Semiconductor device and manufacturing method of the same

A dual-channel region structure with varying acceptor concentrations in n-channel planar vertical MOSFETs addresses the challenge of setting threshold values and resistance, achieving both desired threshold and low resistance through ion implantation, improving mobility and reducing unintended impurity introduction.

JP2024156495AActive Publication Date: 2024-11-06DENSO CORP +2
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
JP2023071005
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-04-24
Publication Date
2024-11-06
Estimated Expiration
2043-04-24

AI Technical Summary

Technical Problem

Existing n-channel planar vertical MOSFETs face challenges in achieving both a desired threshold value and low resistance due to the trade-off between impurity concentration and mobility, particularly when applying techniques from trench gate structures to planar designs.

Method used

The semiconductor device employs a dual-channel region structure with different acceptor concentrations in the first and second channel regions, allowing independent control of the threshold value and reducing resistance through ion implantation of boron ions, thereby forming the channel regions without buried growth.

Benefits of technology

This approach enables the semiconductor device to achieve both a desired threshold value and low resistance by independently setting the threshold value of the first channel region and reducing channel resistance in the second channel region, enhancing mobility and suppressing unintended impurity introduction.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024156495000001_ABST
    Figure 2024156495000001_ABST
Patent Text Reader

Abstract

To provide a semiconductor device that provides a vertical type semiconductor element of an n-type channel planar type capable of achieving both of a desire threshold value and a low resistance.SOLUTION: Each channel region 6 is constructed by a first channel region 6a and a second channel region 6b that are adjusted to each different acceptor concentration. For the first channel region 6a, a control of the threshold value can be performed so as to be separated from the second channel region 6b, and can be set to a preferable threshold value as a vertical type MOSFET. For the second channel region 6b, it can be set to an acceptor concentration that is lower than that of the first channel region 6a. Thus, a semiconductor device capable of achieving both of a desire threshold value and the low resistance can be provided.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present disclosure relates to a semiconductor device having an n-type channel planar vertical semiconductor element made of gallium nitride (hereinafter referred to as GaN), and a method for manufacturing the same. [Background technology]

[0002] In a GaN n-channel planar vertical MOSFET, the threshold voltage can be increased by increasing the impurity concentration in the p-type region that constitutes the channel region. However, the high impurity concentration reduces mobility and increases channel resistance.

[0003] On the other hand, Patent Document 1 discloses a trench-gate vertical MOSFET that has a high threshold voltage and a reduced channel resistance. This vertical MOSFET is constructed by epitaxially growing a first n-layer, a first p-layer, a second p-layer, and a second n-layer in this order on an n-type substrate, forming a trench from the surface of the second n-layer to the first n-layer, and forming a gate electrode in the trench via a gate insulating film. The lower first p-layer has a low acceptor concentration, and the upper second p-layer has a high acceptor concentration, thereby obtaining a high threshold voltage in the high acceptor concentration second p-layer and reducing the channel resistance in the low acceptor concentration first p-layer. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2021-190578 A Summary of the Invention [Problem to be solved by the invention]

[0005] However, when attempting to apply the technology of Patent Document 1 to a planar-type vertical semiconductor MOSFET, the following problems arise.

[0006] First, each layer is formed by epitaxial growth, but if it is to be used in a planar-type vertical MOSFET, the channel part will be composed of a second p-layer epitaxial layer with a high acceptor concentration. In that case, a recess is formed between the n-type source regions formed on both sides of the channel part in advance, and the first p-layer is constructed by embedding the epitaxial layer in the recess, and unintended n-type impurities are introduced during the recess formation. This makes it difficult to set the desired threshold value and makes it impossible to achieve low resistance.

[0007] In addition, it is essential to reduce the resistance of the p-type contact in order to reduce switching loss. In the case of the vertical MOSFET of Patent Document 1, the second p layer also functions as the p-type contact, but if the impurity concentration is increased too much to increase the acceptor concentration, the threshold voltage becomes higher than necessary, and the mobility decreases, making it impossible to achieve low resistance.

