Semiconductor device and manufacturing method

Abstract

A semiconductor device is provided, wherein in a depth direction, a hydrogen concentration distribution has a first hydrogen concentration peak and a second hydrogen concentration peak, a donor concentration distribution has a first donor concentration peak and a second donor concentration peak, the first hydrogen concentration peak and the first donor concentration peak are arranged at a first depth, the second hydrogen concentration peak and the second donor concentration peak are arranged at a second depth, the second depth is deeper than the first depth when the lower surface is taken as a reference, each concentration peak has an upward slope whose concentration value increases from the lower surface toward the upper surface, and a value obtained by normalizing the slope of the upward slope of the second donor concentration peak with the slope of the upward slope of the second hydrogen concentration peak is smaller than a value obtained by normalizing the slope of the upward slope of the first donor concentration peak with the slope of the upward slope of the first hydrogen concentration peak.

Description

Technical Field

[0001] The present invention relates to a semiconductor device and a manufacturing method thereof. Background Art

[0002] Conventionally, it is known that hydrogen is implanted into a semiconductor substrate and diffused to thereby combine with hydrogen in crystal defects existing in a diffusion region to form donors (for example, see Patent Document 1).

[0003] Patent Document 1: Japanese Patent Application No. 2016-204227 Summary of the invention

[0004] Technical issues

[0005] It is preferable that the range and donor concentration of the donor region generated by bonding of crystal defects with hydrogen be controlled with high precision.

[0006] Technical Solution

[0007] In order to solve the above-mentioned problems, in one embodiment of the present invention, a semiconductor device is provided, which has a semiconductor substrate having an upper surface and a lower surface. In the depth direction connecting the upper surface and the lower surface of the semiconductor substrate, the hydrogen concentration distribution may have a first hydrogen concentration peak and a second hydrogen concentration peak. In the depth direction, the donor concentration distribution may have a first donor concentration peak and a second donor concentration peak. The first hydrogen concentration peak and the first donor concentration peak may be arranged at a first depth. The second hydrogen concentration peak and the second donor concentration peak may be arranged at a second depth, and the second depth is deeper than the first depth when the lower surface is taken as a reference. Each concentration peak may have an upward slope whose concentration value increases as it moves from the lower surface toward the upper surface. The value obtained by normalizing the slope of the upward slope of the second donor concentration peak with the slope of the upward slope of the second hydrogen concentration peak may be smaller than the value obtained by normalizing the slope of the upward slope of the first donor concentration peak with the slope of the upward slope of the first hydrogen concentration peak.

[0008] Each concentration peak may have a downward slope whose concentration value decreases from the lower surface toward the upper surface. The upward slope of the second hydrogen concentration peak may have a smaller slope than the downward slope. The upward slope of the second donor concentration peak may have a smaller slope than the downward slope.

[0009] The donor concentration distribution between the first depth and the second depth may have a flat region where the donor concentration is substantially constant. The length of the flat region in the depth direction may be greater than 10% of the thickness of the semiconductor substrate in the depth direction.

[0010] The donor concentration distribution between the first depth and the second depth may have a flat region where the donor concentration is substantially constant. The length of the flat region in the depth direction may be 10 μm or more.

[0011] The minimum value of the donor concentration of the flat region may be higher than the donor concentration of the semiconductor substrate.

[0012] A minimum value of the donor concentration between the first depth and the second depth may be higher than a donor concentration of the semiconductor substrate.

[0013] The concentration value of the second hydrogen concentration peak may be smaller than the concentration value of the first hydrogen concentration peak.

[0014] The semiconductor device may include an N-type drift region disposed in a semiconductor substrate. The semiconductor device may include an emitter region disposed in the semiconductor substrate in contact with an upper surface and having a donor concentration higher than that of the drift region. The semiconductor device may include a P-type base region disposed between the emitter region and the drift region. The semiconductor device may include a P-type collector region disposed in the semiconductor substrate in contact with a lower surface. The semiconductor device may include an N-type buffer region disposed between the collector region and the drift region and having one or more donor concentration peaks having a higher donor concentration than that of the drift region. The first donor concentration peak may be a donor concentration peak of the buffer region.

[0015] The semiconductor device may include an accumulation region provided between the base region and the drift region and having one or more donor concentration peaks having a higher donor concentration than the donor concentration of the drift region. The second donor concentration peak may be a donor concentration peak of the accumulation region.

[0016] The accumulation region may have a donor concentration peak formed of a donor other than hydrogen in addition to the second donor concentration peak.

[0017] The second donor concentration peak may be disposed between the buffer region and the accumulation region.

[0018] The semiconductor device may include a gate trench portion provided on an upper surface of a semiconductor substrate. The second donor concentration peak may be arranged between a bottom of the gate trench portion and an upper surface of the semiconductor substrate.

[0019] The semiconductor device may include an active portion provided on a semiconductor substrate. The semiconductor device may include an edge termination structure portion provided in a manner of surrounding the active portion when the semiconductor substrate is viewed from above. The semiconductor substrate may include a pass-through region through which hydrogen injected into the position of the second hydrogen concentration peak passes. In the depth direction, the pass-through region provided on the edge termination structure portion may be shorter than the pass-through region provided on the active portion, or the pass-through region may not be provided on the edge termination structure portion.

[0020] The semiconductor device may include a transistor portion and a diode portion provided in a semiconductor substrate. In the depth direction, the through region provided in the diode portion may be shorter than the through region provided in the transistor portion, or the through region may not be provided in the diode portion.

[0021] In the depth direction, the through region provided in the transistor portion may be shorter than the through region provided in the diode portion, or no through region may be provided in the transistor portion.

[0022] The first depth may be within a range of 5 μm or less in the depth direction from the lower surface.

[0023] The donor concentration at the first hydrogen concentration peak can be 1×10 15 / cm 3 Above and 1×10 17 / cm 3 the following.

[0024] In a second aspect of the present invention, a semiconductor device is provided, which includes a semiconductor substrate having an upper surface and a lower surface. A hydrogen concentration distribution in a depth direction connecting the upper surface and the lower surface of the semiconductor substrate may include a first hydrogen concentration peak arranged in a range of 5 μm or less along the depth direction from the lower surface and a second hydrogen concentration peak arranged at a position closer to the upper surface than the first hydrogen concentration peak.

[0025] The semiconductor substrate may have an impurity concentration peak between the lower surface and the first hydrogen concentration peak, and the impurity is argon or fluorine.

[0026] In a third embodiment of the present invention, a method for manufacturing a semiconductor device of the first embodiment is provided. The manufacturing method may include a first implantation step of implanting hydrogen from the lower surface of a semiconductor substrate to a first depth. The manufacturing method may include a second implantation step of implanting hydrogen from the lower surface of the semiconductor substrate to a second depth to form a through zone through which hydrogen passes. The manufacturing method may include a diffusion step of heat-treating the semiconductor substrate so that the hydrogen implanted to the first depth diffuses in the through zone. The dose of hydrogen in the first implantation step may be determined so that in the semiconductor substrate heat-treated in the diffusion step, the minimum value of the donor concentration in the through zone becomes higher than the donor concentration of the semiconductor substrate before the hydrogen implantation.

[0027] In the first implantation step, hydrogen may be implanted at a dose greater than a minimum dose determined according to a diffusion coefficient of hydrogen in the semiconductor substrate and the second depth.

[0028] The semiconductor substrate is a silicon substrate. When the second depth from the lower surface is set to x (cm), the dose Q ( / cm 2 ) can satisfy the following formula.

[0029] Q≥2.6×10 10 ×e 0.12x

[0030] In the first implantation step, hydrogen may be implanted to a first depth by plasma doping.

[0031] After plasma doping, the lower surface of the semiconductor substrate may be ground.

[0032] After the plasma doping, laser annealing may be performed on the lower surface of the semiconductor substrate.

[0033] It should be noted that the above-mentioned summary of the invention does not list all the essential features of the present invention. In addition, sub-combinations of these feature groups can also constitute other inventions. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 is a cross-sectional view showing an example of the semiconductor device 100 .

[0035] Figure 2 express Figure 1 The hydrogen concentration distribution, donor concentration distribution, and vacancy defect concentration distribution 175 in the depth direction at the position indicated by the AA line.

[0036] Figure 3A 1 is a diagram for explaining the relationship between the first hydrogen concentration peak 131 and the first donor concentration peak 111 .

[0037] Figure 3B 1 is a diagram for explaining the relationship between the second hydrogen concentration peak 141 and the second donor concentration peak 121 .

[0038] Figure 3C 14 is a diagram illustrating the slope of the upward slope 142 .

[0039] Figure 4A 2 is a diagram illustrating another definition of the normalization of the slope of the upward ramp 112 .

