gate electrode 9 is located at a high position substantially identical to that of the front base surface. The gate electrode 9 is electrically connected, in a part not shown (e.g., the end region of the gate electrode 9), to a gate pad (not shown) via a gate runner (not shown), which becomes a conducting part of the gate electrode 9.

[0043] An interlayer insulating film 14 is provided from the gate electrode 9 across the second mesa portion 32 so as to cover the entire region of the second mesa portion 32 and the gate electrode 9. A contact opening 14a is provided which penetrates the interlayer insulating film 14 in the depth direction and reaches the first mesa portion 31, so that the source electrode 16 is in contact with the source region of the n + -Type 5 and the contact region of the p ++ -type 6 of the first mesa section 31 and is electrically insulated from the gate electrode 9 by the interlayer insulating film 14. Furthermore, the source electrode 16 is electrically connected in a part not shown to the source region of the n + -Type 5 of the second mesa section 32.

[0044] The source electrode 16 may be provided so as to cover a barrier metal 15, or it may be provided only in the contact opening 14a. For example, the barrier metal 15 may be arranged between the source electrode 16 and the interlayer insulating film 14, which prevents the diffusion of metal atoms from the source electrode 16 to the gate electrode 9 side. A source pad 17 is arranged on the source electrode 16 so as to be embedded in the contact opening 14a. On a rear surface of the silicon carbide base 10 (rear surface of a silicon carbide substrate of the n + -Type 1, which leads to a suction region of the n + -type) a suction electrode (second electrode) 18 is arranged.

[0045] A suitable number of removals of the second region of the p + -Type 12 is described. Fig. 2 is a characteristic diagram for the number of distances of the second region from p+ -Type. As in Fig. 2, the higher the number of removals of the second region of the p + -Type 12, the smaller the cell step size w5 can be, which can reduce the element area. If the number of removals of the second region of the p + -Type 12 is two or more, both the second region of the p + -Type 12 containing cell structure as well as the cell structure containing the second region of the p + -type (ie, a cell formed by adjacent second mesa sections 32). The cell pitch w5 of the cell structure, including the second region of the p + -Type 12, corresponds to the description above (see Fig. 1). A minimum value of a cell step size w5' of the cell structure that represents the second region of the p +-type 12 (ie a cell formed by adjacent second mesa sections 32) is, for example, 2.5 µm (see Fig. 15 as described below).

[0046] A range of about 1 or more to 30 or less for the number of removals of the second region of the p + -Type 12 is the limit at which an effect obtained by reducing the cell pitch w5 (reduction of the ON resistance) is achieved. Therefore, the number of removals of the second region 12 of the p + -type 30 or less. Furthermore, the number of removals of the second region of the p + -Type 12 must be an odd number. The reason for this is that the plurality of neighboring cells of the same structure can be arranged continuously. A suitable number of distances of the second region of the p +The number of cells of type 12 can be 1, 3, or a maximum of 9 or less (left side of the dotted vertical line). This is because the effect of reducing the cell step size on the number of removals is large.

[0047] A method for manufacturing a silicon carbide semiconductor device according to the first embodiment will be described. Fig. 3A, Fig. 3B, Fig. 4A, Fig. 4B, Fig. 5A, Fig. 5B, Fig. 6A, Fig. 6B, Fig. 7A, Fig. 7B and Fig. 8 are cross sections of the silicon carbide semiconductor device according to the first embodiment during manufacture. Fig. 3A, Fig. 4A, Fig. 5A, Fig. 6A and Fig. 7A shows a cross section of the active region 41; Fig. 3B, Fig. 4B, Fig. 5B, Fig. 6B and Fig. 7B shows an edge termination region 42. The active region 41 is a region where current flows during the ON state (current control region). The edge termination region 42 is a region that mitigates the electric field on the front base surface of the n-type drift region 2 and ensures the breakdown voltage.

[0048] First, the silicon carbide substrate of the n + -Type 1, which belongs to the runoff region of the n + -type is manufactured. After that, on the front surface of the silicon carbide substrate of the n + -type 1 silicon carbide layer 21a, which becomes the n-type drift region 2, is formed by epitaxial growth. After that, in the surface layer of the n-type silicon carbide layer 21a, the first region of the p + -Type 11 and a region of the p + -type (hereinafter referred to as sub-area of ​​the p +-type) 12a are each selectively formed by photolithography and ion implantation of a p-type impurity. The p-type impurity region + -Type 12a is part of the second region of the p + -Type 12.