[0008] An object of the present disclosure is to provide a semiconductor device having an n-type channel planar type vertical semiconductor element that can achieve both a desired threshold voltage and low resistance, and a method for manufacturing the same. [Means for solving the problem]

[0009] A first aspect of the present disclosure is a semiconductor device having an n-type channel planar vertical semiconductor element made of GaN, An n-type or p-type GaN substrate (1) having a front surface (1a) and a back surface (1b); An n-type drift region (2) formed on the surface of the GaN substrate and having a lower donor concentration than the GaN substrate; a p-type base region (3) formed on the drift region; a p-type contact region (4) provided on the base region or on a surface portion of the base region; an n-type source region (5) formed on the base region and having a higher donor concentration than the drift region; a p-type channel region (6) formed on the base region and disposed on the opposite side of the contact region with the source region therebetween; a JFET region (7) disposed on the drift region and adjacent to the base region and the channel region; a gate insulating film (8) formed on the channel region; a gate electrode (9) formed on a gate insulating film above the channel region; a source electrode (10) connected to the contact region and the source region; A drain electrode (11) connected to the back surface of the GaN substrate, The channel region has a lower acceptor concentration than the contact region, and includes a first channel region (6a) and a second channel region (6b) having different acceptor concentrations in the channel length direction.

[0010] In this way, the first channel region and the second channel region are adjusted to different acceptor concentrations. Therefore, the threshold of the first channel region can be controlled independently of the second channel region, and a suitable threshold can be set for the vertical semiconductor element. Also, the second channel region can be set to a lower acceptor concentration than the first channel region. Therefore, the channel resistance of the second channel region can be reduced when the vertical semiconductor element is turned on, and high mobility can be obtained. Thus, a semiconductor device can be obtained that can achieve both a desired threshold and low resistance.

[0011] Another aspect of the present disclosure is a method for manufacturing a semiconductor device having an n-type channel planar vertical semiconductor element made of GaN, the method comprising the steps of: Providing an n-type or p-type GaN substrate (1) having a front surface (1a) and a back surface (1b); forming an n-type drift region (2) on a surface of a GaN substrate, the n-type drift region having a lower donor concentration than the GaN substrate; forming a p-type base region (3) on the drift region; forming a p-type contact region (4) on or in a surface portion of the base region; forming an n-type JFET region (7) on the drift region from the surface of the contact region to a position deeper than the base region; Ion-implanting an n-type impurity into the contact region to form an n-type source region (5) on the base region, the source region having a higher donor concentration than the drift region; By ion-implanting B into the contact region, the acceptor concentration of the contact region is reduced, and a p-type channel region (6) is formed on the opposite side of the contact region on the base region with the source region in between; forming a gate insulating film (8) on the channel region; forming a gate electrode (9) on the gate insulating film above the channel region; forming a source electrode (10) connected to the contact region and the source region; forming a drain electrode (11) connected to the back surface of the GaN substrate; Forming the channel region includes forming a first channel region (6a) and a second channel region (6b) so that the first channel region (6a) and the second channel region (6b) have a lower acceptor concentration than the contact region and have different acceptor concentrations in the channel length direction; A first channel region is formed by implanting B ions into a region of the contact region where a channel region is to be formed, and further, a second channel region is formed by further implanting B ions into a region of the first channel region where a second channel region is to be formed.

[0012] In this way, the first and second channel regions are formed by ionizing the contact region with B. In other words, they are not formed by embedding growth in which a GaN layer is grown in a trench after the source region is formed. This makes it possible to prevent unintended n-type impurities from being introduced into the first and second channel regions. This makes it possible to manufacture a semiconductor device having an n-channel planar-type vertical semiconductor element that can set a desired threshold value and achieve low resistance, thereby achieving both a desired threshold value and low resistance.

[0013] The reference symbols in parentheses attached to each component indicate an example of the correspondence between the component and specific components described in the embodiments described below. [Brief description of the drawings]

[0014] [Figure 1] 1 is a cross-sectional view of a semiconductor device having an n-type channel planar vertical MOSFET according to a first embodiment. [Figure 2A] 2A to 2C are cross-sectional views illustrating a manufacturing process of the semiconductor device shown in FIG. [Figure 2B] 2B is a cross-sectional view showing a manufacturing process of the semiconductor device subsequent to FIG. 2A. [Figure 2C] 2C is a cross-sectional view showing a manufacturing process of the semiconductor device subsequent to FIG. 2B. [Figure 2D] 2D is a cross-sectional view showing a manufacturing process of the semiconductor device subsequent to FIG. 2C. [Figure 2E] 2E is a cross-sectional view showing a manufacturing process of the semiconductor device subsequent to FIG. 2D. [Figure 3A] 7A to 7C are cross-sectional views showing a manufacturing process of a semiconductor device having an n-type channel planar vertical MOSFET according to a second embodiment. [Figure 3B] 3B is a cross-sectional view showing a manufacturing process of the semiconductor device subsequent to FIG. 3A. [Figure 3C] 3C is a cross-sectional view showing a manufacturing process of the semiconductor device subsequent to FIG. 3B. [Figure 4] FIG. 13 is a cross-sectional view of a semiconductor device having an n-type channel planar type vertical MOSFET according to a modification of the first and second embodiments. [Diagram 5] FIG. 13 is a cross-sectional view of a semiconductor device having an n-type channel planar type vertical MOSFET according to a modification of the first and second embodiments. [Figure 6] FIG. 11 is a cross-sectional view of a semiconductor device having an n-type channel planar vertical MOSFET according to a third embodiment. [Figure 7A] 7A to 7C are cross-sectional views showing a manufacturing process of the semiconductor device shown in FIG. [Figure 7B]It is a cross-sectional view showing the manufacturing process of the semiconductor device following FIG. 7A. [Figure 7C] It is a cross-sectional view showing the manufacturing process of the semiconductor device following FIG. 7B. [Figure 7D] It is a cross-sectional view showing the manufacturing process of the semiconductor device following FIG. 7C. [Figure 7E] It is a cross-sectional view showing the manufacturing process of the semiconductor device following FIG. 7D. [Figure 8] It is a cross-sectional view of a semiconductor device having an n-type channel planar vertical MOSFET according to a modification of the third embodiment. [Figure 9] It is a cross-sectional view of a semiconductor device having an n-type channel planar vertical MOSFET according to a modification of the third embodiment.