[0040] Figure 4B 2 is a diagram illustrating another definition of the normalization of the slope of the upward ramp 122 .

[0041] Figure 5 1 is a diagram illustrating the flat region 150 .

[0042] Figure 6 is a diagram showing a structural example of the semiconductor device 100 .

[0043] Figure 7 Yes means Figure 6 A diagram showing an example of the carrier concentration distribution in the depth direction at the position of the BB line.

[0044] Figure 8 FIG. 1 is a diagram showing another configuration example of the semiconductor device 100 .

[0045] Fig. 9 Yes means Figure 8 A diagram showing an example of the carrier concentration distribution in the depth direction at the position of the CC line.

[0046] Fig.10 express Figure 1 The hydrogen concentration distribution and carrier concentration distribution in the depth direction at the position indicated by the AA line.

[0047] Fig.11 It is a diagram showing an example of arrangement of each element on the upper surface 21 of the semiconductor substrate 10 .

[0048] Fig.12 Yes means Fig.11 FIG. 1 is a diagram of an example of a c-c' section.

[0049] Fig.13 FIG. 1 is a diagram showing another configuration example of the pass-through area 106 .

[0050] Fig.14 FIG. 1 is a diagram showing another configuration example of the pass-through area 106 .

[0051] Fig.15 FIG. 1 is a diagram showing another configuration example of the pass-through area 106 .

[0052] Fig.16A FIG. 1 is a diagram showing another configuration example of the pass-through area 106 .

[0053] Fig. 16B FIG. 1 is a diagram showing another configuration example of the pass-through area 106 .

[0054] Fig.17A FIG. 1 is a diagram showing another configuration example of the pass-through area 106 .

[0055] Fig. 17B FIG. 1 is a diagram showing another configuration example of the pass-through area 106 .

[0056] Fig. 17C 1 is a diagram for explaining the minimum film thickness M of the photoresist film 200 for preventing hydrogen ions from invading the semiconductor substrate 10 .

[0057] Fig.18A FIG. 1 is a diagram showing another configuration example of the pass-through area 106 .

[0058] Fig.18B FIG. 1 is a diagram showing another configuration example of the pass-through area 106 .

[0059] Fig.19 The diagram shows a step of forming the through region 106 in the method of manufacturing the semiconductor device 100 .

[0060] Fig. 20 19 is a diagram showing an example of carrier concentration distribution in the semiconductor substrate 10 after the diffusion step S1904.

[0061] Fig.21: is a graph showing the relationship between the diffusion coefficient D of hydrogen and the first dose Q.

[0062] Fig. 22 It is a graph showing the relationship between the diffusion coefficient D and the annealing temperature T.

[0063] Fig.23 is a graph showing the relationship between the diffusion depth of hydrogen and the first dose.

[0064] Fig.24 It is a graph showing the relationship between the diffusion coefficient D and the diffusion depth x.

[0065] Fig.25 It is a graph of a straight line specifying the minimum dose for each annealing temperature.

[0066] Fig.26 : is a graph showing the relationship between the second dose and the minimum dose of the first dose.

[0067] Fig. 27 This is a diagram illustrating an example of the first depth Z1.

[0068] Fig.28 Another example of the donor concentration distribution and the hydrogen chemical concentration distribution in the depth direction of the semiconductor substrate 10 is shown.

[0069] Fig.29 1 is a diagram showing an example of the hydrogen chemical concentration distribution and the argon chemical concentration distribution in the vicinity of the first hydrogen concentration peak 131 .

[0070] Fig.30 FIG. 1 is a diagram showing another configuration example of the semiconductor device 100 .

[0071] Fig.31 express Fig.30 An example of carrier concentration distribution, hydrogen chemical concentration distribution and boron chemical concentration distribution at the DD line.

[0072] Fig.32 express Fig.30 An example of carrier concentration distribution, hydrogen chemical concentration distribution and phosphorus chemical concentration distribution at the EE line.

[0073] Fig.33 The diagram shows a part of the steps of the method for manufacturing the semiconductor device 100 .

[0074] Fig.34 The diagram shows a part of the steps of the method for manufacturing the semiconductor device 100 .

[0075] Fig.35 An example of a process of implanting hydrogen ions into the first depth Z1 and the second depth Z2 in the lower surface side structure forming step is shown.

[0076] Fig.36 Another example of the step of implanting hydrogen ions into the first depth Z1 and the second depth Z2 in the lower surface side structure forming step is shown.

[0077] Explanation of symbols

[0078] 10: semiconductor substrate, 11: well region, 12: emitter region, 14: base region, 15: contact region, 16: reservoir region, 18: drift region, 20: buffer region, 21: upper surface, 22: collector region, 23: lower surface, 24, 25: peak, 30: dummy groove portion, 32: dummy insulating film, 34: dummy conductive portion, 38: interlayer insulating film, 40: gate groove portion, 42: gate insulating film, 44: gate conductive portion, 48: gate flow channel, 50: gate metal layer, 52: emitter, 54: collector, 70: transistor portion, 72: boundary portion, 80: diode portion, 82: cathode region, 90: edge termination structure portion, 92: guard ring, 94: field plate, 100: semiconductor device, 106: through region, 111: first donor concentration peak, 112, 122, 132, 142, 17 2: Upward slope, 113, 123, 133, 143, 173: Downward slope, 114, 124, 125, 134, 144, 145: Slope, 116: Gate pad, 118: Emitter pad, 120: Active part, 121: Second donor concentration peak, 131: First hydrogen concentration peak, 140: Peripheral end, 141: Second hydrogen concentration peak, 150: Flat area, 151: valley, 161: first case, 162: second case, 163: third case, 171: vacancy concentration peak, 174: channel stop region, 175: vacancy defect concentration distribution, 180: basic doping region, 181: non-doped region, 182: protective film, 184: coating, 190: intermediate boundary region, 192: lifetime control region, 194: hydrogen concentration peak, 196: argon concentration peak, 200: photoresist film DETAILED DESCRIPTION

[0079] Hereinafter, the present invention will be described by way of the embodiments of the invention, but the following embodiments do not limit the inventions involved in the claims. In addition, all combinations of features described in the embodiments are not necessarily required for the solution of the invention.

[0080] In this specification, one side in a direction parallel to the depth direction of the semiconductor substrate is referred to as "upper", and the other side is referred to as "lower". One of the two main surfaces of a substrate, layer, or other component is referred to as an upper surface, and the other is referred to as a lower surface. The directions of "upper" and "lower" are not limited to the direction of gravity or the direction when the semiconductor device is installed.

[0081] In this specification, orthogonal coordinate axes of the X-axis, Y-axis, and Z-axis are sometimes used to explain technical matters. Orthogonal coordinate axes are only used to determine the relative positions of components and do not limit specific directions. For example, the Z-axis is not limited to indicating the height direction relative to the ground. It should be noted that the +Z-axis direction and the -Z-axis direction are opposite directions to each other. When the positive and negative are not recorded but the Z-axis direction is recorded, it refers to the direction parallel to the +Z-axis and the -Z-axis.

[0082] In this specification, the term "same" or "equal" may include a case where there is an error due to manufacturing variation, etc. The error is within 10%, for example.

[0083] In this specification, chemical concentration refers to the concentration of impurities measured regardless of the state of activation. Chemical concentration can be measured, for example, by secondary ion mass spectrometry (SIMS). In this specification, the concentration difference between the donor and the acceptor is sometimes set to the concentration of the larger one of the donor and the acceptor. The concentration difference can be measured by voltage-capacitance determination (CV method). In addition, the carrier concentration measured by the spreading resistance determination (SR) can be used as the concentration of the donor or the acceptor. In addition, in the case where the concentration distribution of the donor or the acceptor has a peak, the peak value can be used as the concentration of the donor or the acceptor in the region. In the case where the concentration of the donor or the acceptor in the region where the donor or the acceptor exists is substantially uniform, the average value of the donor or the acceptor concentration in the region can be used as the donor or the acceptor concentration.

[0084] In this specification, doping concentration refers to the concentration of donor- or acceptor-type impurities. In this specification, the difference between the concentrations of donors and acceptors (i.e., net doping concentration) is sometimes used as the doping concentration. In addition, the peak value of the doping concentration distribution in a doping region is sometimes used as the doping concentration in the doping region.

[0085] In this specification, the concentration per unit volume is expressed in at ms / cm 3 or ions / cm 3 or / cm 3 This unit is used for the concentration of donors or acceptors or chemical concentrations within a semiconductor substrate. The concentration per unit area is expressed in atоms / cm 2 or ions / cm 2 or / cm 2 This unit is used to indicate the amount of atoms or ions implanted into a substrate (dose). The notation of atoms and ions can be omitted.