[0049] Subsequently, an n-type impurity is implanted into the entire n-type silicon carbide layer 21 by ion implantation, thereby forming an n-type region (hereinafter referred to as the n-type partial region) 3a in the entire surface layer of the n-type silicon carbide layer 21. The n-type partial region 3a is a part of the n-type CSL region 3. At this time, the depth of the n-type partial region 3a is made deeper than that of the first region of the p + -Type 11, and the entire extraction side (the side of the silicon carbide substrate of the n + -Type 1) of the first region of the p + -Type 11 and the sub-area of ​​the p +The n-type silicon carbide layer 12a is covered by the n-type portion 3a. A portion of the n-type silicon carbide layer 21a, which lies further toward the suction side than the n-type portion 3a, becomes the n-type drift region 2.

[0050] The order in which the subregion of n-type 3a, the first region of p + -Type 11 and the sub-area of ​​the p + The ion implantation into the silicon carbide layer can be performed at room temperature (below 200 degrees C) or at a high temperature (approximately 200 degrees C to 500 degrees C). For example, when performing ion implantation at room temperature, a resist film is used as a mask, and when performing ion implantation at a high temperature, an oxide film (SiO2) is used as a mask (similarly for all ion implantations described below).

[0051] Next, the n-type silicon carbide layer 21b is formed on the n-type silicon carbide layer 21a by epitaxial growth. The n-type silicon carbide layer 21a and the n-type silicon carbide layer 21b form the n-type silicon carbide layer 21, as described above. Subsequently, the p-type silicon carbide layer portion is + -type 12b is selectively formed by photolithography and ion implantation of a p-type impurity in a part of the n-type silicon carbide layer 21b corresponding to the p + -Type 12a in the depth direction to a depth that covers the sub-area of ​​the p + -Type 12a is achieved. The width and impurity concentration of the sub-region of the p + -Type 12b, for example, are essentially the same as those of the sub-area of ​​the p + -Type 12a.

[0052] The sub-areas of the p +-type 12a, 12b are coupled together in the depth direction (vertical direction), whereby the second region of the p + -type 12 is formed. The second region of the p + -type 12 is formed in the first mesa section 31 and is not formed in the second mesa section 32. Furthermore, the second region of the p + -Type 12 (hereinafter referred to as the outermost region of the p + -type 12') is also formed near the boundary of the active region 41 with the edge termination region 42 and extended in a subsequent process so that it is located further to the outside (chip edge side) than a step 43 (see Fig. 3B), which is formed between the active region 41 and the edge termination region 42.

[0053] Subsequently, an n-type impurity is implanted in the entire n-type silicon carbide layer 21b by ion implantation, whereby in the entire n-type silicon carbide layer --type 21b, an n-type subregion 3b is formed at a depth that reaches the n-type subregion 3a. The impurity concentration of the n-type subregion 3b can be substantially the same as that of the n-type subregion 3a. The n-type subregions 3a, 3b are coupled to each other in the depth direction, thereby forming the n-type CSL region 3. In a subsequent process, the n-type CSL region 3 is extended so that it lies further outward than the step 43 formed between the active region 41 and the edge termination region 42. The order in which the p-type subregion + -type 12b and the n-type 3b subregion are interchangeable.

[0054] Next, the p-type base region 4, which becomes the p-type silicon carbide layer 22, is formed on the n-type silicon carbide layer 21 by epitaxial growth. Through the processes described so far, the silicon carbide base (semiconductor wafer) 10 is formed, in which the n-type silicon carbide layer 21 and the p-type silicon carbide layer 22 are sequentially deposited on the n-type silicon carbide substrate. + -Type 1 silicon carbide layer 22 is then deposited. The p-type silicon carbide layer 22 is then removed from the entire edge termination region 42 by photolithography and etching, exposing the n-type silicon carbide layer 21. At this time, the surface layer of the n-type silicon carbide layer 21 and the p-type silicon carbide layer 22 can be easily removed.

[0055] As a result, the step 43, where the edge termination region 42 is lower than the active region 41, is formed on the front surface of the silicon carbide base 10. At a connecting part (staircase of the step 43) 43a between the front base surfaces (upper and lower) of the active region 41 and the edge termination region 42, the outermost region of the p + -type silicon carbide layer 12' is exposed from the lower step side to the lower step surface (surface of the n-type silicon carbide layer 21). When the p-type silicon carbide layer 22 is removed from the entire edge termination region 42, the p-type silicon carbide layer 22 can be removed up to an outer peripheral portion of the active region 41. The staircase 43a of the step 43 can be inclined with respect to the front base surface.

[0056] Subsequently, a process is carried out in which photolithography and ion implantation are carried out while one set is repeatedly carried out, in the surface layer of the front surface of the silicon carbide base 10, the sequential formation of the source region of the n + -Type 5, the contact region of the p ++ -type 6, a junction termination structure (junction termination extension (JTE)) and a channel stop region of n-type 46, etc. The contact region of the p ++ -Type 6, which is located at the outermost (hereinafter referred to as the outermost contact region of the p ++ -type 6'), is formed from the active region 41 to the edge termination region 42, and the end region of the contact region of the p ++ -Type 6, for example, is extended so that it is further on the outside than the stairs 43a of step 43 in order to form the outermost region of the p + -Type 12' to cover.

