SILICON CARBIDE SEMICONDUCER UNIT AND METHOD FOR PRODUCING THE SAME

By integrating halogen-containing field limiting rings to compensate for trapped charges, the breakdown voltage stability and reliability of silicon carbide semiconductor devices are enhanced, addressing the issue of charge-induced fluctuations in breakdown voltage.

DE112017003653B4Active Publication Date: 2024-11-21MITSUBISHI ELECTRIC CORP
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
DE112017003653
Authority / Receiving Office
DE · DE
Patent Type
Patents
Current Assignee / Owner
Priority Date
2016-07-20
Filing Date
2017-05-19
Publication Date
2024-11-21
Estimated Expiration
2037-05-19

AI Technical Summary

Technical Problem

Trapped charges at the interface between silicon carbide (SiC) and insulating layers in semiconductor devices affect the breakdown voltage characteristic, leading to variations and fluctuations, especially in connection structures designed for high chip density and cost reduction, which existing compensation methods fail to adequately address.

Method used

Incorporating halogen-containing field limiting ring regions in the semiconductor device structure, which compensate for trapped positive charges with negative charges, thereby stabilizing the breakdown voltage performance.

Benefits of technology

The incorporation of halogen-containing field limiting rings effectively reduces the influence of trapped positive charges, maintaining consistent breakdown voltage performance and improving reliability by compensating for charge fluctuations.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

Silicon carbide semiconductor unit comprising the following: - a semiconductor substrate (1); - a semiconductor layer (2) of n-type, arranged on the semiconductor substrate (1), made of silicon carbide and having an elemental region (RE) and a connection region (RT) outside the elemental region (RE); - a plurality of field-limiting ring regions (5) having a p-type, arranged in the terminal region (RT) of the semiconductor layer (2) and spaced apart from each other; and - a field insulating layer (7) arranged on the terminal area (RT) of the semiconductor layer (2) and in contact with the field limiting ring regions (5) and the semiconductor layer (2), wherein each of the field limiting ring regions (5) has a halogen-containing field limiting ring region (5h), wherein the halogen-containing field limiting ring region (5h) is in contact with the field insulating layer (7) and contains atoms from the group of halogens.
Need to check novelty before this filing date? Find Prior Art

Description

TECHNICAL FIELD

[0001] The present invention relates to a silicon carbide semiconductor device and a method for producing the same. STATE OF THE ART

[0002] For power electronic devices, such as inverters, there is a constant demand for further energy savings. Therefore, it is necessary to reduce the losses of the power semiconductor devices used in these devices. In particular, insulated-gate bipolar transistors (IGBTs), metal-oxide-semiconductor field-effect transistors (MOSFETs), positive-intrinsic-negative (PiN) diodes, Schottky barrier diodes (SBDs), and so on require such a reduction in losses.

[0003] As an effective method for reducing losses, a method using silicon carbide (SiC) instead of silicon (Si), the most common semiconductor material, has been studied; its practical application has begun, for example, in the railway field and so on.

[0004] By taking advantage of the fact that SiC can withstand a higher electric field for dielectric breakdown than Si, the thickness of the device is further reduced, thus reducing losses. Furthermore, the use of SiC ensures high-temperature operation, so this is also useful for downsizing the devices used to cool a semiconductor device. Therefore, it is expected that SiC devices will become more popular in the market as their cost reduction progresses.

[0005] SiC devices are expected to be used in the breakdown voltage range of several hundred V to several tens of kV. To ensure sufficient breakdown voltage, such a device is typically provided with a terminal structure, referred to as a field limiting ring (FLR) or a guard ring.

[0006] It is known that certain trapped charges can exist at the interface between SiC and an insulator, such as silicon dioxide (SiO2). According to Non-Patent Document 1, it is stated that a positive charge of +2.4 × 10 12 cm -2exists at the interface between the p-type SiC with the plane orientation (0001) and the SiO2 layer formed by dry oxidation. According to Non-Patent Document 2, when an oxide layer is formed by dry oxidation, a negative charge of -1.4 × 10 exists at the interface between an n-type SiC with a plane orientation (0001) and SiO2. 12 cm -2 exists.

[0007] Trapped charges can affect device characteristics. In Patent Document 1, the influence of a positive surface charge between the oxide and the semiconductor is reduced by disposing a low-concentration p-type surface charge compensation region between the p-type guard ring layers. In Patent Document 2, by utilizing the fact that a positive charge trapped layer is formed by dry oxidation and the fact that a negative charge trapped layer is formed by wet oxidation and wet reoxidation, an electric field relaxation effect is achieved with a depletion layer between the SiC and the interfacial charge trapped layer. PRIOR ART DOCUMENTSPatent documents Patent Document 1: Japanese Patent Application Laid-Open No. JP 2013-62518 A Patent Document 2: Japanese Patent Application Laid-Open No. JP 2003-282 888 A Non-patent documents Non-Patent Document 1: M. Noborio et al., IEEE Trans. Electron Devices, Vol. 56, No. 9, pages 1953 to 1958, Sept. 2009 Non-patent document 2: T. Kimoto et al., Jpn. J. Appl. Phys., Vol. 44, No. 3, pages 1213 to 1218, 2005 BRIEF DESCRIPTIONProblem to be solved by the invention

[0008] Trapped charges can also affect the breakdown voltage characteristics, which are an important characteristic for the power semiconductor device. With this in mind, we investigated n-type devices with p-type field-limiting rings as terminal structures. Each of the Fig. 1 and Fig. Figure 2 is a graphical representation showing a comparison between measurement results and simulation results in a relationship between doses and breakdown voltages for forming a field limiting ring.

[0009] The simulation assumes that there is no trapped charge at the interface between the field confinement ring and the field insulation layer. Fig. 1 represents a case where the field insulation layer on the field limiting ring is made of SiO2, and Fig. Figure 2 illustrates a case where the field insulation layer on the field limiting ring is made of polyimide. In both cases, the measurement result differs from the simulation result.

[0010] In detail, a shift of the dose to the positive side (the right side in the figure) is observed in the measurement result compared to the simulation result (hereinafter referred to as a “dose shift”). Fig. 1 will result in a dose shift of approximately 2.5 x 10 12 cm -2 observed, and in Fig. 2 a dose shift of about 6.5 x 10 12 cm -2 The dose shift causes a variation in the ion implantation amount capable of achieving the designed breakdown voltage value.

[0011] Accordingly, when SiC is used as a semiconductor material, the terminal structure is affected by trapped charges. As a result, the actual breakdown voltage value differs from the design value assuming no trapped charge is present. In particular, when the terminal structure area is reduced to increase the number of chips per wafer to reduce costs, the margin for the breakdown voltage becomes narrow. For this reason, the influence of trapped charges may increase.

[0012] Furthermore, trapped charges can be introduced into the semiconductor device from the outside, causing fluctuations in the breakdown voltage value. The introduction of trapped charges from the outside can occur either during the manufacturing of the semiconductor device, during assembly of the module using the semiconductor device, or during use of the semiconductor device.

[0013] Furthermore, the method according to Patent Document 1 compensates the surface charge between the oxide and the semiconductor to a certain extent. However, due to the design for the breakdown voltage, the surface charge compensation region cannot have a high concentration; the amount of positive surface charge that can be compensated is limited. Furthermore, trapped positive charges at the interface between the guard ring layer and the insulating layer cannot be compensated.