[0086] The unit system in this specification is the SI unit system unless otherwise specified. Sometimes, cm is used to represent the unit of length, but each calculation can be performed after conversion to meter (m).

[0087] Figure 1 1 is a cross-sectional view showing an example of a semiconductor device 100. The semiconductor device 100 includes a semiconductor substrate 10. The semiconductor substrate 10 is a substrate formed of a semiconductor material. As an example, the semiconductor substrate 10 is a silicon substrate. The semiconductor substrate 10 has a donor concentration determined by impurities injected during manufacturing. The conductivity type of the semiconductor substrate 10 in this example is N-type. In this specification, the donor concentration in the semiconductor substrate 10 is sometimes referred to as substrate concentration.

[0088] The semiconductor substrate 10 has an upper surface 21 and a lower surface 23. The upper surface 21 and the lower surface 23 are two main surfaces of the semiconductor substrate 10. In this specification, the orthogonal axes in the plane parallel to the upper surface 21 and the lower surface 23 are referred to as the X-axis and the Y-axis, and the axis perpendicular to the upper surface 21 and the lower surface 23 is referred to as the Z-axis. Semiconductor elements such as IGBTs and FWDs are formed on the semiconductor substrate 10, but Figure 1 In the structure, these components are omitted.

[0089] Hydrogen ions are injected into the semiconductor substrate 10 from the lower surface 23. The hydrogen ions in this example are protons. The hydrogen ions are injected into at least two depths Z1 and Z2 in the depth direction (Z-axis direction) of the semiconductor substrate 10. In this example, the one of the two depths closer to the lower surface 23 is the first depth Z1, and the one deeper than the first depth Z1 as seen from the lower surface 23 is the second depth Z2. Figure 1 In FIG. 1 , the implanted hydrogen is schematically indicated by an × mark, but the hydrogen is also distributed around the implantation positions Z1 and Z2.

[0090] The first depth Z1 may be arranged on the lower surface 23 side in the depth direction of the semiconductor substrate 10. For example, when the lower surface 23 is used as a reference, the first depth Z1 may be arranged in a range of less than half the thickness of the semiconductor substrate 10, or may be arranged in a range of less than 1 / 4 the thickness of the semiconductor substrate 10. The second depth Z2 may be arranged on the upper surface 21 side in the depth direction of the semiconductor substrate 10. For example, when the upper surface 21 is used as a reference, the second depth Z2 may be arranged in a range of less than half the thickness of the semiconductor substrate 10, or may be arranged in a range of less than 1 / 4 the thickness of the semiconductor substrate 10. However, the first depth Z1 and the second depth Z2 are not limited to these ranges.

[0091] The hydrogen ions implanted into the second depth Z2 pass through the through region 106 from the lower surface 23 to the second depth Z2. In the through region 106, vacancy defects such as vacancies (V) and composite vacancies (VV) are generated due to the passage of hydrogen ions. In this specification, unless otherwise specified, vacancies also include composite vacancies. The vacancy density in the through region 106 can be adjusted by the dose of hydrogen ions implanted into the second depth Z2, etc.

[0092] By heat treating the semiconductor substrate 10 after the hydrogen ion implantation, the hydrogen implanted to the first depth Z1 and the second depth Z2 is diffused into the through zone 106. VOH defects are formed by combining vacancies (V) and oxygen (O) existing in the through zone 106 with hydrogen (H). The VOH defects act as donors that supply electrons. As a result, the donor concentration of the through zone 106 can be made higher than the donor concentration Db (or resistivity, basic doping concentration) when the semiconductor ingot that serves as the basis of the semiconductor substrate 10 is manufactured. Therefore, the donor concentration of the semiconductor substrate 10 can be easily adjusted according to the characteristics that the element formed on the semiconductor substrate 10 should have. It should be noted that, unless otherwise specified, in this specification, VOH defects having a distribution similar to the chemical concentration distribution of hydrogen and VOH defects having a distribution similar to the vacancy defects in the through zone 106 are both referred to as hydrogen donors or hydrogen as donors.

[0093] It should be noted that the dopant used to set the basic doping concentration Db is a dopant added when the semiconductor ingot is manufactured. As an example, in the case where the semiconductor ingot is silicon, if it is n-type, it is phosphorus, antimony, arsenic, and if it is p-type, it is boron, aluminum, etc. In addition, in the case of compound semiconductors other than silicon and oxide semiconductors, they can also be respective dopants. In addition, the manufacturing method of the semiconductor ingot can be any of the floating zone (FZ) method, the Czochralski (CZ) method, and the Czochralski (MCZ) method with applied magnetic field.

[0094] Generally, it is necessary to prepare a semiconductor substrate 10 having a basic doping concentration Db corresponding to the characteristics of the element to be formed on the semiconductor substrate 10, in particular, the rated voltage or the withstand voltage. Figure 1 The semiconductor device 100 shown can make the donor concentration of the semiconductor substrate 10 and the range of the through-zone 106 locally higher than the basic doping concentration Db when the semiconductor device 100 is completed by controlling the dose and implantation depth of hydrogen ions. Therefore, even if a semiconductor substrate 10 with a different basic doping concentration is used, an element with a predetermined rated voltage or withstand voltage characteristic can be formed. In addition, although the deviation of the donor concentration during the manufacture of the semiconductor substrate 10 is relatively large, the dose of hydrogen ions can be controlled with relatively high accuracy. Therefore, the concentration of vacancies (V) generated by the implantation of hydrogen ions can also be controlled with high accuracy, and the donor concentration of the through-zone 106 can be controlled with high accuracy.

[0095] Figure 2 express Figure 1 The hydrogen concentration distribution, donor concentration distribution, and vacancy defect concentration distribution 175 in the depth direction at the position indicated by the AA line. Figure 2The horizontal axis of represents the depth position from the lower surface 23, and the vertical axis uses a logarithmic axis to represent the hydrogen concentration, donor concentration, and vacancy defect concentration per unit volume. It should be noted that the vacancy defect concentration distribution 175 is the distribution immediately after the hydrogen ions used to obtain the second hydrogen concentration distribution described below are implanted. When the semiconductor device 100 is completed, the vacancies are reduced or disappeared compared to immediately after the ion implantation, showing the same Figure 2 Different concentration distributions. Figure 2 The hydrogen concentration in is, for example, a chemical concentration measured by the SIMS method. Figure 2 The donor concentration in is, for example, the electrically activated doping concentration measured by the CV method or the SR method. Figure 2 In FIG. 1 , the hydrogen concentration distribution and the vacancy defect concentration distribution 175 are represented by dotted lines, and the donor concentration distribution is represented by a solid line.

[0096] The hydrogen concentration distribution has a first hydrogen concentration peak 131 and a second hydrogen concentration peak 141. The first hydrogen concentration peak 131 has a local maximum at a first depth Z1, and the second hydrogen concentration peak 141 has a local maximum at a second depth Z2.

[0097] The donor concentration distribution has a first donor concentration peak 111 and a second donor concentration peak 121. The first donor concentration peak 111 presents a local maximum at the first depth Z1. The second donor concentration peak 121 presents a local maximum at the second depth Z2. However, the position where the first donor concentration peak 111 presents a local maximum may not be strictly consistent with the first depth Z1. For example, if the position where the first donor concentration peak 111 presents a local maximum is included in the range of the half-value full width of the first hydrogen concentration peak 131 when the first depth Z1 is used as a reference, the first donor concentration peak 111 can be substantially configured at the first depth Z1. Similarly, if the position where the second donor concentration peak 121 presents a local maximum is included in the range of the half-value full width of the second hydrogen concentration peak 141 when the second depth Z2 is used as a reference, the second donor concentration peak 121 can be substantially configured at the second depth Z2.

[0098] Vacancy defect concentration distribution 175 has a first vacancy concentration peak corresponding to first hydrogen concentration peak 131 and a second vacancy concentration peak (vacancy concentration peak 171) corresponding to second hydrogen concentration peak 141. However, the first vacancy concentration peak is omitted from illustration. Vacancy concentration peak 171 has a local maximum at depth Zd2.

[0099] As the concentration peaks are directed from the lower surface 23 of the semiconductor substrate 10 toward the upper surface 21, each concentration peak has an upward slope with an increasing concentration value and a downward slope with a decreasing concentration value. In this example, the first hydrogen concentration peak 131 has an upward slope 132 and a downward slope 133. The second hydrogen concentration peak 141 has an upward slope 142 and a downward slope 143. The first donor concentration peak 111 has an upward slope 112 and a downward slope 113. The second donor concentration peak 121 has an upward slope 122 and a downward slope 123. The vacancy concentration peak 171 has an upward slope 172 and a downward slope 173.