[0057] The JTE structure is formed by a plurality of adjacent p-type regions (in this case two: a p-type region + -type region 44 and a p-type region 45) arranged concentrically around the periphery of the active region 41 and having impurity concentrations that progressively decrease the further outward the p-type region is located. The p-type region + -Type 44 is formed in such a way that it makes contact with the outermost contact region of the p ++ -type 6' on the innermost side (the side of the active region 41) of the edge termination region 42. The p-type region 45 is formed further to the outside than the p + -type 44 and is formed in such a way that it aligns with the region of the p +-type 44 is in contact. The n-type channel stop region 46 is formed separately from the p-type region 45, further outward than the p-type region 45. Activation annealing (heat treatment) of the regions formed by ion implantation is performed.

[0058] Thereafter, an oxide film 51 is deposited on the front base surface, for example, using a chemical vapor deposition process at normal pressure (for example, approximately 700 hPa to 1300 hPa, which results from the installed gas flow rate). The thickness of the oxide film 51 can be, for example, approximately 0.7 µm. Subsequently, the oxide film 51 is patterned by photolithography and etching, exposing an area corresponding to a formation area of ​​the trench 7. The etching can be performed, for example, as an overetch (OE) of 25% more than a predetermined etching amount (= thickness of the oxide film 51) by means of dry etching. The state up to this point is shown in Fig. 3A and Fig. 3B.

[0059] Thereafter, the remaining part of the oxide film 51 is used as a mask and etching is performed, thereby forming the trench 7 so that the source region of the n + -type 5 and the base region of the p-type 4 are penetrated and the first region of the p +-type 11 in the CSL region of n-type 3 is achieved. The depth of trench 7 can be, for example, approximately 1.5 µm. Furthermore, the sidewall of trench 7 can be inclined with respect to the front base surface. For example, the sidewall of trench 7 can be inclined by approximately 85 degrees with respect to the front base surface, whereby the upper side of trench 7 is wider than the lower side. The oxide film 51 is removed. After the etching to pattern trench 7, annealing is performed to round off the corner regions of the opening of trench 7 and the bottom of trench 7. Thereafter, for example, the sacrificial oxidation of the inner wall of trench 7 is performed by dry oxidation, and the surface layer of the inner wall of trench 7 is removed by a thickness of, for example, approximately 10 nm. The state up to this point is shown in Fig. 4A and Fig. 4B.

[0060] Next, a thermal oxide film with a thickness of, for example, approximately 10 nm is formed by dry oxidation along the front surface of the silicon carbide base 10 (surface of the p-type silicon carbide layer 22) and the inner wall of the trench 7. After that, a deposition oxide layer with a thickness of, for example, approximately 500 nm is deposited (formed) on the thermal oxide film by a CVD method at normal pressure so as to be embedded in the trench. The thermal oxide film and the deposition oxide film are sequentially stacked to form a field oxide film 52. Next, heat treatment is performed in a nitrogen (N2) atmosphere at approximately 1000 degrees C for a period of 30 minutes, thereby improving the film quality (e.g., densification) of the field oxide film 52. The state up to this point is shown in Fig. 5A and Fig. 5B.

[0061] Subsequently, the field oxide film 52 is patterned by photolithography and etching (e.g., wet etching) to expose the inner wall of trench 7 and the front base surface (area where the MOS gate is formed) in the active region 41. Thereafter, the inner wall of trench 7 and the exposed portion of the front base surface are subjected to sacrificial oxidation, for example, by dry oxidation, and the surface layer of the front base surface and the inner wall of trench 7 are removed by a thickness of, for example, approximately 10 nm. Subsequently, the inner wall of trench 7 and the exposed portion of the front base surface are subjected to thermal oxidation, forming, for example, a high-temperature oxide (HTO) film 53 along the inner wall of trench 7 and the front base surface, which becomes the gate insulating layer 8.

[0062] The thickness of the HTO film 53 on the sidewall of trench 7 can be, for example, approximately 90 nm. After the formation of the HTO film 53, annealing can be performed to improve the film quality of the HTO film 53 (e.g., densification). Next, a doped polysilicon (poly-Si) layer 54, doped with, for example, an n-type impurity, is deposited on the surface of the HTO film 53 and the field oxide film 52 so that it is embedded in the trench 7. The thickness of the doped polysilicon layer 54 can be, for example, approximately 500 nm. After that, the extent of damage in the doped polysilicon layer 54 is increased by ion implantation of argon (Ar) (improving the non-crystalline quality). The state up to this point is shown in Fig. 6A and Fig. 6B.