[0014] In the method according to Patent Document 2, when the trapped charge layer is formed from a thermal oxide layer, the amount of trapped charges is difficult to control, and the trapped charge layer is uniformly formed on the SiC surface; therefore, individual control of the amounts of trapped charges on the n-type layer and the p-type layer cannot be performed.

[0015] Therefore, the optimal ion implantation amount to achieve the design breakdown voltage may fluctuate excessively due to excessive compensation of trapped charges. Furthermore, when fabricating an n-type device, surface roughening due to pitting may degrade the reliability of the semiconductor device when the field oxide layer is formed by wet oxidation.

[0016] The present invention has been conceived in view of the above-mentioned problems, and the object of the present invention is to provide a silicon carbide semiconductor device and a method for producing the same, which ensure an improvement in the withstand voltage performance. Means of solving the problem

[0017] According to the present invention, a silicon carbide semiconductor device comprises a semiconductor substrate, a semiconductor layer, a plurality of field-limiting ring regions, and a field-insulating layer. The n-type semiconductor layer is disposed on the semiconductor substrate, is made of silicon carbide, and has an element region and a terminal region outside the element region. The plurality of p-type field-limiting ring regions are disposed in the terminal region of the semiconductor layer, and the regions are spaced apart from each other.

[0018] The field insulation layer is arranged on the terminal region of the semiconductor layer and is in contact with the field limiting ring regions and the semiconductor layer. Each of the field limiting regions has a halogen-containing field limiting ring region that is in contact with the field insulation layer and contains atoms from the group of halogens.

[0019] According to the present invention, a method for manufacturing a silicon carbide semiconductor device comprises the following steps. An n-type semiconductor layer made of silicon carbide, including an element region and a terminal region outside the element region, is formed on a semiconductor substrate. A plurality of p-type field-limiting ring regions arranged adjacent to each other are formed in the terminal region of the semiconductor layer by implanting acceptor ions.

[0020] An activation annealing process is performed to electrically activate the acceptor ions. By implanting atoms from the group of halogens into a region of each of the field-limiting ring regions, a field-limiting ring region containing halogens is formed. A field-insulating layer is formed on the terminal region of the semiconductor layer. Effects of the invention

[0021] According to the present invention, each of the field-limiting regions has a halogen-containing field-limiting ring region in contact with the field-insulating layer and containing atoms from the halogen group. Halogen atoms have high electronegativity; therefore, halogen atoms are easily negatively charged. A negative charge compensates for a trapped positive charge that exists at the interface between the p-type field-limiting regions and the field-insulating layer.

[0022] This reduces the influence of the trapped positive charges on the breakdown voltage. Therefore, variation in the breakdown voltage performance is kept low. In other words, the resistance of the breakdown voltage performance is improved.

[0023] These and other objects, features, aspects and advantages of the present invention will become more apparent from the following detailed description of the present invention when taken in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The figures show: Fig. 1 is a graph showing a comparison between measurement results and simulation results in a relationship between doses and breakdown voltages for a formation of a field limiting ring when a material of the field insulating layer is silicon dioxide; Fig. 2 is a graph showing a comparison between measurement results and simulation results in a relationship between doses and breakdown voltages for a formation of a field limiting ring when a material of the field insulating layer is polyimide; Fig. 3 is a local sectional view schematically illustrating a configuration of a Schottky barrier diode (SBD) as a silicon carbide semiconductor device according to Embodiment 1 of the present invention; Fig. 4 is a local sectional view showing a step of a method for manufacturing the SBD according to Fig. 1 schematically; Fig. 5 is a local sectional view showing a step of the method for manufacturing the SBD according to Fig. 1 schematically; Fig. 6 is a local sectional view showing a step of the method for manufacturing the SBD according to Fig. 1 schematically; Fig. 7 is a local sectional view showing a step of the method for manufacturing the SBD according to Fig. 1 schematically; Fig. 8 is a local sectional view showing a step of the method for manufacturing the SBD according to Fig. 1 schematically; Fig. 9 is a local sectional view showing a step of the method for manufacturing the SBD according to Fig. 1 schematically; Fig. 10 is a local sectional view showing a step of the method for manufacturing the SBD according to Fig. 1 schematically; Fig. 11 is a local sectional view showing a step of the method for manufacturing the SBD according to Fig. 1 schematically; Fig. 12 is a local sectional view showing a step of the method for manufacturing the SBD according to Fig. 1 schematically; Fig. 13 a flow chart showing the process for producing the SBD according to Fig. 1 represents; Fig. 14 is a flowchart illustrating the method for manufacturing the SBD as a silicon carbide semiconductor device according to Embodiment 2; Fig. 15 is a flowchart illustrating the method for manufacturing an SBD of a modification according to Embodiment 2; Fig. 16 is a local sectional view schematically illustrating a configuration of an SBD as a silicon carbide semiconductor device according to Embodiment 3 of the present invention; Fig. 17 is a flow chart showing the process for producing the SBD according to Fig. 16 represents; Fig. 18 is a local sectional view schematically illustrating a configuration of an SBD as a silicon carbide semiconductor device according to Embodiment 4 of the present invention; Fig. 19 is a flow chart showing the process for producing the SBD according to Fig. 18 represents; Fig. 20 is a local sectional view schematically illustrating a configuration of an SBD as a silicon carbide semiconductor device according to Embodiment 5 of the present invention; Fig. 21 is a local sectional view schematically illustrating the configuration of an SBD as the silicon carbide semiconductor device according to Embodiment 5 of the present invention; Fig. 22 is a local sectional view schematically illustrating a configuration of an SBD as a silicon carbide semiconductor device according to Embodiment 6 of the present invention; Fig. 23 is a local sectional view schematically illustrating the configuration of an SBD as the silicon carbide semiconductor device according to Embodiment 6 of the present invention; Fig. 24 is a local sectional view schematically illustrating a configuration of a MOSFET as a silicon carbide semiconductor device according to Embodiment 7 of the present invention; Fig. 25 is a local sectional view showing a step of a method for manufacturing the MOSFET according to Fig. 24 schematically; Fig. 26 is a local sectional view showing a step of the method for manufacturing the MOSFET according to Fig. 24 schematically; Fig. 27 is a local sectional view showing a step of the method for manufacturing the MOSFET according to Fig. 24 schematically; Fig. 28 is a local sectional view showing a step of the method for manufacturing the MOSFET according to Fig. 24 schematically; Fig. 29 is a local sectional view showing a step of the method for manufacturing the MOSFET according to Fig. 24 schematically; Fig. 30 is a local sectional view showing a step of the method for manufacturing the MOSFET according to Fig. 24 schematically; Fig. 31 is a local sectional view showing a step of the method for manufacturing the MOSFET according to Fig. 24 schematically; Fig. 32 is a local sectional view showing a step of the method for manufacturing the MOSFET according to Fig. 24 schematically; Fig. 33 is a local sectional view showing a step of the method for manufacturing the MOSFET according to Fig. 24 schematically; Fig. 34 is a local sectional view showing a step of the method for manufacturing the MOSFET according to Fig. 24 schematically; Fig. 35 is a local sectional view showing a step of the method for manufacturing the MOSFET according to Fig. 24 schematically; Fig. 36 is a local sectional view showing a step of the method for manufacturing the MOSFET according to Fig. 24 schematically; Fig. 37 is a local sectional view schematically illustrating a configuration of a MOSFET as a silicon carbide semiconductor device according to Embodiment 8 of the present invention; Fig. 38 is a flowchart showing the method for manufacturing the MOSFET according to Fig. 37 represents; Fig. 39 is a local sectional view schematically illustrating a configuration of a MOSFET as a silicon carbide semiconductor device according to Embodiment 9 of the present invention; Fig. 40 is a local sectional view schematically illustrating a configuration of a MOSFET as a silicon carbide semiconductor device of a modification according to Embodiment 9 of the present invention; Fig. 41 is a local sectional view schematically illustrating a configuration of a MOSFET as a silicon carbide semiconductor device according to Embodiment 10 of the present invention; Fig. 42 is a local sectional view schematically illustrating a configuration of a MOSFET as a silicon carbide semiconductor device of a modification according to Embodiment 10 of the present invention. DESCRIPTION OF EMBODIMENTS