[0100] The donors of the semiconductor substrate 10 include donors present in the semiconductor substrate 10 before the hydrogen ion is injected, that is, the basic doping concentration (concentration Db), donors activated by the injected hydrogen, and the above-mentioned VOH defects. The ratio of hydrogen activated as a donor is, for example, about 1%. In the area separated to a certain extent from the first depth Z1 and the second depth Z2 in the through zone 106, the ratio of donors formed by VOH defects corresponding to the vacancy defect concentration is higher than that of donors formed by VOH defects corresponding to the hydrogen chemical concentration, and the donor concentration is limited by the concentration of vacancy defects. VOH defects corresponding to the hydrogen chemical concentration refer to VOH defects that are determined by the hydrogen chemical concentration compared with the vacancy defect concentration. VOH defects corresponding to the vacancy defect concentration refer to VOH defects that are determined by the vacancy defect concentration compared with the hydrogen chemical concentration.

[0101] Here, the VOH defect determined by the hydrogen chemical concentration distribution means the following. When vacancies, oxygen, and hydrogen form clusters to form VOH defects, since the hydrogen chemical concentration is sufficiently greater than the vacancy defect concentration, the distribution of the donor concentration formed by the VOH defect is similar to the distribution of the hydrogen chemical concentration. As an example, when the hydrogen chemical concentration is greater than the vacancy defect concentration at a certain depth and a depth near it, it can be said that the donor concentration distribution of the VOH defect is determined by the hydrogen chemical concentration distribution.

[0102] On the other hand, the VOH defect determined by the vacancy defect concentration distribution 175 refers to a situation where the distribution of the donor concentration formed by the VOH defect is similar to the distribution of the vacancy defect concentration because the vacancy defect concentration is sufficiently large relative to the hydrogen chemical concentration. As an example, when the vacancy defect concentration is greater than the hydrogen chemical concentration at a certain depth and a depth near it, it can be said that the donor concentration distribution of the VOH defect determined by the vacancy defect concentration distribution 175 is obtained.

[0103] In the passing region 106, except for the vicinity of the first depth Z1 and the second depth Z2, it is considered that vacancies (V, VV, etc.) generated by the passing of hydrogen are as follows: Figure 2As shown, the concentration distribution is substantially uniform in the depth direction. In addition, it is considered that oxygen (O) implanted during the manufacture of the semiconductor substrate 10 is also uniformly distributed in the depth direction. In addition, in the through region 106, since the hydrogen of the first hydrogen concentration peak 131 and the hydrogen of the second hydrogen concentration peak 141 diffuse from both sides in the depth direction, there is a sufficient amount of hydrogen. They form a flat donor distribution as VOH defects.

[0104] In the vicinity of the second hydrogen concentration peak 141, it is considered that vacancy defects, especially dangling bonds, formed by the injection of hydrogen ions are terminated by the injected hydrogen. Therefore, the second depth Zd2 of the vacancy concentration peak 171 is sometimes located slightly closer to the injection surface side of the ion injection or the first hydrogen concentration peak 131 side than the second depth Z2 of the second hydrogen concentration peak 141. The same is true for the first vacancy concentration peak, but the position shift is sufficiently small compared to the shift of Z2 and Zd2, so it is omitted in the figure.

[0105] Therefore, in the through region 106 except for the vicinity of the first depth Z1 and the second depth Z2, there is a flat region 150 in which the VOH defects functioning as donors are substantially uniformly distributed. The donor concentration distribution in the flat region 150 is substantially constant in the depth direction. The donor concentration being substantially constant in the depth direction may mean, for example, that a region in which the difference between the maximum and minimum values ​​of the donor concentration is within 50% of the maximum value of the donor concentration is continuous in the depth direction. The difference may be less than 30% of the maximum value of the donor concentration in the region, or less than 10%.

[0106] Alternatively, with respect to the average concentration of the donor concentration distribution in the predetermined range in the depth direction, the value of the donor concentration distribution may be within ±50%, within ±30%, or within ±10% of the average concentration of the donor concentration distribution. As an example, the predetermined range in the depth direction may be as follows. That is, the length from the first depth Z1 to the second depth Z2 is set to Z L , the center Zc between Z1 and Z2 can be separated by 0.25Z to the first depth Z1 side and the second depth Z2 side respectively. L The length between the two points is 0.5Z L The length of the predetermined range can be set to 0.75Z corresponding to the length of the flat area 150. L , or it can be set to 0.3Z L , you can also set it to 0.9Z L .

[0107] The range of setting the flat region 150 can be controlled by the position of the second hydrogen concentration peak 141. The flat region 150 is set between the first hydrogen concentration peak 131 and the second hydrogen concentration peak 141. In addition, the donor concentration of the flat region 150 can be controlled by the dose of hydrogen ions at the second hydrogen concentration peak 141. By increasing the dose of hydrogen ions, more vacancies (V) are generated in the passing region 106, and the donor concentration increases.

[0108] It should be noted that when hydrogen ions are injected into a position deeper than the first hydrogen concentration peak 131, the acceleration energy of the hydrogen ions can be increased to a value at which hydrogen penetrates (penetrates) the semiconductor substrate 10. That is, the second hydrogen concentration peak 141 may not remain in the semiconductor substrate 10. Thus, the concentration of vacancy defects can also be increased. On the other hand, when the acceleration energy is too large, the damage to the substrate during ion injection is too strong and sometimes the distribution of vacancy defects through the region 106 cannot maintain flatness. Therefore, the second hydrogen concentration peak 141 can be located inside the semiconductor substrate 10.

[0109] Since the second hydrogen concentration peak 141 is located deeper than the first hydrogen concentration peak 131, the peak extension is likely to be larger than the first hydrogen concentration peak 131. Therefore, the second donor concentration peak 121 is likely to be larger than the first donor concentration peak 111. That is, the second donor concentration peak 121 is likely to be a gentler peak than the first donor concentration peak 111.

[0110] In addition, in this example, the concentration value of the first hydrogen concentration peak 131 is greater than the concentration value of the second hydrogen concentration peak 141. The concentration value of the first hydrogen concentration peak 131 may be more than 10 times the concentration value of the second hydrogen concentration peak 141, or more than 100 times. In another example, the concentration value of the first hydrogen concentration peak 131 may be less than the concentration value of the second hydrogen concentration peak 141.

[0111] In this example, since the concentration value of the first hydrogen concentration peak 131 is high, the amount of hydrogen activated as a donor is also relatively large at the first hydrogen concentration peak 131. That is, the ratio of the donor of the VOH defect determined by the hydrogen chemical concentration distribution becomes higher relative to the vacancy defect concentration distribution 175. At this time, the shape of the first donor concentration peak 111 is similar to the shape of the first hydrogen concentration peak 131.

[0112] On the other hand, the concentration value of the second hydrogen concentration peak 141 is relatively low, and the amount of hydrogen activated as a donor at the second hydrogen concentration peak 141 is relatively small. That is, the ratio of the donors of the VOH defects determined by the vacancy defect concentration distribution 175 is relatively high compared to the donors of the VOH defects determined by the hydrogen chemical concentration distribution. At this time, the similarity between the shape of the second donor concentration peak 121 and the shape of the second hydrogen concentration peak 141 is smaller than the similarity between the shape of the first donor concentration peak 111 and the shape of the first hydrogen concentration peak 131. Since it is believed that the VOH defects are basically uniformly distributed in most of the through-region 106, the second donor concentration peak 121 becomes a more gentle shape. The similarity of the peak shape can be an indicator that the greater the difference in the slope of the corresponding slope between the hydrogen concentration peak and the donor concentration peak, the smaller the value represented by the similarity of the peak shape.

[0113] With such a configuration, the flat region 150 can be provided between the first depth Z1 and the second depth Z2. Since hydrogen diffuses from both sides in the depth direction relative to the through region 106, it is easy to form a long flat region 150 in the depth direction. The length of the flat region 150 in the depth direction may be 10% or more, 30% or more, or 50% or more of the thickness of the semiconductor substrate 10 in the depth direction. In addition, the length of the flat region 150 in the depth direction may be 10 μm or more, 30 μm or more, 50 μm or more, or 100 μm or more.

[0114] The minimum value of the donor concentration of the flat region 150 may be higher than the basic doping concentration Db of the semiconductor substrate 10. That is, the donor concentration of the flat region 150 may be higher than the basic doping concentration Db throughout the flat region 150. The difference between the donor concentration of the flat region 150 and the basic doping concentration Db of the semiconductor substrate 10 may be adjusted by, for example, the hydrogen dosage at the second hydrogen concentration peak 141.

[0115] The minimum value of the donor concentration between the first depth Z1 and the second depth Z2 may be higher than the basic doping concentration of the semiconductor substrate 10. An N-type region may be continuously provided between the first hydrogen concentration peak 131 and the second hydrogen concentration peak 141. In addition, the minimum value of the donor concentration between the second depth Z2 and the lower surface 23 of the semiconductor substrate 10 may be higher than the donor concentration of the semiconductor substrate 10.