Embodiments for Carrying Out the Invention

[0015] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the following embodiments, parts that are the same or equivalent to each other will be described with the same reference numerals.

[0016] (First Embodiment) The first embodiment of the present invention will be described. First, referring to FIG. 1, a semiconductor device having an n-type channel planar vertical MOSFET made of GaN according to this embodiment will be described. In FIG. 1, P1 to P4 indicate the acceptor concentration of p-type GaN, and in this embodiment, the relationship is P1 < P2 and P4 < P3 < P2.

[0017] The vertical power MOSFET is formed using an n- + type GaN substrate 1. The GaN substrate 1 has a structure with the upper surface as the main surface 1a and the lower surface, which is the opposite surface of the main surface 1a, as the back surface 1b, and is highly doped with Si (silicon) as a dopant.

[0018] On the main surface 1a of the GaN substrate 1, an n made of GaN doped with Si at a donor concentration lower than that of the GaN substrate 1 -A molded drift region 2 is deposited.

[0019] A base region 3 made of p-type GaN with a predetermined thickness is formed in a predetermined region on the drift region 2. The base region 3 is doped with Mg (magnesium) at a desired dopant concentration to have a predetermined acceptor concentration P1.

[0020] In addition, a contact region 4 made of p-type GaN with a high acceptor concentration is formed on the base region 3, and a source region 5 made of n-type GaN with a high donor concentration is formed. The contact region 4 is doped with Mg at a higher dopant concentration than the base region 3, thereby giving it a higher acceptor concentration P2 than the base region 3. The source region 5 is doped with Si at a higher dopant concentration than the drift region 2, thereby giving it a high donor concentration.

[0021] Furthermore, a channel region 6 made of p-type GaN is formed on the base region 3 on the opposite side of the contact region 4 with the source region 5 in between. The channel region 6 has a lower acceptor concentration than the contact region 4, and includes a first channel region 6a and a second channel region 6b with different acceptor concentrations. The channel region 6 is doped with Mg at a predetermined dopant concentration and with ions that inactivate the p-type dopant, thereby setting the acceptor concentration different from that of the contact region 4.

[0022] In this embodiment, the first channel region 6a is formed in contact with a JFET region 7 described later, and the second channel region 6b is disposed between the first channel region 6a and the source region 5 in contact with the source region 5. In other words, the first channel region 6a is formed away from the source region 5.

[0023] The first channel region 6a has an acceptor concentration P3 lower than that of the contact region 4 and higher than that of the second channel region 6b, and functions as a portion for setting the threshold value of the vertical MOSFET. The acceptor concentration P3 of the first channel region 6a is set to a concentration different from the acceptor concentration P2 of the contact region 4. Therefore, the contact region 4 has a low contact resistance, while the acceptor concentration P3 of the first channel region 6a is set to a concentration suitable for setting the threshold value.

[0024] On the other hand, the second channel region 6b has an acceptor concentration P4 lower than that of the first channel region 6a, and this second channel region 6b enables a lower resistance, as will be described later.

[0025] Furthermore, a JFET region 7 is formed adjacent to the base region 3 and the channel region 6. The JFET region 7 may be formed on the drift region 2 and separately from the drift region 2, or may be composed of a portion located above the drift region 2. In the case of this embodiment, the JFET region 7 is composed of an epitaxial growth layer grown on the drift region 2 or an ion-implanted layer formed by ion-implanting n-type impurities into GaN formed on the drift region 2.