[0025] Embodiments of the present invention will be described below with reference to the drawings. Note that identical or equivalent components are denoted by the same reference numerals in the following drawings, and descriptions thereof will not be repeated. Embodiment 1

[0026] Fig. 3 is a local sectional view schematically illustrating a configuration of an SBD (a silicon carbide semiconductor device) 101 according to Embodiment 1. The SBD 101 includes a drift layer (a semiconductor layer) 2, a plurality of field limiting ring regions (FLR regions) 5, and a field insulating layer 7.

[0027] A semiconductor substrate 1 is made of SiC. SiC with a polytype of 4H is preferred. The semiconductor substrate 1 is n-type. The impurity concentration of the semiconductor substrate 1 is, for example, approximately 1 × 10 19 cm -3The semiconductor substrate 1 has an upper surface (a first surface) and a lower surface (a second surface opposite the first surface). The semiconductor substrate 1 has, for example, a thickness of 50 µm or more and 500 µm or less. The upper surface of the semiconductor substrate 1 has an offset angle of several degrees with respect to a plane orientation (0001).

[0028] The drift layer 2 is arranged on the upper surface of the semiconductor substrate 1. The drift layer 2 is made of SiC and has an n-type. The drift layer 2 has an element region RE and a terminal region RT outside the element region RE. The terminal region RT is arranged outside the element region RE and typically encloses the element region RE. A structure for achieving a function of the SBD 101 as an element per se, specifically a function as a diode element, is formed in the element region RE of the drift layer 2.

[0029] The impurity concentration and the thickness of the drift layer are defined by a breakdown voltage specification. For example, the impurity concentration is equal to 5 × 10 14 cm -3 or higher and equal to 1 × 10 18 cm -3 or less, and the thickness is equal to or greater than 4 µm and equal to or less than 100 µm.

[0030] The plurality of field limiting ring regions 5, or FLR regions 5 for short, are arranged in the terminal region RT in the drift layer 2. Each of the plurality of FLR regions 5 is arranged such that it is spaced apart from one another. Each of the FLR regions 5 is separated from the semiconductor substrate 1. Each of the FLR regions 5 has the shape of a ring in a plane layout. Each of the FLR regions 5 is made of SiC. The FLR regions 5 are p-type due to the addition of acceptors. Aluminum or boron, for example, can be used for the acceptors.

[0031] The FLR regions 5 function as an electric field relaxation layer. One of the FLR regions 5 located at the innermost periphery encloses the element region RE in the plane layout. The one of the FLR regions 5 located at the innermost periphery may include a high-concentration region 5a arranged in the shape of a trough. The high-concentration region 5a contains a higher impurity concentration than all the FLR regions 5 except for the high-concentration region 5a.

[0032] The field insulating layer 7 is formed on the terminal region RT and is in contact with the FLR regions 5 and the drift layer 2. The field insulating layer 7 is made of silicon dioxide, for example. An inner end portion of the field insulating layer 7 (the left end portion in the figure) is disposed on the high-concentration region 5a and is separated from the two ends of the high-concentration region 5a. The field insulating layer 7 has a thickness of, for example, 0.5 μm or more and 3 μm or less.

[0033] Each of the FLR regions 5 includes a halogen-containing field limiting ring region 5h that is in contact with the field insulating layer 7 and contains atoms from the group of halogens. The depth of the halogen-containing field limiting ring region 5h is smaller than the depth of the FLR region 5, and is, for example, approximately equal to or greater than 0.2 μm and equal to or less than 1 μm. The halogen-containing field limiting ring region 5h includes a halogen-containing high-concentration region 5ah, which is a part of the high-concentration region 5a. The inner end region (the left end region in the figure) of the halogen-containing high-concentration region 5ah is located outside the inner end region of the high-concentration region 5a.

[0034] The position of an inner end portion of the halogen-containing field-limiting ring region 5h, formed by the high-concentration halogen-containing region 5ha, coincides with the position of the inner end portion of the field-insulating layer 7. The halogen-containing field-limiting ring region 5h forms a negatively charged layer 61 containing atoms from the group of halogens. For example, fluorine (F) atoms or chlorine (Cl) atoms can be used for the atoms from the group of halogens.

[0035] The SBD 101 further comprises the semiconductor substrate 1, a rear ohmic electrode 3, a cathode electrode 4, a Schottky electrode 8, an anode electrode 9, and a protective insulating layer 10. The Schottky electrode 8 is brought into a Schottky junction with the element region RE of the drift layer 2. An end region of the Schottky electrode 8 extends into the interior of the connection region RT and is arranged on the field insulating layer 7. The anode electrode 9 is arranged directly on the Schottky electrode 8.

[0036] The anode electrode 9 has an area larger than the area of ​​the Schottky electrode 8 and covers the Schottky electrode 8. The protective insulating layer 10 covers an edge region of the anode electrode 9, the field insulating layer 7, and an edge region of the drift layer 2. The rear ohmic electrode 3 is brought into ohmic contact with the lower surface of the semiconductor substrate 1. The cathode electrode 4 is in contact with the rear ohmic electrode 3.

[0037] According to the SBD 101 of Embodiment 1, each of the FLR regions 5 has a halogen-containing field confinement ring region 5h in contact with the field insulating layer 7 and containing halogen atoms. Halogen atoms have high electronegativity; therefore, halogen atoms are easily negatively charged. The negative charge-retained layer 61 formed by the negative charge compensates for trapped positive charges existing at the interface between the p-type FLR regions 5 and the field insulating layer 7. Thus, the influence of the trapped positive charge on the breakdown voltage is mitigated.

[0038] Specifically, a dose shift (see the Fig. 1 and Fig. 2) is suppressed, thereby achieving a breakdown voltage close to the design breakdown voltage, assuming no trapped charge. Therefore, variation in breakdown voltage performance attributed to the manufacturing process or manufacturing environment can be minimized. In other words, the robustness of the breakdown voltage performance is improved.

[0039] Next, the process for manufacturing the SBD 101 is described below.

[0040] Referring to Fig. 4, the drift layer 2 is formed on the semiconductor substrate 1. For this purpose, epitaxial growth is carried out, for example, using a chemical vapor deposition (CVD) process.

[0041] Referring to Fig. The FLR regions 5 are formed. For this purpose, impurity ions, such as aluminum or boron to achieve p-type, i.e., acceptor ions, are implanted on the surface side of the drift layer 2. The depth of the ion implantation does not exceed the thickness of the drift layer 2 and is, for example, approximately equal to or greater than 0.2 µm and equal to or less than 3 µm.

[0042] Referring to Fig. 6, the high-concentration region 5a is formed on the one of the FLR regions 5 located at the innermost periphery. For this purpose, a local ion implantation is performed using an implantation mask on the surface of the drift layer 2 where the FLR region 5 has been formed ( Fig. 13: Step S10). The depth of the ion implantation is less than the depth of the FLR region 5. A resist mask or a hard mask formed by a photomechanical process can be used as the implantation mask.