[0116] Figure 3A 1 is a diagram illustrating the relationship between the first hydrogen concentration peak 131 and the first donor concentration peak 111. In this example, the slope 114 of the upward slope 112 of the first donor concentration peak 111 is normalized using the slope 134 of the upward slope 132 of the first hydrogen concentration peak 131. As an example, the normalization is a process of dividing the slope 134 by the slope 114.

[0117] The slope of the upward slope may be the slope of the position where the concentration presents a maximum value and the position where the concentration has a predetermined ratio relative to the maximum value. The predetermined ratio may be 80%, 50%, 10%, 1%, or other ratios. In addition, at the first hydrogen concentration peak 131 and the first donor concentration peak 111, the slope of the concentration distribution between the first depth Z1 and the lower surface 23 of the semiconductor substrate 10 may be used. Figure 3A In the example shown, the slope 134 of the first hydrogen concentration peak 131 is given by (H1-aH1) / (Z1-Z3), and the slope 114 of the first donor concentration peak 111 is given by (D1-aD1) / (Z1-Z4). H1 is the hydrogen concentration at the first depth Z1, D1 is the donor concentration at the first depth Z1, a is a predetermined ratio, Z3 is the depth at which the hydrogen concentration becomes aH1 at the upward slope 132 of the first hydrogen concentration peak 131, and Z4 is the depth at which the donor concentration becomes aD1 at the upward slope 112 of the first donor concentration peak 111. For example, if the slope 114 is normalized by the slope 134, it becomes (D1-aD1)(Z1-Z3) / {(H1-aH1)(Z1-Z4)}. The slope obtained by normalizing the slope 114 by the slope 134 is set to α.

[0118] Figure 3B 1 is a diagram illustrating the relationship between the second hydrogen concentration peak 141 and the second donor concentration peak 121. In this example, the slope 124 of the upward slope 122 of the second donor concentration peak 121 is normalized using the slope 144 of the upward slope 142 of the second hydrogen concentration peak 141.

[0119] exist Figure 3B In the example shown, the slope 144 of the second hydrogen concentration peak 141 is given by (H2-aH2) / (Z2-Z5), and the slope 124 of the second donor concentration peak 121 is given by (D2-aD2) / (Zd2-Z6). H2 is the hydrogen concentration at the second depth Z2, D2 is the donor concentration at the second depth Z2, a is a predetermined ratio, Z5 is the depth at which the hydrogen concentration becomes aH2 at the upward slope 142 of the second hydrogen concentration peak 141, and Z6 is the depth at which the donor concentration becomes aD2 at the upward slope 122 of the second donor concentration peak 121. The ratio a used to normalize the slope of the second donor concentration peak 121 may be the same as or different from the ratio a used to normalize the slope of the first donor concentration peak 111. For example, if the slope 124 is normalized by the slope 144, it becomes (D2-aD2)(Z2-Z5) / {(Zd2-Z6)(H2-aH2)}. The slope obtained by normalizing the slope 124 by the slope 144 is denoted as β.

[0120] The normalized slope β of the upward slope 122 of the second donor concentration peak 121 is smaller than the normalized slope α of the upward slope 112 of the first donor concentration peak 111. That is, the second donor concentration peak 121 becomes a more gentle peak relative to the hydrogen concentration peak than the first donor concentration peak 111. By injecting hydrogen ions in a manner to form such a second donor concentration peak 121, a flat region 150 can be formed. In addition, by making the second donor concentration peak 121 a gentle shape, the change in donor concentration at the front end of the flat region 150 can also be alleviated. The normalized slope β of the upward slope 122 of the second donor concentration peak 121 can be less than 1 times the normalized slope α of the upward slope 112 of the first donor concentration peak 111, or less than 0.1 times, or less than 0.01 times.

[0121] In addition, the slope 144 of the upward slope 142 of the second hydrogen concentration peak 141 can be smaller than the slope 145 of the downward slope 143. The concentration distribution of hydrogen ions injected into a position deeper than the lower surface 23 sometimes introduces a gentle downward swing on the side of the lower surface 23, so by comparing the slope 144 of the upward slope 142 with the slope 145 of the downward slope 143, it is sometimes possible to determine whether the hydrogen of the second hydrogen concentration peak 141 is injected from the side of the lower surface 23. The slope 145 is given by (H2-aH2) / (Z7-Z2). The slope 125 is given by (D2-aD2) / (Z7-Zd2). It should be noted that in Figure 3B , the slope 124 of the upward slope 122 of the second donor concentration peak 121 is greater than the slope 125 of the downward slope 123 , but similarly to the second hydrogen concentration peak 141 , the slope 124 of the upward slope 122 of the second donor concentration peak 121 may be smaller than the slope 125 of the downward slope 123 .

[0122] Figure 3C 142 is a graph illustrating the slope of the upward ramp 142. The slope of the upward ramp 142 can be considered as follows. Figure 3C As recorded, at the second hydrogen concentration peak 141, the width (10% full width) between the two positions Z8 and Z9 where the concentration becomes 10% (0.1×H2) of the peak concentration H2 is set to FW10%M. The two positions Z8 and Z9 are the two positions closest to the peak position Z2 among the points where the hydrogen concentration becomes 0.1×H2 across the peak position Z2. The position on the side of the first hydrogen concentration peak in the two positions Z8 and Z9 is set to Z8. The slope of the donor concentration at position Z8 is basically flat. The slope of the hydrogen concentration at position Z8 exceeds 100 times the slope of the donor concentration at position Z8. For example, the slope of the hydrogen concentration at position Z8 may be more than 100 times the slope of the donor concentration at position Z8, or may be more than 1000 times.

[0123] Figure 4AThis is a diagram illustrating another definition of the normalization of the slope of the upward slope 112. In the normalization of the slope of the upward slope 112, for example, the following index γ is introduced. In the example of 3A, position Z3 is different from position Z4, but in this example, position Z3 and position Z4 are set to the same position (Z3=Z4). Position Z3 here is a predetermined position. Position Z3 can be a position where the hydrogen concentration distribution and the donor concentration distribution are closer to the lower surface side than position Z1 to form the upward slope 132, 112. The hydrogen concentration at position Z3 is set to a×H1, and the donor concentration is set to b×D1. a is the ratio of the hydrogen concentration at position Z3 to the concentration H1 of the first hydrogen concentration peak 131 at position Z1. b is the ratio of the donor concentration at position Z3 to the concentration D1 of the first donor concentration peak 111 at position Z1. Here, the ratio of each slope of the hydrogen concentration and the donor concentration in the interval Z3 to Z1 and the slope ratio γ obtained by normalizing the slope ratio are introduced. The slope ratio of the hydrogen concentration in the interval Z3 to Z1 is defined as (H1 / aH1) / (Z1-Z3). Similarly, the slope ratio of the donor concentration in the interval Z3 to Z1 is defined as (D1 / bD1) / (Z1-Z3). Furthermore, using the slope ratio of the hydrogen concentration in the interval Z3 to Z1, the slope ratio γ obtained by normalizing the slope ratio of the donor concentration is defined as {(D1 / bD1) / (Z1-Z3)} / {(H1 / aH1) / (Z1-Z3)}. The normalized slope ratio γ becomes a simple ratio a / b by calculating the above formula.

[0124] Figure 4B 1 is a diagram for explaining another definition of normalization of the slope of the upward slope 122. In the normalization of the slope of the upward slope 122, for example, an index ε similar to the index γ is introduced. Figure 3BIn the example, position Z5 is different from position Z6, but in this example, position Z5 and position Z6 are set to the same position (Z5=Z6). Position Z5 here is a predetermined position. Position Z5 can be a position where the hydrogen concentration distribution and the donor concentration distribution form an upward slope 142, 122 on the lower surface side compared to position Z2. The hydrogen concentration at position Z5 is set to c×H2, and the donor concentration is set to d×D2. c is the ratio of the hydrogen concentration at position Z5 to the concentration H2 of the second hydrogen concentration peak 141 at position Z2. d is the ratio of the donor concentration at position Z5 to the concentration D2 of the second donor concentration peak 121 at position Z1. Here, the ratio of the slopes of the hydrogen concentration and the donor concentration in the interval Z5 to Z2 and the slope ratio ε obtained by normalizing the ratio of the slopes are introduced. The ratio of the slopes of the hydrogen concentrations in the interval Z5 to Z2 is defined as (H2 / cH2) / (Z2-Z5). Similarly, the ratio of the slopes of the donor concentrations in the interval Z5 to Z2 is defined as (D2 / dD2) / (Zd2-Z5). Furthermore, using the ratio of the slopes of the hydrogen concentrations in the interval Z5 to Z2, the slope ratio ε obtained by normalizing the ratio of the slopes of the donor concentrations is defined as {(D2 / dD2) / (Z2-Z5)} / {(H2 / cH2) / (Zd2-Z5)}. The normalized slope ratio ε is calculated by the above formula to become a simple ratio (c / d){(Zd2-Z5) / (Z2-Z5)}. In addition, if Zd2 is close enough to Z2, the slope ratio ε can be approximated to c / d. As an example, if |Zd2-Z2| is less than 10% of Z2-Z5, the slope ratio ε can be approximated to c / d.