[0026] A gate insulating film 8 made of a silicon oxide film or the like is formed on the channel region 6 and the JFET region 7. Furthermore, a gate electrode 9 is formed on the gate insulating film 8, and the gate electrode 9 is covered with an insulating film made of a silicon oxide film or the like (not shown).

[0027] A source electrode 10 is formed on the main surface 1a of the GaN substrate 1 so as to be connected to the source region 5 and the contact region 4. A drain electrode 11 is formed on the back surface 1b of the GaN substrate 1. With this structure, a planar type vertical MOSFET made of GaN is formed. The region in which this vertical MOSFET is formed is set as a cell region, and a breakdown voltage structure is formed in an outer peripheral region surrounding the cell region, thereby forming a semiconductor device.

[0028] Next, a method for manufacturing a semiconductor device having the vertical power MOSFET shown in Fig. 1 will be described with reference to Fig. 2A to Fig. 2E. Note that in Fig. 2B to Fig. 2E, the GaN substrate 1 and the portion below the drift region 2 are omitted.

[0029] [Step shown in FIG. 2A] First, a GaN substrate 1 is prepared. Then, a drift region 2 is crystal-grown on the main surface 1a of the GaN substrate 1, and a p-type base region 3 and a contact region 4 are crystal-grown on the drift region 2 in sequence. At this stage, the base region 3 and the contact region 4 are formed on the entire surface of the GaN substrate 1, and are also formed in the portion that will become the JFET region 7, and the contact region 4 is also formed in the portion that will become the source region 5 and the channel region 6. The drift region 2 is doped with Si to form n-type GaN, and the base region 3 and the contact region 4 are doped with Mg to form p-type GaN. The crystal growth conditions, such as the amount of doping material introduced, are adjusted for the base region 3 and the contact region 4 to change the acceptor concentration.

[0030] [Step shown in FIG. 2B] The JFET region 7 is formed so as to extend from the surface of the contact region 4 to a position deeper than the base region 3 and reach the drift region 2. The JFET region 7 is formed by burying growth in which a mask opening in the region where the JFET region 7 is to be formed is placed on the surface of the contact region 4, a trench penetrating the contact region 4 and the base region 3 is formed using the mask, and the trench is filled with n-type GaN. Alternatively, the JFET region 7 can also be formed by ion implanting Si from the surface of the contact region 4 to a position deeper than the base region 3 using the mask.

[0031] [Step shown in FIG. 2C] After a mask 20 is placed on the surfaces of the contact region 4 and the JFET region 7, the mask 20 is patterned to open a region where the source region 5 is to be formed. Then, Si is ion-implanted from above the mask 20. As a result, the source region 5 is formed.

[0032] [Step shown in FIG. 2D] After removing the mask 20, a mask 21 is placed on the surfaces of the contact region 4, the source region 5, and the JFET region 7, and the mask 21 is patterned to open the region where the channel region 6 is to be formed. Then, B (boron) is ion-implanted from above the mask 21. As a result, boron is introduced into the portion that will become the channel region 6, that is, into both the first channel region 6a and the second channel region 6b.

[0033] [Step shown in FIG. 2E] After removing the mask 21, a mask 22 is placed on the surfaces of the contact region 4, the source region 5, the channel region 6, and the JFET region 7, and the mask 22 is patterned to open a region where the second channel region 6b is to be formed. Then, B (boron) is additionally ion-implanted from above the mask 22. As a result, more boron is introduced into the portion that will become the second channel region 6b than into the portion that will become the first channel region 6a.

[0034] Thereafter, an activation annealing process is performed by a heat treatment at a high temperature to activate the B ions implanted in the channel region 6. As a result, the Mg ions in the channel region 6 are inactivated according to the amount of the B ions introduced. Specifically, the first channel region 6a has a higher acceptor concentration than the second channel region 6b because the amount of inactivated Mg ions is smaller in the first channel region 6a than in the second channel region 6b. In addition, the first channel region 6a and the second channel region 6b both have B ions implanted in them, so that the acceptor concentration is lower than that of the contact region 4.

[0035] Although the activation annealing treatment of the B ions has been described here, the activation annealing treatment is also performed on the Si ions implanted in the source region 5. The activation annealing treatment of the Si ions implanted in the source region 5 may be performed together with the activation annealing treatment of the B ions, or may be performed after the Si ions are implanted and before the B ions are implanted in the portion that will become the channel region 6.