[0063] Thereafter, a resist mask 55 is formed on the surface of the doped polysilicon layer 54 by photolithography, covering a portion corresponding to a formation region of a gate runner 19. Subsequently, the doped polysilicon layer 54 is etched using the resist mask 55 as a mask, with portions of the doped polysilicon layer 54 becoming the gate electrode 9 and the gate runner 19 remaining. Specifically, the doped polysilicon layer 54 is etched (i.e., etched back) until the gate insulating layer 8 is exposed, leaving the doped polysilicon layer 54, which becomes the gate electrode 9 in the trench 7. The height position on the upper surface of the gate electrode 9 is a position at the same height as or below that of the front base surface ( Fig. 7A).

[0064] In addition, near the boundary between the active region 41 and the edge termination region 42, the part of the doped polysilicon layer 54 covered by the resist mask 55 is left as a gate runner ( Fig. 7B). The gate electrode 9 and the gate runner 19 are connected near the boundary between the active region 41 and the edge termination region 42 ( Fig. 8). This etching can, for example, be a chemical dry etching (CDE). Because no resist mask is used in the formation of the gate electrode 9, there is no need to ensure a safety margin for patterning deviations of the resist mask, which allows the cell pitch w5 to be reduced by an amount corresponding to this safety margin. The state up to this point is shown in Fig. 7A, Fig. 7B and Fig. 8 shown.

[0065] Thereafter, the interlayer insulating film 14 is formed on the entire front base surface so as to cover the gate electrode 9. Next, the interlayer insulating film 14 and the gate insulating layer 8 are patterned and the contact opening is formed, thus forming the source region of the n + -Type 5 and the contact region of the p ++ -type 6 are exposed. Thereafter, the barrier metal 15 is formed and patterned to cover the interlayer insulating film 14 and the source region of the n + -Type 5 and the contact region of the p ++ -type 6. Subsequently, the source electrode 16 is formed so that contact is made with the source region of the n + -Type 5 and the contact region of the p ++ -Type 6. The source electrode 16 can be formed to cover the barrier metal 15 and can remain in the contact opening.

[0066] Next, the source contact pad 17 is formed so that it is embedded in the contact opening. A portion of the metal layer deposited to form the source contact pad 17 can be used as a gate contact pad. The gate contact pad can be formed so that it is in contact with the gate runner 19. On the rear surface of the n-type silicon carbide substrate, + -Type 1, an extraction electrode 18 is formed. After that, the semiconductor wafer is cut (separated) into individual chips, whereby the Fig. 1 shown trench-gate SiC MOSFET is completed.

[0067] As described above according to the first embodiment, the first region of the p + -type covering the trench floor, and the second region of the p +-type, which is in contact with the p-type base region of a mesa section directly below the p-type base region (extraction side), thereby ensuring a predetermined breakdown voltage. Furthermore, the second region of the p + -type arranged so that two or more trenches are formed between adjacent second regions of the p + -type. In addition, in a mesa section (second mesa section) in which no second region of the p + -type is provided, no contact region of the p ++ -type or contact opening. Consequently, the cell pitch can be reduced by an amount corresponding to the non-arrangement of either the second region of the p + -type or contact opening, thereby enabling a reduction of the on-resistance (RonA). Furthermore, the mesa section, in which no second region of the p +-type region is arranged at a predetermined pitch, whereby the ON resistance of the entire device (entire chip) can be reduced substantially uniformly. Therefore, the ON resistance can be reduced while maintaining the predetermined resistance determined by arranging the first and second regions of the p + -type breakdown voltage is maintained. Furthermore, according to the first embodiment, the p-type base region and the n-type source region are + -type of the mesa section, in which no second region of the p + -type, for example, electrically connected to a source electrode near the end region of the trench and the potential of the source electrode is fixed. Therefore, even with selective arrangement of the second region of the p + -type prevents the breakdown voltage in the mesa section, where no second region of the p + -type is arranged, is reduced.

[0068] In a second embodiment, a planar arrangement of the silicon carbide semiconductor component according to the first embodiment (see Fig. 1) described. Fig. 9 is a plan view of a planar arrangement of the silicon carbide semiconductor device according to the second embodiment. The planar arrangement represents a planar shape and arrangement configuration of the components as viewed from the front surface of the silicon carbide base 10 side. Fig. Figure 9 shows a planar arrangement of trench 7 and the contact region 6 of the p ++ -type in the active region 41; regions provided on the surface layer of the front base surface and the gate insulating layer 8 are not shown (similar to Fig. 10 to 12).

[0069] A planar arrangement in a case where the number of removals of the second region of the p + -Type 12 is one, is in Fig. 9. In other words, a cross section along the section line AA' in Fig. 9 corresponds to Fig. 1. The trench 7 is arranged in a striped planar configuration and extends in a predetermined direction (hereinafter referred to as the first direction) X. An end region 7a of the length dimension of all trenches 7 is connected to the gate runner 19. The gate runner 19 is arranged, for example, in a linear planar configuration and extends in a direction (hereinafter referred to as the second direction) Y that is orthogonal to the first direction X.