[0043] The hard mask can be formed by depositing a silicon dioxide layer using a CVD process and patterning the layer using a dry etching process. For the ions to be implanted, aluminum or boron, for example, is used. The high-concentration region 5a has a particularly high impurity concentration; therefore, the temperature for ion implantation thereof is preferably 150°C or higher. As a result, the high-concentration region 5a is formed, which has a low sheet resistance.

[0044] Next, a high-temperature annealing process (an activation annealing process) is performed to electrically activate the implanted conductive impurities ( Fig. 13: Step S20). The activation annealing process is performed in an inert atmosphere, such as an argon atmosphere, in a temperature range of 1500°C to 2000°C for a period of 30 seconds to 1 hour. At the time of the activation annealing process, the surface of the drift layer 2 may be covered with a carbon layer to prevent surface roughness.

[0045] Referring to Fig. 7, atoms from the group of halogens are implanted into the surface areas of the FLR regions 5 ( Fig. 13: Step S30). Accordingly, the field-limiting ring regions 5h containing halogens are formed. To selectively implant the atoms from the group of halogens into the FLR regions 5, an implantation mask can be used. The implantation mask can be formed using a photomechanical process. The implantation is performed using ion implantation. The accelerating voltage for the ion implantation is, for example, 30 keV or higher and 700 keV or lower.

[0046] The ion implantation amount is determined so that the trapped negative charge density due to the implanted atoms from the group of halogens is approximately equal to 1 × 10 12 cm -2 or higher and equal to 1 × 10 13 cm -2 or less, preferably approximately equal to 2 × 10 12 cm -2The implantation depth of the halogen atoms does not exceed the thickness of the FLR region 5 and is specified at approximately 0.2 µm or greater and 1 µm or less. Depending on the amount of trapped negative charge introduced, the temperature for ion implantation can be 150 °C or higher.

[0047] Referring to Fig. 8 the field insulation layer 7 is formed ( Fig. 13: Step S40). For this purpose, a deposition process using a CVD method, a photomechanical process, and an etching process are carried out.

[0048] Next, referring to Fig. 9, the back ohmic electrode 3 is formed on the lower surface of the semiconductor substrate 1. Specifically, a metal layer, such as a Ni layer, is formed, and to achieve ohmic connection, the metal layer is subjected to heat treatment at 600°C or higher and 1100°C or lower. The heat treatment can also contribute to the restoration of a crystalline damaged region of the semiconductor that was damaged by the above-described ion implantation of halogen atoms.

[0049] Next, referring to Fig. 10, the Schottky electrode 8 is formed. For this purpose, Ti, Ni, or Mo, for example, is deposited on the surface of the drift layer 2 by a sputtering process, and a photomechanical process and an etching process are performed.

[0050] Next, referring to Fig. 11, the anode electrode 9 is formed on the Schottky electrode 8. For this purpose, Al, for example, is deposited on the surface of the Schottky electrode 8 by a sputtering process, and a photomechanical process and an etching process are performed.

[0051] Next, referring to Fig. 12, the protective insulating layer 10 is formed to cover a portion of the anode electrode 9 and the drift layer 2. For this purpose, a deposition of an insulator, a photomechanical process, and an etching process are performed.

[0052] Next, again on Fig. Referring to Figure 3, the cathode electrode 4 is formed on the backside ohmic electrode 3. Specifically, a metal layer made of, for example, Ti, Ni, Ag, Au, or Al is formed by a sputtering or vapor deposition process. As a result, the SBD 101 is completed.

[0053] According to the present manufacturing process, after an activation tempering process ( Fig. 13: Step S20), which is a heat treatment at a high temperature and over a long period of time, an ion implantation of atoms from the group of halogens is carried out ( Fig. 13: Step S30). This prevents desorption of the implanted halogen atoms during the activation annealing process.

[0054] The trapped charge density formed at the interface between a SiC region and the field insulating layer 7 depends on the material of the field insulating layer 7. Therefore, the ion implantation amount of the atoms from the halogen group is adjusted according to the material of the field insulating layer 7. Embodiment 2

[0055] In the above-described embodiment 1, the ion implantation of atoms from the group of halogens ( Fig. 13: Step S30) after the activation tempering process ( Fig. 13: Step S20), but the sequence can be changed. In this case, too, the layer 61 is fixed with the charge ( Fig. 3) is formed, and the effect of the same is achieved. Referring to Fig. 14, the ion implantation of atoms from the group of halogens (step S30) can be carried out in detail after an implantation of conductive impurities (step S10) and before an activation annealing process (step S20). Or referring to Fig. 15, the ion implantation of atoms from the group of halogens (step S30) can be carried out after an implantation of conductive impurities (step S10) and an activation annealing process (step S20). Embodiment 3

[0056] Fig. 16 is a local sectional view schematically illustrating a configuration of an SBD (silicon carbide semiconductor device) 102 according to Embodiment 3. In the SBD 102, the field insulating layer 7 includes a thermal oxide layer 7a and a CVD layer 7b (a deposition layer). The thermal oxide layer 7a is disposed on the terminal region RT and is in contact with the FLR regions 5 and the drift layer 2. The CVD layer 7b is disposed on the thermal oxide layer 7a.

[0057] Referring to Fig. 17, the step of forming the field insulating layer 7 in Embodiment 3 includes a step of forming the thermal oxide layer 7a (step S41) and a step of forming the CVD layer 7b (step S42). In step S41, for example, the thermal oxide layer 7a is formed on the drift layer 2 by thermal oxidation in a dry atmosphere with a film thickness of 10 nm to 50 nm. This thermal oxidation process is performed in a temperature range of 1000°C to 1200°C.

[0058] This thermal oxidation step also serves as a step for restoring a region of crystalline damage attributed to the ion implantation of halogen atoms to form the charge-trapped layer 61. In step S42, the silicon dioxide CVD layer 7b is deposited on the thermal oxide layer 7a by a CVD process. Then, a laminated body of the thermal oxide layer 7a and the CVD layer 7b is patterned by a photomechanical process and an etching process.

[0059] The configuration except for the above is substantially the same as that of Embodiments 1 and 2 described above, therefore, the same or corresponding elements are denoted by the same reference numerals, and a description thereof will not be repeated. Embodiment 4

[0060] Fig. 18 is a local sectional view schematically illustrating a configuration of an SBD (a silicon carbide semiconductor device) 103 according to Embodiment 4. As with the SBD 102 ( Fig. 16: Embodiment 3), the field insulating layer 7 in the SBD 103 also includes a thermal oxide layer 7a. In Embodiment 4, unlike Embodiment 3, the thermal oxide layer 7a includes a halogen-containing field insulating region 7ah. The halogen-containing field insulating region 7ah contains atoms from the group of halogens and is located on the halogen-containing field limiting regions 5h.

[0061] Therefore, in the SBD 103, a negative charge trapped layer 62 caused by atoms from the group of halogens has regions formed from the field limiting ring regions 5h containing halogens and has regions formed from the field insulating regions 7ah containing halogens on the field limiting ring regions 5h containing halogens.