[0036] Although the subsequent steps are not shown, steps such as forming a gate insulating film 8, a gate electrode 9, forming an insulating film covering the gate electrode 9, forming a source electrode 10, and forming a drain electrode 11 on the back surface of the GaN substrate 1 are performed. This completes the semiconductor device having the planar-type vertical MOSFET made of GaN shown in FIG.

[0037] Next, the operation of the n-type channel planar vertical MOSFET provided in the semiconductor device according to this embodiment will be described.

[0038] First, in a vertical MOSFET, in the off state before a gate voltage is applied to the gate electrode 9, no inversion layer is formed in the channel region 6. Therefore, even if a positive voltage is applied to the drain electrode 11, electrons do not flow from the source region 5 into the channel region 6, and no current flows between the source electrode 10 and the drain electrode 11.

[0039] When a predetermined gate voltage is applied to the gate electrode 9, a channel is formed on the surface of the channel region 6 that is in contact with the gate insulating film 8. Therefore, electrons injected from the source electrode 10 pass through the channel formed from the source region 5 to the channel region 6, then pass through the JFET region 7 and the drift region 2 and flow to the drain electrode 11. As a result, a current flows between the source electrode 10 and the drain electrode 11, and the semiconductor device is turned on.

[0040] At this time, the threshold value of the vertical MOSFET is determined by the acceptor concentration P3 of the first channel region 6a, which has a high acceptor concentration among the channel regions 6. Since the acceptor concentration P3 of the first channel region 6a is different from the acceptor concentration P2 of the contact region 4, it is possible to set the threshold value to a desired value. The second channel region 6b, which has a low acceptor concentration, is more likely to be inverted to n-type and contributes to low resistance by making electrons flow more easily. For this reason, the channel length is the length of the entire channel region 6 including the first channel region 6a and the second channel region 6b from the source region 5 to the JFET region 7, and the longer the dimension of the second channel region 6b in the channel length, the lower the resistance can be achieved. Therefore, it is preferable to set the dimension of the first channel region 6a to a value that allows the threshold value to be set accurately, while making the dimension of the second channel region 6b as long as possible. For example, it is preferable to make the dimension of the second channel region 6b larger than 1 / 2 of the channel length and larger than the dimension of the first channel region 6a.

[0041] According to the semiconductor device having an n-type channel planar vertical MOSFET made of GaN according to the present embodiment described above, the following effects can be obtained.

[0042] (1) The first channel region 6a and the second channel region 6b are adjusted to different acceptor concentrations. Therefore, the threshold of the first channel region 6a can be controlled independently of the second channel region 6b, and a suitable threshold can be set for a vertical MOSFET. The second channel region 6b can be set to a lower acceptor concentration than the first channel region 6a. Therefore, the channel resistance of the second channel region 6b can be reduced when the vertical MOSFET is turned on, and high mobility can be obtained. This makes it possible to provide a semiconductor device that can achieve both a desired threshold and low resistance.

[0043] (2) The contact region 4, the first channel region 6a, and the second channel region 6b are adjusted to have different acceptor concentrations. Therefore, the contact region 4 is set to an acceptor concentration that reduces the contact resistance with the source electrode 10, while the first channel region 6a can be set to an acceptor concentration suitable for the threshold value, and the second channel region 6b can be set to an acceptor concentration that can reduce resistance. Therefore, by increasing the acceptor concentration of the contact region 4, it is possible to prevent the threshold value from becoming higher than necessary, and to suppress a decrease in mobility.

[0044] (3) In this embodiment, the first channel region 6a is separated from the source region 5. This makes it possible to set the threshold voltage more accurately, suppress the normally-on state, and further reduce the resistance.

[0045] It is preferable that the acceptor concentration of the first channel region 6a is set with high accuracy in order to set the threshold value, but if it is adjacent to the source region 5, n-type impurities may be introduced by lateral diffusion. In this case, the threshold value may fluctuate and become normally on, so it is necessary to ensure a certain degree of dimension of the first channel region 6a in the longitudinal direction of the channel length, that is, in the left-right direction of the paper in FIG. 1, to suppress the normally on state. On the other hand, if the dimension of the first channel region 6a in the longitudinal direction of the channel length becomes large, it will hinder the reduction in resistance. For this reason, it is preferable to make the dimension in that direction as small as possible while accurately setting the threshold value.

[0046] Therefore, when the first channel region 6a is in contact with the source region 5, it is necessary to set the size of the first channel region 6a to be larger in consideration of the lateral diffusion of n-type impurities, which may hinder the reduction in resistance. For this reason, in this embodiment, the first channel region 6a is separated from the source region 5. This makes it possible to suppress the lateral diffusion of n-type impurities into the first channel region 6a, allowing the threshold value to be set more accurately and enabling the reduction in resistance.