[0125] For the first hydrogen concentration peak 131 and the first donor concentration peak 111, the hydrogen concentration distribution and the donor concentration distribution are mostly similar in shape. Here, becoming similar in shape means that, for example, when the horizontal axis is set to depth and the vertical axis is set to the common logarithm of the concentration, the donor concentration distribution presents a distribution that reflects the hydrogen concentration distribution. That is, in the predetermined interval Z3 to Z1, by performing ion implantation of hydrogen ions and then performing thermal annealing, the donor concentration distribution becomes a distribution that reflects the hydrogen concentration distribution. As an example, if H1 of the first hydrogen concentration peak 131 is set to 1×10 17 at ms / cm 3 , and the hydrogen concentration aH1 at position Z3 is set to 2×10 16 at ms / cm 3 , then a becomes 0.2. On the other hand, if D1 of the first donor concentration peak 111 is set to 1×10 16 at ms / cm 3 , and the donor concentration bD1 at position Z3 is set to 2×10 15 at ms / cm 3, then b becomes 0.2. Therefore, since the normalized slope ratio γ is a / b, it is 1. That is, at the depth position Z1 close to the lower surface, the slope ratio a of the hydrogen concentration distribution and the slope ratio b of the donor concentration distribution become substantially the same value, and can be said to be similar shapes.

[0126] On the other hand, for the second hydrogen concentration peak 141 and the second donor concentration peak 121, the hydrogen concentration distribution and the donor concentration distribution may not be similar in shape. That is, in the predetermined interval Z5 to Z2, the donor concentration distribution may not reflect the hydrogen concentration distribution. As an example, if H2 of the second hydrogen concentration peak 141 is set to 1×10 16 at ms / cm 3 , and the hydrogen concentration cH2 at position Z5 is set to 1×10 15 at ms / cm 3 , then c is 0.1. On the other hand, if D2 of the second donor concentration peak 121 is set to 3×10 14 at ms / cm 3 , and the donor concentration dD2 at position Z5 is set to 1.5×10 14 at ms / cm 3 , then d is 0.5. Therefore, since the normalized slope ratio ε is c / d, it is 0.2. That is, at the depth position Z2 which is deep enough from the lower surface, the slope ratio c of the hydrogen concentration distribution is 0.2 times the slope ratio d of the donor concentration distribution, which can be said to present a shape different from the similarity.

[0127] If the normalized slope ratio γ and ε are compared, when the peak position of the hydrogen concentration distribution is close to the lower surface, γ is close to 1, and when the peak position of the hydrogen concentration distribution is deep enough from the lower surface, ε can be a value much smaller than 1. That is, the normalized slope ratio ε can be smaller than the normalized slope ratio γ. In addition, the slope ratio ε can be less than 0.9, less than 0.5, or less than 0.2. Alternatively, it can be less than 0.1, or less than 0.01.

[0128] In addition, as another example of the second donor concentration peak 121, due to the decrease in carrier mobility, the donor concentration, that is, the carrier concentration, calculated based on the spreading resistance is sometimes lower than the carrier concentration at the previous and next positions at the depth position Z2. In this case, since the upward slope 122 is a decreasing slope, the sign of d is negative. That is, d becomes a negative number with an absolute value greater than 1. Thus, ε becomes a negative number. That is, the normalized slope ratio ε can be less than the normalized slope ratio γ. In addition, the slope ratio ε can be less than 0.9, less than 0, or less than -1. Alternatively, it can be less than -10, or less than -100.

[0129] It should be noted that the actual position of the second hydrogen concentration peak 141 and the actual position of the second donor concentration peak 121 are sometimes different. In addition, the position of the first hydrogen concentration peak may not be completely consistent with the position of the first donor concentration peak, and the position of the second hydrogen concentration peak may not be completely consistent with the position of the second donor concentration peak. In this way, when the position of the hydrogen concentration peak is inconsistent with the position of the donor concentration peak, the position of the hydrogen concentration peak is set to Z1, and for the donor concentration, for convenience, the concentration at Z1 can be used as the peak position. Thus, calculation based on the above definition can be performed.

[0130] The important point in the above description is that the second hydrogen concentration peak 141 has a local value. That is, the distribution of hydrogen concentration has a local value at the depth Z2. The fact that the second hydrogen concentration peak 141 has a local value enables the comparison of the normalized slope ratios described above.

[0131] Figure 5 1 is a diagram for explaining the flat region 150. As described above, the donor concentration distribution in the flat region 150 is substantially constant in the depth direction. The flat region 150 is a portion where the region where the donor concentration is a value between a predetermined maximum value max and a predetermined minimum value min is continuous in the depth direction. The maximum value max may be the maximum value of the donor concentration in the region. The minimum value min may be 50%, 70%, or 90% of the maximum value max.

[0132] Alternatively, as described above, relative to the average concentration of the donor concentration distribution within a predetermined range in the depth direction, the value of the donor concentration distribution may be within ±50%, within ±30%, or within ±10% of the average concentration of the donor concentration distribution. The predetermined range in the depth direction may also be the same as described above.

[0133] Figure 6 1 is a diagram showing a structural example of a semiconductor device 100. The semiconductor device 100 of this example functions as an IGBT. The semiconductor device 100 of this example includes a semiconductor substrate 10, an interlayer insulating film 38, an emitter 52, and a collector 54. The interlayer insulating film 38 is formed so as to cover at least a portion of the upper surface 21 of the semiconductor substrate 10. Through holes such as contact holes are formed in the interlayer insulating film 38. The upper surface 21 of the semiconductor substrate 10 is exposed through the contact holes. The interlayer insulating film 38 may be a silicate glass such as PSG or BPSG, or may be an oxide film or a nitride film.

[0134] The emitter electrode 52 is formed on the upper surface of the semiconductor substrate 10 and the interlayer insulating film 38. The emitter electrode 52 is also formed inside the contact hole and is in contact with the upper surface 21 of the semiconductor substrate 10 exposed through the contact hole.

[0135] The collector electrode 54 is formed on the lower surface 23 of the semiconductor substrate 10. The collector electrode 54 may be in contact with the entire lower surface 23 of the semiconductor substrate 10. The emitter electrode 52 and the collector electrode 54 are formed of a metal material such as aluminum.

[0136] In the semiconductor substrate 10 of this example, an N-type drift region 18, an N+ type emitter region 12, a P-type base region 14, an N+ type accumulation region 16, an N+ type buffer region 20 and a P+ type collector region 22 are provided.

[0137] The emitter region 12 is provided in contact with the upper surface 21 of the semiconductor substrate 10, and is a region having a higher donor concentration than the drift region 18. The emitter region 12 contains N-type impurities such as phosphorus.

[0138] The base region 14 is disposed between the emitter region 12 and the drift region 18. The base region 14 includes P-type impurities such as boron. Figure 6 The base region has a P-type contact region (not shown) arranged alternately with the emitter region 12 in the Y-axis direction. The contact region is formed on the upper surface 21 of the base region and can be formed deeper than the emitter region 12. The contact region suppresses latching of the IGBT when it is turned off.

[0139] The accumulation region 16 is provided between the base region 14 and the drift region 18, and has one or more donor concentration peaks having a higher donor concentration than the donor concentration of the drift region 18. The accumulation region 16 may contain N-type impurities such as phosphorus, or may contain hydrogen.

[0140] The collector region 22 is provided in contact with the lower surface 23 of the semiconductor substrate 10. The acceptor concentration of the collector region 22 may be higher than that of the base region 14. The collector region 22 may contain the same P-type impurities as the base region 14, or may contain different P-type impurities.

[0141] The buffer region 20 is provided between the collector region 22 and the drift region 18, and has one or more donor concentration peaks having a higher donor concentration than that of the drift region 18. The buffer region 20 has N-type impurities such as hydrogen. The buffer region 20 can function as a field stop layer that prevents the depletion layer extending from the lower surface side of the base region 14 from reaching the collector region 22.