[0062] Referring to Fig. 19, in Embodiment 4, unlike Embodiment 3, the implantation of halogen atoms (step S30) is performed after formation of the thermal oxide layer 7a (step S41). Accordingly, the halogen atoms are implanted into the drift layer 2 through the thermal oxide layer 7a. As a result, not only the halogen-containing field limiting ring regions 5h are formed, but also the halogen-containing field isolation regions 7ah are formed in the thermal oxide layer 7a. The CVD layer 7b is formed on the thermal oxide layer 7a. Thereafter, a laminated body of the thermal oxide layer 7a and the CVD layer 7b is patterned by a photomechanical process and an etching process.

[0063] The configuration except for the above is substantially the same as that of Embodiment 3 described above, therefore, the same or corresponding elements are denoted by the same reference numerals, and a description thereof will not be repeated.

[0064] According to Embodiment 4, the field insulating layer 7 includes the halogen-containing field insulating regions 7ah containing atoms from the group of halogens and located on the halogen-containing field limiting ring regions 5h. In this case, the halogen-containing field limiting ring regions 5h are formed by ion implantation through the field insulating layer 7, specifically the thermal oxide layer 7a. As a result, the atoms from the group of halogens are prevented from desorbing from the halogen-containing field limiting ring regions 5h due to the influence of the formation of the field insulating layer 7, in particular the influence of the heat treatment.

[0065] It should be noted that the heat treatment for forming the thermal oxide layer 7a in Embodiment 4 does not serve as a step for restoring a crystalline damage region attributed to the ion implantation of the halogen group atoms. However, the heat treatment for forming the back ohmic electrode 3 is performed after the implantation of the halogen group atoms; therefore, this heat treatment also contributes to the restoration of a crystalline damage region in Embodiment 4. Embodiment 5

[0066] Fig. 20 is a local sectional view schematically illustrating a configuration of an SBD (silicon carbide semiconductor device) 104 according to Embodiment 5. In the SBD 104, the drift layer 2 includes halogen-containing semiconductor regions 2h. Each of the halogen-containing semiconductor regions 2h contains atoms from the group of halogens and is in contact with the field insulating layer 7 between the FLR regions 5.

[0067] Therefore, in the SBD 104, a negative charge trapped layer 63 caused by atoms from the group of halogens has regions formed from the field limiting ring regions 5h containing halogens and has regions formed from the semiconductor regions 2h containing halogens that are adjacent to the field limiting ring regions 5h containing halogens.

[0068] The configuration except for the above is substantially the same as that of Embodiment 1 described above, so the same or corresponding elements are denoted by the same reference numerals, and a description thereof will not be repeated. The same effects as in Embodiment 1 are also achieved in Embodiment 5.

[0069] Furthermore, according to Embodiment 5, the negative charge-retaining layer 63 caused by atoms from the group of halogens includes the halogen-containing semiconductor regions 2h in the drift layer 2. As a result, a positive charge attributed to donor ions in the drift layer 2 is compensated. Therefore, the donor concentration in the drift layer 2 is effectively reduced. Therefore, a depletion layer easily spreads into the drift layer 2.

[0070] Therefore, the electric field concentration at the edge regions of the FLR regions 5 is relaxed. The occurrence of the electric field concentration at the edge regions of the FLR regions 5 is particularly likely when the impurity concentration of the FLR regions 5 increases due to a variation in ion implantation; according to Embodiment 5, resistance to such a process variation is improved.

[0071] Even if a positive charge from the outside reaches the interface between the field insulation layer 7 and the drift layer 2, the electric flux lines from the positive charge are limited by a negative charge in the halogen-containing semiconductor regions 2h. Accordingly, the suppression of a breakdown voltage fluctuation attributed to the positive electric charge from the outside is ensured without changing the implantation amount of the acceptor ions forming the FLR regions 5.

[0072] Referring to Fig. 21, in an SBD (silicon carbide semiconductor device) 105 of a modification as in Embodiment 4, the field insulating layer 7 includes a thermal oxide layer 7a and a CVD layer 7b, and a thermal oxide layer 7a includes a field insulating region 7ah containing halogens.

[0073] Therefore, in the SBD 105, a negative charge-retaining layer 64 caused by atoms from the group of halogens includes a region formed by the halogen-containing field isolation region 7ah. Note that, as in Embodiment 3, it is possible for the halogen-containing field isolation region 7ah not to be disposed in the thermal oxide layer 7a. Embodiment 6

[0074] Fig. 22 is a local sectional view schematically illustrating a configuration of an SBD (a silicon carbide semiconductor device) 106 according to Embodiment 6. In the SBD 106, each halogen-containing semiconductor region 2h has a pinned negative charge density larger than the pinned negative charge density of each halogen-containing field-limiting ring region 5h.

[0075] Therefore, in the SBD 106, a negative charge trapped layer 65 caused by atoms from the group of halogens has regions formed from the halogen-containing field-limiting ring regions 5h and regions formed from the halogen-containing semiconductor regions 2h located adjacent to the halogen-containing field-limiting ring regions 5h, and the latter regions have a higher trapped negative charge density. It should be noted that the trapped charge density of each region of the charge trapped layer 65 is preferably in a range of 1 × 10 12 cm -2 or more and of 1 × 10 13 cm -2 or less.

[0076] The dose level for forming the halogen-containing semiconductor regions 2h is preferably greater than the dose level for forming the halogen-containing field-limiting ring regions 5h. In other words, the density of atoms from the group of halogens per unit area in each halogen-containing semiconductor region 2h is preferably higher than the density of atoms from the group of halogens per unit area in each halogen-containing field-limiting ring region 5h.

[0077] According to Embodiment 6, each halogen-containing semiconductor region 2h has a pinned negative charge density higher than the pinned negative charge density of each halogen-containing field-limiting ring region 5h. Accordingly, the effect described in Embodiment 5, by which the electric field concentration at each edge region of the FLR region 5 is relaxed, is enhanced. Furthermore, the effect described in Embodiment 5, by which the fluctuation in the breakdown voltage attributed to the external positive charge is kept low, is enhanced.

[0078] Referring to Fig. 23, in an SBD (silicon carbide semiconductor device) 107 of a modification as in Embodiment 4, the field insulating layer 7 includes a thermal oxide layer 7a and a CVD layer 7b. The thermal oxide layer 7a includes a halogen-containing field insulating region forming a negative charge-pinning region of the layer 66 caused by atoms from the group of halogens. The halogen-containing field insulating region includes halogen-containing field insulating regions 7ahp on the halogen-containing field-limiting ring regions 5h and halogen-containing field insulating regions 7ahn on the halogen-containing semiconductor regions 2h.

[0079] The pinned negative charge density of each halogen-containing field insulating region 7ahn is higher than the pinned negative charge density of each halogen-containing field insulating region 7ahp. Note that, as in Embodiment 3, it is possible for the halogen-containing field insulating regions 7ahp and the halogen-containing field insulating regions 7ahn not to be arranged in the thermal oxide layer 7a. Embodiment 7

[0080] Fig. 24 is a local sectional view schematically illustrating a configuration of a MOSFET (a silicon carbide semiconductor device) 201 according to Embodiment 7. The MOSFET 201 has, in a manner similar to the SBD 101 ( Fig. 3: Embodiment 1) comprises a semiconductor substrate 1, a drift layer 2, a back ohmic electrode 3, a plurality of FLR regions 5, a field insulating layer 7, and a protective insulating layer 10.

[0081] In a similar manner to the SBD 101, the negative charge pinned layer 71 in the MOSFET 201 is formed by the halogen-containing field limiting ring regions 5h of the FLR regions 5. Accordingly, the withstand voltage breakdown performance is improved in a similar manner to Embodiment 1.