[0047] Although the second channel region 6b is adjacent to the source region 5, the acceptor concentration in the second channel region 6b is set low to reduce resistance, so that the resistance does not increase even if the n-type impurities diffuse laterally.

[0048] (4) Furthermore, according to the manufacturing method of this embodiment, the contact region 4 is formed over the entire surface of the drift region 2, while the first channel region 6a and the second channel region 6b are formed by ion implantation of B. In other words, the channel region 6 is not formed by forming a recess after the formation of the source region 5 and then growing a GaN layer in the recess. This makes it possible to prevent unintended introduction of n-type impurities into the first channel region 6a and the second channel region 6b. This makes it possible to more accurately set a desired threshold value and achieve low resistance.

[0049] (5) The second channel region 6b is formed by ion implantation of B onto the first channel region 6a. This allows the channel length, i.e., the overall length of the channel region 6 including the first channel region 6a and the second channel region 6b from the source region 5 to the JFET region 7, to be set with high precision, and the channel length does not need to be made unnecessarily long. Of course, the overall length of the channel region 6 may vary due to mask misalignment, but since the variation is on the order of the mask misalignment, the overall length of the channel region 6 can be set with sufficient precision.

[0050] In addition to using B ions, there is also a method of inactivating p-type semiconductors by ion implantation of hydrogen or fluorine. However, the method of ion implantation of hydrogen or fluorine is thermally unstable, and there is a possibility that the implanted hydrogen ions or fluorine ions may be removed due to the thermal history during the manufacturing process of the semiconductor device, so here, B ions are used as a more preferable method. In addition, the temperature of the activation annealing process for activating B ions is 800 to 1000°C, which is lower than the temperature of 1300 to 1400°C for activating Mg ions, and does not need to be so high that GaN decomposes. Therefore, it is possible to suppress damage to the GaN surface, such as roughening of the GaN surface, caused by the activation annealing process.

[0051] Second embodiment A second embodiment will be described. This embodiment is a semiconductor device manufacturing method modified from that of the first embodiment, and other aspects are the same as those of the first embodiment, so only the differences from the first embodiment will be described.

[0052] This embodiment differs from the first embodiment in that the contact region 4 is formed by ion implantation. The manufacturing method of this embodiment will be described with reference to Figures 3A to 3C. Note that the GaN substrate 1 and the lower portion of the drift region 2 are omitted in Figures 3B to 3C.

[0053] 3A, a GaN substrate 1 is prepared, and then a drift region 2 and a base region 3 are grown by crystal growth in this order on a main surface 1a of the GaN substrate 1. At this time, the thickness of the base region 3 is set to be the thickness of the portion to be ultimately left as the base region 3 plus the thickness of the contact region 4.

[0054] Next, in a step shown in FIG. 3B, an n-type JFET region 7 is formed so as to extend from the surface of the base region 3 to a position deeper than the base region 3 and reach the drift region 2 using a technique similar to that used in the step shown in FIG. 2B.

[0055] Then, in the step shown in FIG. 3C, a mask (not shown) having openings in areas where the contact region 4 and the channel region 6 are to be formed is disposed, and Mg ions are implanted using this mask, thereby forming the contact region 4.

[0056] At this stage, the source region 5 doped with a high concentration of n-type impurities has not yet been formed. Therefore, it is not necessary to consider the lateral diffusion of n-type impurities. Therefore, after arranging a mask (not shown) that opens the regions where the contact region 4 and the channel region 6 are to be formed, the base region 3 may be partially etched using the mask to form trenches, and the contact region 4 may be formed by embedding growth.

[0057] 2B, a semiconductor device having a vertical power MOSFET can be manufactured by performing the same steps as those shown in FIG. 2C and thereafter. In this manner, the contact region 4 may be formed after the formation of the JFET region 7, and in this case, the same effect as in the first embodiment can be obtained.

[0058] (Modifications of the first and second embodiments) In the first and second embodiments, the channel region 6 has a structure in which the first channel region 6a is separated from the source region 5 and the second channel region 6b is in contact with the source region 5. However, as shown in FIG. 4, the configuration of the channel region 6 in the first embodiment may be changed so that the first channel region 6a is in contact with the source region 5.

[0059] Even in the case of such a structure, even if n-type impurities penetrate into the first channel region 6a due to lateral diffusion of the n-type impurities from the source region 5, it is sufficient that the threshold voltage can be set by the remaining portion and the vertical MOSFET does not become normally on.