[0142] The gate groove portion 40 penetrates the emitter region 12, the base region 14 and the accumulation region 16 from the upper surface 21 of the semiconductor substrate 10 to reach the drift region 18. The accumulation region 16 of this example is arranged at a position above the lower end of the gate groove portion 40. The accumulation region 16 can be provided in a manner covering the entire lower surface of the base region 14. By providing an accumulation region 16 having a higher concentration than the drift region 18 between the drift region 18 and the base region 14, the carrier injection promotion effect (IE effect, Injection-Enhancement effect) can be improved, thereby reducing the on-voltage of the IGBT.

[0143] The gate trench portion 40 includes a gate trench formed on the upper surface side of the semiconductor substrate 10, a gate insulating film 42, and a gate conductive portion 44. The gate insulating film 42 is formed in a manner covering the inner wall of the gate trench. The gate insulating film 42 can be formed by oxidizing or nitriding the semiconductor of the inner wall of the gate trench. The gate conductive portion 44 is formed inside the gate trench at a position closer to the inner side than the gate insulating film 42. That is, the gate insulating film 42 insulates the gate conductive portion 44 from the semiconductor substrate 10. The gate conductive portion 44 is formed of a conductive material such as polysilicon.

[0144] The gate conductive portion 44 includes a region facing the base region 14 via a gate insulating film 42. The gate groove portion 40 in this cross section is covered by the interlayer insulating film 38 on the upper surface of the semiconductor substrate 10, but the gate conductive portion 44 is connected to the gate electrode in other cross sections. If a predetermined gate voltage is applied to the gate conductive portion 44, a channel formed by an inversion layer of electrons is formed in the surface layer of the interface in contact with the gate groove portion 40 in the base region 14.

[0145] The first donor concentration peak 111 may be provided in the buffer region 20. The second donor concentration peak 121 may be provided in an N-type region above the buffer region 20. The second donor concentration peak 121 may be provided between the buffer region 20 and the reservoir region 16. The second donor concentration peak 121 of this example is provided in the drift region 18. The second donor concentration peak 121 may be provided at a position lower than the lower end of the gate trench portion 40, may be provided in contact with the lower end of the gate trench portion 40, or may be provided at a position higher than the lower end of the gate trench portion 40.

[0146] Figure 7 Yes means Figure 6 An example of the carrier concentration distribution in the depth direction at the position of the BB line is shown in FIG. Figure 7 In FIG. 1 , a part of the hydrogen concentration distribution is shown in combination. Figure 7 The vertical axis is a logarithmic axis.

[0147] The carrier concentration distribution in the buffer region 20 of this example has a plurality of peaks 24 arranged at different positions in the depth direction. The peak 24 is a peak of the donor concentration. The peak 24 may have hydrogen as an impurity. By providing a plurality of peaks 24, it is possible to further suppress the depletion layer from reaching the collector region 22. The first donor concentration peak 111 may function as the peak 24 in the buffer region 20.

[0148] As an example, the first donor concentration peak 111 may function as a peak farthest from the bottom surface 23 of the semiconductor substrate 10 among the peaks 24 of the buffer region 20. The flat region 150 is disposed between the first donor concentration peak 111 and the second donor concentration peak 121 included in the buffer region 20.

[0149] The donor concentration of the first donor concentration peak 111 may be higher than the donor concentration of the peak 24 that follows the first donor concentration peak 111 and is away from the lower surface 23 among the plurality of peaks 24 of the buffer zone 20. By increasing the concentration of the first donor concentration peak 111, it is easy to form a flat region 150 even if the first donor concentration peak 111 is separated from the second donor concentration peak 121. The hydrogen chemical concentration distribution may have one or more hydrogen concentration peaks 194 between the first depth Z1 and the lower surface 23. The hydrogen concentration peaks 194 may be arranged at Figure 6 The buffer zone 20 described in the above description. The hydrogen concentration peak 194 may be arranged at the same depth position as the peak 24 .

[0150] The accumulation region 16 of this example has a plurality of peaks 25. The peak 25 is a peak of donor concentration. The second donor concentration peak 121 is arranged closer to the lower surface 23 than the accumulation region 16. A region (basic doping region 180) having a basic doping concentration Db of the substrate may be provided between the second donor concentration peak 121 and the accumulation region 16. In another example, the donor concentration between the second donor concentration peak 121 and the accumulation region 16 may be higher than the basic doping concentration Db of the semiconductor substrate.

[0151] In addition, the semiconductor device 100 may use a non-doped substrate in which a dopant such as phosphorus (P) is not doped throughout the semiconductor ingot during the manufacture of the semiconductor ingot as the semiconductor substrate 10. In this case, the basic doping concentration Dn of the semiconductor substrate 10 is lower than the basic doping concentration Db. Figure 7 In the example of FIG. 1 , a region with a doping concentration of a basic doping concentration Dn is set as the non-doped region 181. The basic doping concentration Dn of the non-doped region 181 is, for example, 1×10 10 atoms / cm 3 Above and 5×10 12 atoms / cm 3 The basic doping concentration Dn can also be 1×10 11 atoms / cm 3 The basic doping concentration Dn can be 5×10 12 atoms / cm 3 Hereinafter, each concentration in the present specification may be a value at room temperature. As an example, the value at room temperature may be a value at 300K (Kelvin) (about 26.9°C).

[0152] Figure 8 1 is a diagram showing another structural example of the semiconductor device 100. The semiconductor device 100 of this example is Figure 6 and Figure 7The semiconductor device 100 described in the above is different in that the second donor concentration peak 121 (and the second hydrogen concentration peak 141) are arranged in the accumulation region 16. Other structures may be different from those Figure 6 and Figure 7 The semiconductor device 100 is the same as that described in .

[0153] Fig. 9 Yes means Figure 8 An example of the carrier concentration distribution in the depth direction at the position of the CC line. Fig. 9 In FIG. 1 , a part of the hydrogen concentration distribution is shown in combination. Fig. 9 The vertical axis is a logarithmic axis.

[0154] The carrier concentration distribution in the accumulation region 16 of this example has multiple peaks arranged at different positions in the depth direction. The peak is a peak of donor concentration. The peak in the accumulation region 16 may have hydrogen or phosphorus as an impurity. By providing multiple peaks in the accumulation region 16, the displacement current to the gate groove portion can be suppressed in a structure in which the gate groove portion and the dummy groove portion are adjacently arranged (for example, refer to WO2018 / 030440). The dummy groove portion has the same structure as the gate groove portion and is a groove portion to which the emitter potential is applied.

[0155] The second donor concentration peak 121 of this example functions as any donor concentration peak in the accumulation region 16. As an example, the second donor concentration peak 121 can function as a peak farthest from the upper surface 21 of the semiconductor substrate 10 among the multiple peaks of the accumulation region 16. The flat region 150 of this example is arranged between the first donor concentration peak 111 included in the buffer region 20 and the second donor concentration peak 121 included in the accumulation region 16.

[0156] The donor concentration of the second donor concentration peak 121 may be lower than or equal to the donor concentration of other peaks 25 in the buffer region 20, or higher than the donor concentration of other peaks 25 in the buffer region 20. In the case where the buffer region 20 has three or more peaks, the donor concentrations of the peaks 25 other than the second donor concentration peak 121 may be the same. The donor concentration of the second donor concentration peak 121 may be determined based on the donor concentration that the flat region 150 should have.

[0157] The carrier concentration of the drift region 18 in this example can be higher than the base doping concentration Db of the substrate in the entire depth direction. With such a configuration, the overall carrier concentration of the drift region 18 can be adjusted with high precision.

[0158] Peak 25 other than the second donor concentration peak 121 may be a peak formed by donors other than hydrogen. For example, peak 25 is a peak in which phosphorus acts as a donor. By using phosphorus as a donor, VOH defects are less likely to occur, and the donor concentration of peak 25 and its vicinity can be easily controlled by the concentration of phosphorus. In addition, the width of the second donor concentration peak 121 in the depth direction may be made wider than the width of peak 25 in the depth direction. Thus, the displacement current to the gate trench portion can be further suppressed. In addition, between the second donor concentration peak 121 and peak 25 in the storage region 16, as shown in FIG. Fig. 9 As shown, there is a valley in the donor concentration distribution. Alternatively, it can be Fig. 9 The dotted line between the two peaks indicates that the donor concentration distribution in the accumulation region 16 may be a knot instead of a valley.

[0159] Fig.10 express Figure 1 The hydrogen concentration distribution and carrier concentration distribution in the depth direction at the position shown by the AA line. The carrier concentration is measured by the SR method. Near the range (Z2) of the second hydrogen concentration peak 141, more defects are likely to be generated than through the zone 106. Due to the defects that remain without combining with hydrogen, sometimes a valley 151 is generated in the carrier concentration distribution near the second depth Z2.