[0070] In the first and second mesa sections 31, 32 (between trenches 7) a p-type base region and an n-type source region are + -type (not shown). The contact region of the p ++-type 6 in the first mesa section 31 is arranged in a striped planar configuration extending in the first direction X. An end region 6a of the length dimension of all contact regions of the p ++ -Type 6 is connected to a contact region (hereinafter referred to as connection part (sixth semiconductor region)) 6b and arranged in a comb-shaped planar configuration.

[0071] The connecting part 6b of the contact region of the p ++ -type 6 is arranged, for example, at a position facing the gate runner 19 via the trench 7, in a linearly designed planar configuration substantially parallel to the second direction Y. In other words, an electrical contact region (source contact) in contact with the source electrode (not shown) is arranged in a comb-shaped planar configuration substantially the same as that of the contact region of the p ++-Type 6 and its connecting part 6b. The source region of the n + -type (not shown) of the second mesa section 32 is connected to the connecting part 6b of the contact region of the p ++ -Type 6 connected.

[0072] Fig. 10, Fig. 11 and Fig. 12 are plan views of other examples of the planar arrangement of the silicon carbide semiconductor device according to the second embodiment. In a case where the number of removals of the second region of the p + -Type 12, as in Fig. 10, between adjacent first mesa sections 31 is two, two of the second mesa sections 32 are arranged so as to be adjacent to each other along the second direction Y. Except for the arrangement of the second mesa section 32, the configuration of the silicon carbide semiconductor device according to the second embodiment shown in Fig. 1 is the same as that of the silicon carbide semiconductor device according to the Fig. 9 shown second embodiment.

[0073] Furthermore, in a case where the number of removals of the second region of the p + -Type 12 is two or more, as in Fig. 11 and Fig. 12, the end regions 7a of the trenches 7 are connected in such a way that a single trench with a meandering planar design is arranged. Fig. 11 and Fig. 12 represent a case where the number of removals of the second region of the p + -type 12 is six. Connected sections 7b of the trenches 7 have, for example, an arcuate planar shape. The gate runner 19 can be connected to both end regions 7c of the single trench of the meandering planar arrangement ( Fig. 11). In this case, the gate runner 19, which has a substantially rectangular planar shape, can be arranged separately near each of the end regions 7c of the individual trench of the meandering planar arrangement.

[0074] Furthermore, the gate runner 19 can be connected to each of the connected sections 7b of the trenches 7 ( Fig. 12). In this case, the gate runner 19 can be arranged, for example, in a comb-shaped planar configuration formed by a linear portion 19a of a linear shape extending along the second direction Y and a plurality of portions 19b connected to the connected portions 7b of the trenches 7 and the linear portion 19a. The end portions 7c of the individual trenches of the meandering planar configuration can each be connected to the linear portion 19a of the gate runner 19 by the connecting portions 19b, or they can extend on the gate runner 19 side to be connected to the linear portion 19a of the gate runner 19.

[0075] The contact region of the p ++ -Type 6 of the first mesa section 31 is arranged in a striped planar configuration running along the first direction X. A contact region of the p ++ -type (hereinafter referred to as partial contact region of the p ++-type (sixth semiconductor region) 6c is selectively separated from the contact region of the p ++ -type 6 of the first mesa section 31, at a position facing the second mesa section 32 along the first direction X and located between the connected sections 7b of adjacent trenches 7 along the second direction Y. The partial contact region of the p ++ -type 6c is connected to the source electrode (not shown) and the source region of the n + -type (not shown) of the second mesa section 32. The source contact is arranged in a planar layout that is essentially the same as that of the contact region of the p ++ -Type 6 and the partial contact region of the p ++ -Type 6c.

[0076] As described above, according to the second embodiment, application to the first embodiment is possible, and effects identical to those of the first embodiment can be obtained.

[0077] A structure of the silicon carbide semiconductor device according to a third embodiment will be described. Fig. 13 is a cross-sectional view of the structure of the silicon carbide semiconductor device according to the third embodiment. Fig. 13 is a cross section along the section line BB' in Fig. 14. Fig. 14 is a plan view of a planar arrangement of the silicon carbide semiconductor device according to the third embodiment. In Fig. 14, Trench 7 is shown in black (similar to Fig. 16 to 18). The silicon carbide semiconductor device according to the third embodiment differs from the silicon carbide semiconductor device according to the first embodiment in that the source electrode 16 makes contact with the source region of the n + -Type 5 of the second mesa section 32.

[0078] In particular, as in Fig. 13, a contact hole (hereinafter referred to as the first contact hole) 14a penetrating the interlayer insulating film 14 in the depth direction is added in the first mesa portion 31, and a second contact hole 14b penetrating the interlayer insulating film 14 in the depth direction is provided in the second mesa portion 32. The source electrode 16 is connected to the source region of the n + -Type 5 and the contact region of the p ++ -type 6 in the first mesa section 31 through the first contact opening 14a, and it is connected to the source region of the n +-Type 5 of a second mesa section 32 through the second contact opening 14b.