[0082] The MOSFET 201 further includes a well region 11, a terminal well region 12, a high-concentration well region 13, a source region 14, a source contact electrode 16, an interlayer insulating layer 17, a source electrode 19, a gate electrode 20, a gate insulating layer 21, a gate wiring 22, a field stop region 24, and a drain electrode 54.

[0083] The terminal well region 12 is disposed on a region adjacent to the terminal region RT in the element region RE of the drift layer 2. The terminal well region 12 is separated from the semiconductor substrate 1 and has a depth of approximately 0.2 µm to 3 µm. The terminal well region 12 is disposed adjacent to the inner side of one of the FLR regions 5 located at the innermost periphery. The terminal well region 12 is made of SiC. The terminal well region 12 is p-type by adding acceptors. Aluminum or boron, for example, are used for the acceptors.

[0084] The terminal well region 12 has an impurity concentration higher than the impurity concentration of the FLR regions 5. The terminal well region 12 has a high-concentration terminal well region 12a arranged in the shape of a well. The high-concentration terminal well region 12a has a higher impurity concentration than the terminal well region 12 except for the high-concentration terminal well region 12a.

[0085] The negative charge-containment layer 71 may include, in addition to the region formed by the halogen-containing field-limiting ring regions 5h, a region formed by the halogen-containing terminal well region 12h. The halogen-containing terminal well region 12h is a region in contact with the field insulating layer 7 and containing atoms from the group of halogens in the terminal well region 12. The halogen-containing terminal well region 12h may include a halogen-containing high-concentration terminal well region 12ah, which is a high-concentration region of the terminal well region 12a.

[0086] The well region 11 is arranged on the element region RE of the drift layer 2 and is arranged according to the respective unit cell of the MOSFET. The well region 11 is separated from the semiconductor substrate 1 and has a depth of approximately 0.2 µm to 3 µm. The well region 11 is made of SiC. The well region 11 has a p-type by adding acceptors. For the acceptors, aluminum or boron, for example, is used. The well region 11 has an impurity concentration that is higher than the impurity concentration of the FLR regions 5 and has a value in a range of 1 × 10 15 cm -3 up to 1 × 10 19 cm -3 on.

[0087] Well region 11 may have the same impurity concentration as the terminal well region 12. A p-type high-concentration well region 13 and an n-type source region 14 are selectively formed on well region 11. A source contact electrode 16 is ohmically connected to the surface of the high-concentration well region 13 and the source region 14.

[0088] The source electrode 19 is connected to the source contact electrode 16 on the well region 11 through the source contact hole 18 arranged in the insulating interlayer 17. The source electrode 19 is connected to the source contact electrode 16 on the terminal well region 12 through a terminal well contact hole 18a arranged in the insulating interlayer 17 and the field insulating layer 7.

[0089] The gate electrode 20 is disposed on the well region 11 with the gate insulating layer 21 interposed therebetween. The gate electrode 20 has a region disposed on the field insulating layer 7. This region is disposed on the halogen-containing terminal well region 12h through the field insulating layer 7 interposed therebetween. Specifically, the region is disposed on the halogen-containing terminal well region 12ah via the field insulating layer 7 with a high concentration. The gate insulating layer 21 has a thickness of 100 µm or less.

[0090] The insulating interlayer 17 is disposed on the gate electrode 20. The gate wiring 22 is connected to a region of the gate electrode 20 disposed on the field insulating layer 7 through the gate contact hole 23. The field stop region 24 is of n-type. The field stop region 24 is disposed at a chip end portion of the MOSFET 201. The field stop region 24 is separated from and surrounds one of the FLR regions 5 located at the outermost periphery. In Embodiment 7, the protective insulating layer 10 covers the source electrode 19, the insulating interlayer 17 on the field insulating layer 7, the gate wiring 22, and the drift layer 2. The drain electrode 54 is in contact with the backside ohmic electrode 3.

[0091] Next, the method for manufacturing the MOSFET 201 is described below.

[0092] Referring to Fig. 25, in a manner similar to the above-described embodiment 1, the drift layer 2 is formed on the semiconductor substrate 1. Next, alignment marks (not shown) are formed on the surface of the drift layer 2 by means of a photomechanical process and a dry etching process. Next, referring to Fig. 26, a resist mask (not shown) is formed on the surface of the drift layer 2 by a photomechanical process. Next, in the same manner as in Embodiment 1, local ion implantation is performed using the resist mask as an implantation mask. Accordingly, the FLR regions 5 are formed.

[0093] Referring to Fig. 27, an implantation mask (not shown) is reworked, and local ion implantation is performed using the mask. Accordingly, the well region 11 and the terminal well region 12 are formed. Both regions can be formed simultaneously.

[0094] Referring to Fig. 28, an implantation mask (not shown) is reworked, and local ion implantation is performed using the mask. Accordingly, the source region 14 and the field stop region 24 are formed. Both regions can be formed simultaneously. The ions to be implanted are donor ions, for example, nitrogen or phosphorus. The concentration of the ions to be implanted is in the range of 1 × 10 18 cm -3 up to 1 × 10 21 cm -3and extends beyond the impurity concentration of well region 11, which has the opposite conductivity type. The depth of the ion implantation is shallower than the depth of well region 11.

[0095] Referring to Fig. 29, an implantation mask (not shown) is reworked, and local ion implantation is performed using the mask. Accordingly, the high-concentration well region 13 and the high-concentration terminal well region 12a are formed. Both regions can be formed simultaneously. A hard mask is preferably used as the implantation mask, although a resist mask formed by a photomechanical process can also be used.

[0096] The hard mask is, for example, a silicon dioxide layer formed by a CVD process and a dry etching process. The temperature for ion implantation is preferably 150°C or higher, so that the high-concentration well region 13 and the high-concentration terminal well region 12a are obtained with low sheet resistance. The depth of the ion implantation does not exceed the thickness of the well region 11 or the terminal well region 12.

[0097] Next, in the same manner as in Embodiment 1, a high-temperature annealing process (an activation annealing process) is performed to electrically activate the implanted conductive impurities. The order of the above-described plurality of ion implantation steps is arbitrary.

[0098] Referring to Fig. 30, atoms from the group of halogens are implanted into the surface regions of the FLR regions 5 and the surface region of the terminal well region 12 adjacent to the FLR region 5. Thus, the halogen-containing field limiting ring regions 5h and the halogen-containing terminal well region 12h are formed. The implantation is performed by ion implantation. The ion implantation amount is determined such that the trapped negative charge density due to the implanted atoms from the group of halogens is approximately equal to 1 × 10 12 cm -2 or higher and equal to 5 × 10 12 cm -2 or less, preferably equal to 2 × 10 12 cm -2 .

[0099] The implantation depth of the halogen atoms does not exceed the thickness of the FLR region 5 or the terminal well region 12 and is specified to be approximately 0.2 µm or greater and 1 µm or less. Depending on the amount of trapped negative charge introduced, the temperature for ion implantation can be 150°C or higher. The trapped negative charge layer 71 is formed by the above-described implantation of the halogen atoms.

[0100] The negative charge trapped layer 71 is disposed only under the field insulating layer 7, which is to be formed by a process to be described later. Therefore, the charge trapped layer 71 is not formed on the inner peripheral side of the terminal well region 12. The region where the charge trapped layer 71 is not formed, as described above, includes the inner peripheral side of the high-concentration terminal well region 12a.

[0101] Referring to Fig. 31, the field insulation layer 7 is formed. Specifically, a silicon dioxide layer is deposited using a CVD process. The layer is then structured using a photomechanical process and an etching process.