[0060] 5, the second channel region 6b may be disposed at the center of the channel region 6, and the first channel region 6a may be formed separately on both sides of the second channel region 6b, that is, on the source region 5 side or the JFET region 7 side of the second channel region 6b. Even in this case, n-type impurities may be laterally diffused on the first channel region 6a side on the source region 5 side, as long as the threshold value can be set in the first channel region 6a located on the JFET region 7 side.

[0061] Considering the mask misalignment, it may not be possible to accurately form the second channel region 6b at the boundary with the source region 5. However, it is sufficient if the threshold voltage can be accurately set by the first channel region 6a on the JFET region 7 side.

[0062] Third embodiment The third embodiment will be described. In the third embodiment, the acceptor concentration is set to P2, which allows the base region 3 to become the contact region 4, in comparison with the first embodiment. Since the other points are the same as those in the first embodiment, only the points different from the first embodiment will be described.

[0063] 6, in this embodiment, the base region 3 is increased to an acceptor concentration P2 that allows it to function as the contact region 4 described in the first embodiment. In this way, by increasing the base region 3 to match the acceptor concentration P2 of the contact region 4, the surface layer of the base region 3 can be made to function as the contact region 4, and these can be formed at the same time.

[0064] Specifically, in the step shown in FIG. 7A, after a GaN substrate 1 is prepared, a drift region 2 and a p-type base region 3 which also functions as a contact region 4 are sequentially grown by crystal growth on a main surface 1a of the GaN substrate 1.

[0065] 2B, an n-type JFET region 7 is formed from the surface of the base region 3 to a position deeper than the base region 3 and to reach the drift region 2. Note that the GaN substrate 1 and the portion below the drift region 2 are omitted in FIGS. 7B to 7E.

[0066] 7C to 7E, the same steps as those in FIG. 2C and thereafter are performed to manufacture a semiconductor device having a vertical power MOSFET. In this way, the manufacturing process of the semiconductor device can be further simplified by making the base region 3 have an acceptor concentration P2 that allows the base region 3 to function as the contact region 4. Of course, since the base region 3 is formed thick with a high acceptor concentration P2, the process of forming the base region 3 becomes more difficult, but this can be dealt with by lengthening the crystal growth time, for example.

[0067] (Modification of the third embodiment) In the third embodiment, similarly to the first and second embodiments, the channel region 6 has a structure in which the first channel region 6a is separated from the source region 5 and the second channel region 6b is in contact with the source region 5. However, as shown in FIG. 8, the configuration of the channel region 6 in the third embodiment may also be changed to a structure in which the first channel region 6a is in contact with the source region 5.

[0068] Even in such a structure, the threshold value can be set in the first channel region 6a by the lateral diffusion of the n-type impurities from the source region 5, as long as the vertical MOSFET does not become normally-on.

[0069] Also, as shown in FIG. 9, the second channel region 6b may be disposed at the central position within the channel region 6, and the first channel regions 6a may be formed separately on both sides of the second channel region 6b, that is, on the source region 5 side and the JFET region 7 side with respect to the second channel region 6b. Even in this case, as long as the threshold value can be set in the first channel region 6a located on the JFET region 7 side, the n-type impurities may laterally diffuse to the first channel region 6a side on the source region 5 side.

[0070] (Other Embodiments) Although the present disclosure has been described in accordance with the above-described embodiments, it is not limited to the embodiments, and includes various modifications and modifications within the equivalent scope. In addition, various combinations and forms, and further other combinations and forms including only one element, more than one element, or less than one element thereof, fall within the scope and spirit of the present disclosure.

[0071] For example, in the embodiment, the acceptor concentrations are set such that P1 < P2 and P4 < P3 < P2, but this relationship does not necessarily have to be as such. For example, P1 < P3 may be satisfied. That is, P3 may be determined according to the desired threshold value, and the magnitude relationship between P1 and P3 may be determined according to the required specifications. Also, P1 and P4 may be the same, or either one may be higher than the other.

[0072] Also, in each of the above embodiments, in the direction of the channel length, the dimension of the second channel region 6b is made longer than the dimension of the first channel region 6a, but the dimension of the first channel region 6a may be longer than the dimension of the second channel region 6b.