[0079] In this case, as in Fig. 14, for example, two adjacent trenches 7 are connected by the connecting part 7b at the end regions 7a, and they form a substantially annular planar arrangement surrounding the periphery of the first mesa section 31. The gate runner 19 is arranged at each of the end regions 7a of the trenches 7. The gate runners 19 are arranged, for example, in a comb-shaped planar configuration formed by a linear part 19a of a linear shape extending along the second direction Y and the connecting parts 19b each connecting the connected parts 7b of the trenches 7 and the linear part 19a.

[0080] The source region of the n + -type (not shown) and the contact region of the p ++-Type 6 of the first mesa section 31 are arranged in a striped planar layout running along the first direction X. The source region of the n + -type (not shown) and the contact region of the p ++ -type 6 of the first mesa section 31 are arranged inside the annular planar arrangement of the trench 7. The partial contact region of the p ++ -Type 6c is selectively separated from the contact region of the p ++ -type 6 of the first mesa section 31, at a position facing the second mesa section 32 along the first direction X and located between the connected parts 7b of adjacent trenches 7 along the second direction Y.

[0081] As described above, according to the third embodiment, effects identical to those of the first embodiment can be achieved. Furthermore, according to the third embodiment, the source region of the n +-type of the second mesa section are connected directly to the source electrode and fixed at the source potential.

[0082] A structure of the silicon carbide semiconductor device according to a fourth embodiment will be described. Fig. 15 is a cross-sectional view of the structure of the silicon carbide semiconductor device according to the fourth embodiment. Fig. 15 is a cross section along the section line CC' in Fig. 16. Fig. 16 is a plan view of a planar arrangement of the silicon carbide semiconductor device according to the fourth embodiment. The silicon carbide semiconductor device according to the fourth embodiment is a trench-gate SiC MOSFET in a case where, in the first embodiment, the number of removals of the second region of the p + -Type 12 is three.

[0083] In other words, three adjacent second mesa sections 32 are arranged along the second direction Y between adjacent first mesa sections 31. The three adjacent second mesa sections 32 along the second direction Y, which are arranged between the adjacent first mesa sections 31, are covered by the interlayer insulating film 14. The cell pitch w5' of a cell structure in which no second region is separated from the p + -Type 12 is the sum of the width w1 of the first region of the p + -Type 1 and half the width w4 of the JFET region 13b of the second mesa section 32 × 2; and the corresponding minimum value is 2.5 µm (= 1.5 µm + (1.0 µm×1 / 2)×2).

[0084] The planar arrangements of trench 7, the contact region of the p ++ -Type 6, the partial contact region of the p ++ -Type 6c and Gate-Runner 19 can be the same as those in Fig. 11, Fig. 12 of the second embodiment. In other words, the end regions 7a of the trenches 7 can be connected to form a single trench with a meandering planar layout. As shown in Fig. 16, the trenches 7 may further comprise a planar arrangement in which the end regions 7a of adjacent trenches 7 sandwiching the first mesa section 31 are connected and extend such that they form the contact region of the p ++ -Type 6 meandering.

[0085] As described above, according to the fourth embodiment, effects identical to those of the first to third embodiments can be achieved. Furthermore, according to the fourth embodiment, the setting of the number of removals of the second region of the p + -Type 12 on three or more allow the arrangement of a cell structure in which no second region of the p +-Type 12, which allows the cell pitch to be further reduced. This allows the total ON resistance of the element and the element area (chip area) to be reduced.

[0086] A structure of the silicon carbide semiconductor device according to a fifth embodiment will be described. Fig. 17 is a plan view of a planar arrangement of the silicon carbide semiconductor device according to the fifth embodiment. Fig. Fig. 18 is a plan view of another example of the planar arrangement of the silicon carbide semiconductor device according to the fifth embodiment. The cross section along the section line DD' in Fig. 17 corresponds to the Fig. 13 of the third embodiment. The cross section at the section line EE' in Fig. 18 corresponds to the Fig. 15 of the fourth embodiment. In Fig. 17 and Fig. 18 are a planar arrangement of the first and second regions 11, 12 (hatched area) of the p + -type and the JFET regions 13a, 13b (white filled areas) are shown.

[0087] The silicon carbide semiconductor device according to the fifth embodiment differs from the silicon carbide semiconductor device according to the third and fourth embodiments in that the adjacent first and second regions of the p + -types 11, 12 and the neighboring first regions of the p +-type 11 are partially connected at predetermined intervals and respectively separate the JFET regions 13a, 13b. In other words, a plurality of the JFET regions 13a, 13b are arranged at predetermined intervals along the second direction Y such that the JFET regions (the JFET regions 13a, 13b) are arranged in a matrix-like planar configuration. The JFET regions have, for example, a linear shape extending along the second direction Y.