[0102] Referring to Fig. 32, the gate insulating layer 21 made of silicon dioxide is formed, for example, by thermal oxidation or a CVD process. Next, the gate electrode 20 is formed. Specifically, a polycrystalline silicon layer is deposited using a CVD process. The layer is structured using a photomechanical process and an etching process.

[0103] Referring to Fig. 33, for example, an insulating interlayer 17 is deposited on the semiconductor layer provided with the field insulating layer 7, the gate insulating layer 21, and the gate electrode 20 using a CVD process. The layer is patterned using a photomechanical process and an etching process. This forms the source contact hole 18. During patterning, a terminal well contact hole 18a is also formed, which penetrates the insulating interlayer 17 and the field insulating layer 7 and extends to the high-concentration terminal well region 12a.

[0104] Referring to Fig. 34, the source contact electrode 16 is formed at the bottom of the source contact hole 18 and the terminal well contact hole 18a. Specifically, a metal layer, such as a Ni layer, is first formed. A nickel silicide layer is then formed by heat treatment at 600°C to 1100°C. The unreacted metal layer on the interlayer insulating layer 17 is then removed by etching. This forms the source contact electrode 16. Using the same method, the back ohmic electrode 3 is formed on the back surface of the semiconductor substrate 1.

[0105] Referring to Fig. 35, the source electrode 19 and the gate wiring 22 are formed on the insulating interlayer 17. Specifically, the gate contact hole 23 is first formed using a photomechanical process and an etching process. Subsequently, a metal layer, made of aluminum, for example, is formed using a sputtering process or a vapor deposition process. A photomechanical process and an etching process are performed again to impart a structure to the metal layer. This results in the source electrode 19 and the gate wiring 22.

[0106] Referring to Fig. 36, the protective insulating layer 10 is formed by means of layer formation and structuring using a photomechanical process and an etching process.

[0107] Referring again to Fig. 24, a metal layer, such as titanium, nickel, silver, gold, or aluminum, is formed on the backside ohmic electrode 3 using a sputtering process or a vapor deposition process. The drain electrode 54 is formed accordingly. As a result, the MOSFET 201 is completed.

[0108] According to Embodiment 7, substantially the same effects as those of Embodiment 1 are also achieved in the MOSFET. According to the manufacturing method described above, a region with crystalline damage resulting from the implantation step of atoms from the group of halogens ( Fig. 30) is restored by a heating step carried out thereafter.

[0109] The step comprising heating to an extent effective for restoring crystallinity may be the thermal oxidation step for forming the gate insulating layer 21 or a silicidation step for forming the source contact electrode 16 and the back ohmic electrode 3. Embodiment 8

[0110] Fig. 37 is a local sectional view schematically illustrating a configuration of a MOSFET (a silicon carbide semiconductor device) 202 according to Embodiment 8. In the MOSFET 202, the field insulating layer 7 has a thickness of 10 mm, unlike the MOSFET 201 ( Fig. 24: Embodiment 7) Halogen-containing field insulating regions 7h. The halogen-containing field insulating region 7h contains atoms from the group of halogens and is located on the halogen-containing field limiting ring region 5h and the halogen-containing terminal well region 12h.

[0111] Therefore, in the MOSFET 202, a negative charge trapped layer 72 caused by atoms from the group of halogens has regions formed from the field limiting ring regions 5h containing halogens and has regions formed from the field insulating regions 7h containing halogens on the field limiting ring regions 5h containing halogens.

[0112] Further referring to Fig. 38, the implantation of atoms from the group of halogens (step S30) in the manufacture of the MOSFET 202 is performed after the field insulating layer 7 is formed (step S40), the gate insulating layer 21 and the gate electrode 20 are formed (step S45), and before the interlayer insulating layer 17 is formed (step S50). In the above-described Embodiment 7, the implantation of atoms from the group of halogens (step S30) is performed before the field insulating layer 7 is formed (step S40).

[0113] In the step of implanting atoms from the group of halogens, which is performed after the formation of the field insulating layer 7, it is necessary that a portion of the atoms from the group of halogens penetrate the field insulating layer 7. Therefore, in Embodiment 8, the field insulating layer 7 must be sufficiently thin to allow sufficient penetration of atoms from the group of halogens, and it preferably has a thickness of, for example, about 100 nm or more and 500 nm or less.

[0114] According to Embodiment 8, unlike Embodiment 7, the step of forming the gate insulating layer 21 and the gate electrode 20 is not performed after the implantation of atoms from the group of halogens.

[0115] Accordingly, desorption of halogen atoms attributed to this step is avoided. In particular, in the case where a thermal oxidation process involving extended high-temperature treatment is used to form the gate insulating layer 21, desorption of halogen atoms is likely to occur if the halogen atoms are implanted before the treatment.

[0116] In Embodiment 8, the thermal oxidation step for forming the gate insulating layer 21 does not contribute to the recovery of a crystalline damage region attributed to the implantation step of halogen atoms. However, during heating in the silicidation step for forming the source contact electrode 16 and the back ohmic electrode 3, the crystalline damage region is recovered to some extent. Embodiment 9

[0117] Fig. 39 is a local sectional view schematically illustrating a configuration of a MOSFET (a silicon carbide semiconductor device) 203 according to Embodiment 9. In the MOSFET 203, the drift layer 2 has a similar configuration to the SBD 104 ( Fig. 20: Embodiment 5) Halogen-containing semiconductor regions 2h.

[0118] Therefore, in the MOSFET 203, a negative charge trapping layer 73 caused by atoms from the group of halogens includes regions formed by the halogen-containing semiconductor regions 2h adjacent to the regions formed by the halogen-containing field-limiting ring regions 5h. This configuration enables substantially the same effects to be achieved in the MOSFET as in Embodiment 5.

[0119] Referring to Fig. 40, the field insulating layer 7 of the MOSFET (silicon carbide semiconductor device) 204 of a modification has a halogen-containing field insulating region 7h on the halogen-containing field limiting ring regions 5h and the halogen-containing semiconductor regions 2h.

[0120] Therefore, in the MOSFET 204, the negative charge trapping layer 74 caused by the halogen atoms includes a region formed by the field isolation region 7h containing halogens. The MOSFET 204 is manufactured by substantially the same manufacturing method as the manufacturing method described in Embodiment 8. Accordingly, desorption of the implanted halogen atoms is suppressed. Embodiment 10

[0121] Fig. 41 is a local sectional view schematically illustrating a configuration of a MOSFET (a silicon carbide semiconductor device) 205 according to Embodiment 10. In the MOSFET 204, each halogen-containing semiconductor region 2h has a silicon carbide semiconductor device 2h in a manner similar to the SBD 106 ( Fig. 22: Embodiment 6) has a trapped negative charge density higher than the trapped negative charge density of each halogen-containing field limiting ring region 5h.

[0122] Therefore, in the MOSFET 205, a negative charge trapped layer 75 caused by atoms from the group of halogens includes regions formed by the halogen-containing field-limiting ring regions 5h and regions formed by the halogen-containing semiconductor regions 2h adjacent to the halogen-containing field-limiting ring regions 5h, and the latter regions have a higher trapped negative charge density. This configuration enables substantially the same effects to be achieved in the MOSFET as in Embodiment 6.