[0073] In addition, in the above embodiments, an n-channel planar vertical MOSFET is used as an example of a semiconductor element, but an n-channel planar IGBT may be used instead. The IGBT differs from the above embodiments in that the conductivity type of the GaN substrate 1 is changed from n-type to p-type, and other structures and manufacturing methods are the same as those of the above embodiments. [Explanation of symbols]

[0074] 1... GaN substrate, 2... drift region, 3... base region, 4... contact region 5...source region, 6...channel region, 6a, 6b...first and second channel regions 7...JFET region, 8...gate insulating film, 9...gate electrode, 10...source electrode 11...Drain electrode

Claims

1. A semiconductor device having an n-type channel planar type vertical semiconductor element made of GaN, An n-type or p-type GaN substrate (1) having a front surface (1a) and a back surface (1b); An n-type drift region (2) formed on the surface of the GaN substrate and having a lower donor concentration than the GaN substrate; a p-type base region (3) formed on the drift region; A p-type contact region (4) provided on the base region or on a surface layer of the base region; an n-type source region (5) formed on the base region and having a higher donor concentration than the drift region; a p-type channel region (6) formed on the base region and disposed on the opposite side of the contact region with the source region therebetween; a JFET region (7) disposed on the drift region and in contact with the base region and the channel region; a gate insulating film (8) formed on the channel region; a gate electrode (9) formed on the gate insulating film above the channel region; a source electrode (10) connected to the contact region and the source region; A drain electrode (11) connected to the rear surface of the GaN substrate; the channel region has a lower acceptor concentration than the contact region, and includes a first channel region (6a) and a second channel region (6b) having different acceptor concentrations in a channel length direction.

2. 2. The semiconductor device according to claim 1, wherein the channel region is made p-type GaN by introducing Mg and B, the amount of B introduced is greater in the second channel region than in the first channel region, and an acceptor concentration (P3) in the first channel region is higher than an acceptor concentration (P4) in the second channel region.

3. 3. The semiconductor device according to claim 2, wherein a dimension of said channel region in a direction from said source region to said JFET region is defined as a channel length, and a dimension of said second channel region in said channel length direction is larger than a dimension of said first channel region.

4. 3. The semiconductor device according to claim 2, wherein a dimension of said channel region in a direction from said source region to said JFET region is defined as a channel length, and a dimension of said first channel region in said channel length direction is larger than a dimension of said second channel region.

5. 5. The semiconductor device according to claim 2, wherein said second channel region is in contact with said source region, and said first channel region is in contact with said JFET region.

6. 5. The semiconductor device according to claim 2, wherein said first channel region is in contact with said source region, and said second channel region is in contact with said JFET region.

7. A method for manufacturing a semiconductor device having an n-type channel planar type vertical semiconductor element made of GaN, comprising the steps of: Providing an n-type or p-type GaN substrate (1) having a front surface (1a) and a back surface (1b); forming an n-type drift region (2) on a surface of the GaN substrate, the n-type drift region having a lower donor concentration than the GaN substrate; forming a p-type base region (3) on the drift region; forming a p-type contact region (4) on the base region or in a surface portion of the base region; forming an n-type JFET region (7) on the drift region from a surface of the contact region to a position deeper than the base region; Ion-implanting an n-type impurity into the contact region to form an n-type source region (5) on the base region, the n-type source region having a higher donor concentration than the drift region; Implanting B into the contact region to reduce an acceptor concentration in the contact region, and forming a p-type channel region (6) on the opposite side of the contact region across the source region on the base region; forming a gate insulating film (8) on the channel region; forming a gate electrode (9) on the gate insulating film above the channel region; forming a source electrode (10) connected to the contact region and the source region; forming a drain electrode (11) connected to the rear surface of the GaN substrate; forming the channel region forming a first channel region (6 a) and a second channel region (6 b) so that the first channel region (6 a) and the second channel region (6 b) have a lower acceptor concentration than the contact region and have different acceptor concentrations in a channel length direction; 2. A method for manufacturing a semiconductor device, comprising: forming the first channel region by implanting B ions into a region of the contact region where the channel region is to be formed; and further forming the second channel region by further implanting B ions into a region of the first channel region where the second channel region is to be formed.

8. 8. The method for manufacturing a semiconductor device according to claim 7, wherein, in forming the contact region, an acceptor concentration of the base region is adjusted to match an acceptor concentration of the contact region, so that the contact region is constituted by a surface layer portion of the base region.

9. 9. The method for manufacturing a semiconductor device according to claim 7, wherein the forming of the channel regions comprises ion implantation of B to form the first channel region and the second channel region such that the second channel region is in contact with the source region and the first channel region is in contact with the JFET region.

Citation Information

Patent Citations

  • Semiconductor device

    JP2012059744A

  • Semiconductor device

    JP2012234848A

  • Compound semiconductor device and manufacturing method for the same

    JP2017152490A

  • Method for manufacturing semiconductor device, and semiconductor device

    JP2018010892A

  • Manufacturing method for nitride semiconductor device

    JP2021129018A