[0088] As described above, according to the fifth embodiment, effects identical to those of the first to fourth embodiments can be achieved. Furthermore, according to the fifth embodiment, adjacent first and second regions of the p + -type as well as neighboring first regions of the p + -type partially connected at predetermined intervals, which allows a fixation of the potential of the region of the p +-type and avoidance of reductions in breakdown voltage due to floating potential.

[0089] A structure of the silicon carbide semiconductor device according to a sixth embodiment will be described. Fig. Fig. 19 is a cross-sectional view of the structure of the silicon carbide semiconductor device according to the sixth embodiment. The silicon carbide semiconductor device according to the sixth embodiment differs from the silicon carbide semiconductor device according to the first embodiment in that in the n-type CSL region 3 directly below the second p-type region + -type 12 (on the suction side), an n-type CSL region (hereinafter referred to as partial n-type CSL region) 61 is selectively provided, which is in contact with the second region of the p +-type 12. The partial n-type CSL region 61 has an impurity concentration higher than that of the n-type CSL region 3. The width of the partial n-type CSL region 61 may, for example, be the same as the width w2 of the second region of the p + -Type 12.

[0090] As described above, according to the sixth embodiment, effects identical to those of the first to fifth embodiments can be achieved.

[0091] According to the sixth embodiment, furthermore, the partial CSL region of the n-type is directly below the second region of the p + -type, which allows the breakdown voltage to be reduced near the second region of the p + -type is lower than that near the first region of the p + -type. Consequently, the occurrence of an avalanche breakthrough near the second region of the p +-type compared to the proximity of the first region of the p + -type, which makes it possible to avoid the occurrence of an avalanche breakthrough on the trench floor.

[0092] In the present invention described above, various modifications that do not deviate from the spirit of the invention are possible, and in the embodiments described above, for example, the dimensions and impurity concentrations of the components, the number of removals of the second region of the p + -type or similar can be selected in various ways depending on the required specifications. Furthermore, the arrangement of the trench forming the MOS gate of the trench-gate structure can be changed in various ways; the active region can be divided into several sections, and the trenches can be arranged in the sections in respective predetermined planar configurations.

[0093] In the embodiments described above, although a MOSFET was described as an example without limiting the present invention, a broad application is possible with respect to other silicon carbide semiconductor devices of a trench-gate structure. For example, an insulated gate bipolar transistor (IGBT) or the like can be cited as an example of another silicon carbide semiconductor device of a trench-gate structure. Furthermore, the present invention is implemented similarly when the conductivity types (n-type, p-type) are reversed.

[0094] Although in general, the smaller the cell pitch w104 is, the lower the ON resistance (RonA) can be made, the cell pitch w104 in the conventional structures above (see Fig.20) cannot be made smaller than 4.5 µm due to process limitations of semiconductor manufacturing equipment. Therefore, a reduction of the on-resistance below the on-resistance achievable by a cell pitch w104 of 4.5 µm is impossible, and the advantage of a trench gate structure in terms of the ability to reduce (shrink) the element area compared to a planar gate structure is not fully realized.

[0095] According to the present invention, the third semiconductor region covering the trench bottom is disposed, and the fourth semiconductor region in contact with the first semiconductor region is disposed directly below the first semiconductor region (p-type base region) between trenches (first mesa portion), thus ensuring and maintaining a predetermined breakdown voltage. Furthermore, according to the invention described above, the fourth semiconductor region is selectively disposed so that two or more trenches are located between adjacent fourth semiconductor regions, thereby reducing the cell pitch.

[0096] The silicon carbide semiconductor device and the method for manufacturing a silicon carbide semiconductor device according to the present invention achieve an effect that the breakdown voltage can be maintained and the ON resistance can be reduced.

[0097] As described above, the silicon carbide semiconductor device and the method for manufacturing a silicon carbide semiconductor device according to the present invention are useful for power semiconductor devices for use in power conversion devices and power supply devices such as industrial machines, and are particularly suitable for silicon carbide semiconductor devices of a trench gate structure.