[0123] Referring to Fig.42, the field insulating layer 7 of the MOSFET (silicon carbide semiconductor device) 206 of a modification includes halogen-containing field insulating regions 7hp on the halogen-containing field limiting ring regions 5h and the halogen-containing terminal well region 12h. Furthermore, the field insulating layer 7 includes halogen-containing field insulating regions 7hn on the halogen-containing semiconductor regions 2h. Therefore, the negative charge pinning layer 76 caused by atoms from the group of halogens in the MOSFET 206 includes the halogen-containing field insulating regions 7hp and the halogen-containing field insulating regions 7hn.

[0124] The pinned negative charge density of each halogen-containing field isolation region 7hn is higher than the pinned negative charge density of each halogen-containing field isolation region 7hp. MOSFET 206 is manufactured by substantially the same method as the manufacturing method described in Embodiment 8. Accordingly, desorption of the implanted atoms from the group of halogens is suppressed.

[0125] In Embodiments 1 to 6 described above, Schottky barrier diodes or SBDs 101 to 107 are described; however, as a modification, a diode element other than an SBD may also be configured. Specifically, a junction barrier Schottky (JBS) diode can be formed by disposing a p-type semiconductor layer region in a portion of the drift layer 2 under a Schottky electrode 8, which is an active region.

[0126] Alternatively, a mixed-pin Schottky diode, or MPS diode, can be configured with the Schottky electrode 8, which has both a Schottky junction and an ohmic junction to the drift layer 2. Alternatively, a pin (PiN) diode can be configured by disposing a p-type semiconductor region in the drift layer 2 under the Schottky electrode 8, which is the active region, and disposing an ohmic contact electrode instead of the Schottky electrode 8.

[0127] In Embodiments 7 to 10 described above, the MOSFETs 201 to 205 are described as planar type; however, as a modification, a trench type MOSFET may also be configured. Alternatively, a metal-insulator-semiconductor field-effect transistor (MISFET) may be configured in addition to the MOSFET. Alternatively, a junction FET (JFET) or a metal-semiconductor field-effect transistor (MESFET) may be configured instead of the MISFET. Alternatively, the IGBT may be configured instead of the FET.

[0128] In each of the above embodiments, a semiconductor device using SiC as a semiconductor material is described. However, semiconductor materials other than SiC may be used as long as charges are trapped at the interface between the semiconductor layer and the insulating layer. For example, a compound semiconductor material such as GaN may be used.

[0129] It should be noted that embodiments and modifications of the present invention can be arbitrarily combined and, where appropriate, modified or features omitted without departing from the scope of the invention. Although the invention has been described in detail, the foregoing description is only illustrative and not restrictive in all aspects. It is understood that numerous other modifications and variations can be devised without departing from the scope of the invention. EXPLANATION OF REFERENCE SYMBOLS RE element area RT connection area 101 to 107 SBD (silicon carbide semiconductor unit) 201 to 206 MOSFET (silicon carbide semiconductor device) 1 semiconductor substrate 2 Drift layer (semiconductor layer) 2h Halogen-containing semiconductor region 3 rear ohmic electrode 4 Cathode electrode 5 Field limiting ring area (FLR area) 5a Area with high concentration 5h Halogen-containing field limiting ring region 5ah halogen-containing area with high concentration 7 Field insulation layer 7a Layer of a thermal oxide 7b CVD layer (deposition layer) 7h, 7ah, 7hp field isolation area containing halogens 7ahn, 7ahp halogen-containing field insulation region 8 Schottky electrode 9 Anode electrode 10 Protective insulation layer 11 Trough area 12 Connection trough area 12a Connection trough area with high concentration 12h halogen-containing connection recess area 12ah halogen containing connection recess area with 13 high concentration Trough area with high concentration 14 Source area 16 Source contact electrode 17 insulating intermediate layer 18 source contact hole 18a terminal well contact hole 19 Source electrode 20 Gate electrode 21 Gate insulating layer 22 Gate wiring 23 Gate contact hole 24 Field stop area 54 Drain electrode 61 to 66 shift with retained charges 71 to 76 layer with retained charges

Claims

[1] Silicon carbide semiconductor device comprising: - a semiconductor substrate (1); - a semiconductor layer (2) having an n-type, arranged on the semiconductor substrate (1), made of silicon carbide and having an element region (RE) and a terminal region (RT) outside the element region (RE); - a plurality of field-limiting ring regions (5) which have a p-type, are arranged in the connection region (RT) of the semiconductor layer (2) and are arranged at a distance from one another; and - a field insulation layer (7) arranged on the connection region (RT) of the semiconductor layer (2) and in contact with the field limiting ring regions (5) and the semiconductor layer (2), each of the field limiting ring regions (5) having a field limiting ring region (5h) containing halogens, the field limiting ring region (5h) containing halogens being in contact with the field insulation layer (7) and containing atoms from the group of halogens. [2] A silicon carbide semiconductor device according to claim 1, wherein the field insulating layer (7) has a halogen-containing field insulating region (7h, 7ah, 7hn, 7hp, 7ahn, 7ahp) which is located on the halogen-containing field limiting ring region (5h) and contains atoms from the group of halogens. [3] A silicon carbide semiconductor device according to claim 1 or 2, wherein the semiconductor layer (2) has a halogen-containing semiconductor region (2h) which is in contact with the field insulating layer (7) between the field limiting ring regions (5) and contains atoms from the group of halogens. [4] A silicon carbide semiconductor device according to claim 3, wherein the halogen-containing semiconductor region (2h) has a pinned negative charge density higher than a pinned negative charge density of the halogen-containing field limiting ring region (5h). [5] A method for producing a silicon carbide semiconductor device, comprising the following steps: - forming a semiconductor layer (2) on a semiconductor substrate (1), wherein the semiconductor layer (2) has an n-type, is made of silicon carbide and has an element region (RE) and a terminal region (RT) outside the element region (RE); - forming a plurality of field-limiting ring regions (5) in the connection region (RT) of the semiconductor layer (2) by implanting acceptor ions, wherein the plurality of field-limiting ring regions (5) have a p-type and the regions are arranged at a distance from one another; - performing an activation annealing process to electrically activate the acceptor ions; - forming a field-limiting ring region (5h) containing halogens by implanting atoms from the group of halogens into a region of each of the field-limiting ring regions (5); and - Forming a field insulation layer (7) on the connection region (RT) of the semiconductor layer (2). [6] A method for producing a silicon carbide semiconductor device according to claim 5, wherein the step of performing an activation annealing process for electrically activating the acceptor ions is performed before the step of forming the field limiting ring region (5h) containing halogens. [7] A method for producing a silicon carbide semiconductor device according to claim 5, wherein the step of performing an activation annealing process for electrically activating the acceptor ions is performed after the step of forming the field limiting ring region (5h) containing halogens. [8] A method for manufacturing a silicon carbide semiconductor device according to any one of claims 5 to 7, wherein the field insulating layer (7) comprises a layer (7a) of a thermal oxide. [9] A method for producing a silicon carbide semiconductor device according to any one of claims 5 to 8, wherein at least a portion of the field insulating layer (7) is formed before the step of forming the field limiting ring region (5h) containing halogens, and the atoms from the group of halogens are implanted through the at least a portion of the field insulating layer (7) in the step of forming the field limiting ring region (5h) containing halogens.

Citation Information

Patent Citations

  • Semiconductor device and its manufacturing method

    JP2003282888A

  • Edge termination structure for silicon carbide device and manufacturing method of silicon carbide device including edge termination structure

    JP2013062518A

  • Semiconductor device and method of manufacturing the same

    JP2004158603A

  • Semiconductor device

    US20160049484A1

  • JP002004158603A