Claims

[1] A silicon carbide semiconductor device comprising: a plurality of trenches (7) provided in a silicon carbide substrate (10) of a first conductivity type, the plurality of trenches (7) being provided at a predetermined depth from a front surface of the silicon carbide substrate (10); a first semiconductor region (4) of a second conductivity type provided between adjacent trenches (7) of the plurality of trenches (7); a second semiconductor region (5) of the first conductivity type selectively provided in the first semiconductor region (4); a third semiconductor region (11) of the second conductivity type selectively provided in the silicon carbide substrate (10) to cover a bottom of a trench (7) of the plurality of trenches (7); a fourth semiconductor region (12) of the second conductivity type selectively provided in the silicon carbide substrate (10) to be in contact with the first semiconductor region (4) between adjacent trenches (7) of the plurality of trenches (7); a gate electrode (9) provided on a gate insulating layer (8) in the trench (7) of the plurality of trenches (7); a unit structure arranged in a plurality with a predetermined pitch, each of which has an insulated gate structure formed by the gate electrode (9) in the trench (7) of the plurality of trenches (7); a first electrode (16) connected to the first semiconductor region (4) and the second semiconductor region (5); and a second electrode (18) connected to a rear surface of the silicon carbide substrate (10), characterized by , that two or more trenches (7) of the plurality of trenches (7) are arranged between the fourth semiconductor region (12) and an adjacent fourth semiconductor region (12); and a front surface of the second semiconductor region (5) between the two or more trenches (7) is covered by an interlayer insulating film (14). [2] The silicon carbide semiconductor device according to claim 1, characterized by that the fourth semiconductor region (12) is provided separately from the third semiconductor region (11). [3] The silicon carbide semiconductor device according to claim 1, characterized by that the fourth semiconductor region (12) is partially connected to the third semiconductor region (11) adjacent to it. [4] The silicon carbide semiconductor device according to claim 3, characterized by , that three or more trenches (7) of the plurality of trenches (7) are arranged between the fourth semiconductor region (12) and the adjacent fourth semiconductor region (12), and the third semiconductor region (11) and an adjacent third semiconductor region (11) are partially connected to each other. [5] The silicon carbide semiconductor device according to any one of claims 1 to 4, characterized by that the plurality of trenches (7) are arranged in a striped layout running along a direction parallel to the front surface of the silicon carbide substrate (10). [6] The silicon carbide semiconductor device according to claim 5, comprising: a fifth semiconductor region (6) of the second conductivity type selectively provided in the first semiconductor region (4); and a sixth semiconductor region of the second conductivity type selectively provided in the first semiconductor region (4), wherein the fifth semiconductor region (6) is arranged in a position facing the fourth semiconductor region (12) in a depth direction, the sixth semiconductor region is arranged near end regions of two or more trenches (7) of the plurality of trenches (7) between the fourth semiconductor region (12) and the adjacent fourth semiconductor region (12), and the first electrode (16) is connected to the first semiconductor region (4) through the fifth semiconductor region (6) and the sixth semiconductor region. [7] The silicon carbide semiconductor device according to any one of claims 1 to 6, characterized by that the specified step size is 4 µm or less. [8] A method of manufacturing a silicon carbide semiconductor device having a plurality of unit structures each having an insulated gate structure formed by a trench provided in the silicon substrate (10) of a first conductivity type at a predetermined depth from a front surface of the silicon carbide substrate (10), and a gate electrode (9) provided on a gate insulating layer (8) in the trench (7), the method comprising: Forming a plurality of trenches (7) at a predetermined depth from the front surface of the silicon carbide substrate (10); Forming the gate insulating layer (8) along an inner wall of the trench (7); Forming a polysilicon layer on the surface of the gate insulating layer (8) and the front surface of the silicon carbide substrate (10) so as to be enclosed in the trench (7); Etching back the polysilicon layer until the gate insulating layer (8) is exposed, whereby the polysilicon layer becomes the gate electrode (9) in the trench (7) and thus forms the plurality of unit structures each having the insulated gate structure, wherein the insulated gate structure comprises: a first semiconductor region (4) of a second conductivity type provided between adjacent trenches (7) of the plurality of trenches (7); and a second semiconductor region (5) of the first conductivity type selectively provided in the first semiconductor region (4), the method comprising, prior to forming the plurality of trenches (7): Depositing a first silicon carbide layer of the first conductivity type on a front surface of a starting substrate containing silicon carbide; selectively forming a third semiconductor region (11) of a second conductivity type in the first silicon carbide layer; selectively forming a fourth semiconductor region (12) of the second conductivity type in the first silicon carbide layer to be exposed on a surface of the first silicon carbide layer; Applying a second silicon carbide layer of the second conductivity type so as to cover the third semiconductor region (11) and the fourth semiconductor region (12), wherein the second silicon carbide layer becomes the first semiconductor region (4) and the silicon carbide substrate (10) is formed, which is obtained by sequentially applying the starting substance, the first silicon carbide layer and the second silicon carbide layer; and selectively forming the second semiconductor region (5) in the second silicon carbide layer, wherein: the plurality of trenches (7) are formed to penetrate the second semiconductor region (5) and the second silicon carbide layer and reach the third semiconductor region (11); and a front surface of the second semiconductor region (5) between the two or more trenches (7) is covered by an interlayer insulating film (14). [9] The method according to claim 8, after etching back the polysilicon, comprising: Forming a first electrode (16) connected to the second semiconductor region (5) and the second silicon carbide layer; and Forming a second electrode (18) connected to a rear surface of the silicon carbide substrate (10).

Images