[0160] If the valley 151 is generated, it is sometimes more difficult to calculate the slope of the upward slope 122 of the second donor concentration peak 121. Therefore, the slope of the upward slope 122 is preferably calculated by removing the influence of the valley 151. For example, the concentrations at the second depth Z2 and the concentrations at the depth Zc can be used to calculate the slopes of the upward slopes of the second donor concentration peak 121 and the second hydrogen concentration peak 141. The depth Zc can be a position closer to the lower surface 23 than the valley 151. In this example, the depth Zc is the depth of the center of the flat area 150 in the depth direction. As another example, the position where the hydrogen concentration distribution becomes extremely small between the first hydrogen concentration peak 131 and the second hydrogen concentration peak 141 can be set to Zc. In this way, the influence of the valley 151 can be removed, and the slope of the upward slope 122 of the second donor concentration peak 121 can be normalized. As described above, the normalization can use the slope of the concentration difference or the slope of the concentration ratio.

[0161] Fig.11 2 is a diagram showing an example of arrangement of elements on the upper surface 21 of the semiconductor substrate 10. Fig.11 In the embodiment, the outer peripheral end of the semiconductor substrate 10 is defined as the outer peripheral end 140 .

[0162] The semiconductor device 100 includes an active portion 120 and an edge termination structure portion 90. The active portion 120 is a region where a main current flows between the upper surface 21 and the lower surface 23 of the semiconductor substrate 10 when the semiconductor device 100 is controlled to be in an on state. That is, it is a region where a current flows in the depth direction inside the semiconductor substrate 10 from the upper surface 21 to the lower surface 23 of the semiconductor substrate 10, or from the lower surface 23 to the upper surface 21.

[0163] The active portion 120 of this example is provided with a transistor portion 70 and a diode portion 80. The transistor portion 70 and the diode portion 80 may be arranged side by side in the X-axis direction. Fig.11 In the example of , the transistor portion 70 and the diode portion 80 are arranged to contact each other alternately in the X-axis direction. In the active portion 120, the transistor portion 70 may be provided at both ends in the X-axis direction. The emitter 52 may cover the transistor portion 70 and the diode portion 80. The active portion 120 may also refer to a region covered by the emitter 52.

[0164] The transistor portion 70 of this example has Figure 6 to Figure 10 The diode section 80 of this example has a FWD (freewheeling diode). In each diode section 80, an N+ type cathode region 82 is provided in a region in contact with the lower surface 23 of the semiconductor substrate 10. Fig.11 In FIG. 1 , the diode portion 80 indicated by the solid line is a region where the cathode region 82 is provided on the lower surface 23 of the semiconductor substrate 10. In the semiconductor device 100 of this example, the collector region 22 is provided in a region other than the cathode region 82 in a region in contact with the lower surface 23 of the semiconductor substrate 10.

[0165] The diode section 80 is a region formed by projecting the cathode region 82 along the Z-axis direction. The transistor section 70 is a region in which the collector region 22 is provided on the lower surface 23 of the semiconductor substrate 10, and unit structures including the emitter region 12 are periodically provided on the upper surface 21 of the semiconductor substrate 10. The boundary between the diode section 80 and the transistor section 70 in the Y-axis direction is the boundary between the cathode region 82 and the collector region 22. The region formed by projecting the cathode region 82 extends in the Y-axis direction to the end of the active section 120 or the portion of the gate flow channel 48 (at Fig.11 The portion indicated by the dotted line extending the diode portion 80 may also be included in the diode portion 80. The emitter region 12 is not provided in this extended portion.

[0166] The semiconductor device 100 of this example further includes a gate metal layer 50 and a gate flow channel 48. In addition, the semiconductor device 100 may include pads such as a gate pad 116 and an emitter pad 118. The gate pad 116 is electrically connected to the gate metal layer 50 and the gate flow channel 48. The emitter pad 118 is electrically connected to the emitter 52.

[0167] The gate metal layer 50 may be provided in a manner to surround the active portion 120 in a plan view. The gate pad 116 and the emitter pad 118 may be arranged in a region surrounded by the gate metal layer 50. The gate metal layer 50 may be formed of a metal material such as aluminum or an aluminum-silicon alloy. The gate metal layer 50 is insulated from the semiconductor substrate 10 by the interlayer insulating film 38. Fig.11 In the embodiment, the interlayer insulating film 38 is omitted. In addition, the gate metal layer 50 is provided in a manner separated from the emitter 52. The gate metal layer 50 transmits the gate voltage applied to the gate pad 116 to the transistor portion 70. The gate conductive portion 44 of the transistor portion 70 is directly connected to the gate metal layer 50 or indirectly connected via other conductive components.

[0168] The gate runner 48 connects the gate metal layer 50 and the gate conductive portion 44 . The gate runner 48 may be formed of a semiconductor material such as polysilicon doped with impurities. A portion of the gate runner 48 may be disposed above the active portion 120 . Fig.11 The gate runner 48 shown is provided so as to cross the active portion 120 in the X-axis direction. Thus, it is possible to suppress a decrease and a delay in the gate voltage even inside the active portion 120 separated from the gate metal layer 50. A portion of the gate runner 48 may be arranged so as to surround the active portion 120 along the gate metal layer 50. The gate runner 48 may be connected to the gate conductive portion 44 at the end of the active portion 120.

[0169] The edge terminal structure portion 90 is provided between the active portion 120 and the peripheral end 140 of the semiconductor substrate 10 on the upper surface 21 of the semiconductor substrate 10. In this example, a gate metal layer 50 is arranged between the edge terminal structure portion 90 and the active portion 120. The edge terminal structure portion 90 can be arranged in a ring shape in a manner of surrounding the active portion 120 on the upper surface 21 of the semiconductor substrate 10. The edge terminal structure portion 90 of this example is arranged along the peripheral end 140 of the semiconductor substrate 10. The edge terminal structure portion 90 alleviates the electric field concentration on the upper surface 21 side of the semiconductor substrate 10. The edge terminal structure portion 90 has, for example, a guard ring, a field plate, a surface electric field reduction portion, and a structure formed by combining these.

[0170] Fig.12 Yes means Fig.11 Figure 1 shows an example of a c-c' section in FIG. Fig.12 In Figures 1 to 10 The configuration example of the through area 106 in the cross section is described in FIG. Fig.12 In FIG. 1 , the through area 106 is shaded with oblique lines. Fig.12 , only the passing region 106 in the drift region 18 is shown, and the passing regions 106 in the buffer region 20 , the collector region 22 , and the cathode region 82 are omitted.

[0171] Fig.12 The cross section shown is an XZ plane including the edge termination structure 90, the transistor section 70, and the diode section 80. It should be noted that although the gate metal layer 50 and the gate flow channel 48 are arranged between the edge termination structure 90 and the transistor section 70, Fig.12 The structure of the transistor portion 70 is the same as that of Figure 6 to Figure 10 The same as the IGBT described in .

[0172] The diode portion 80 includes a base region 14, a drift region 18, a cathode region 82, and a dummy trench portion 30 inside the semiconductor substrate 10. The base region 14 and the drift region 18 are the same as those in the transistor portion 70.

[0173] In the diode portion 80, a base region 14 or a contact region 15 may be provided in a region in contact with the upper surface 21 of the semiconductor substrate 10. The contact region 15 is a P+ type region having a doping concentration higher than that of the base region 14. In the diode portion 80 of this example, the emitter region 12 is not provided. In addition, the diode portion 80 may or may not be provided with an accumulation region 16.

[0174] The dummy trench portion 30 has the same structure as the gate trench portion 40. However, the dummy trench portion 30 is electrically connected to the emitter 52. The dummy trench portion 30 is provided from the upper surface 21 of the semiconductor substrate 10 through the base region 14 to the drift region 18. The dummy trench portion 30 may also be provided in the transistor portion 70. In the transistor portion 70, the dummy trench portion 30 and the gate trench portion 40 may be arranged at a predetermined period.

[0175] An intermediate boundary region 190 may be provided between the transistor portion 70 and the diode portion 80. The intermediate boundary region 190 is a region where the operation of the transistor portion 70 and the operation of the diode portion 80 are not directly performed. As an example, the region of the intermediate boundary region 190 in contact with the upper surface 21 may have the same structure as the upper surface 21 side of the diode portion 80. In addition, in a plan view, the collector region of the transistor portion 70 may be extendedly provided in the region of the intermediate boundary region 190 in contact with the lower surface 23. Fig.12 In FIG. 1 , only the arrow is used to indicate the range of the middle boundary area 190. Fig.12 , the range illustrated as the intermediate boundary region 190 also has the same structure as the transistor portion 70 .

[0176] In the drift region 18 of the diode portion 80, a lifetime control region 192 may be provided at a position closer to the upper surface 21 than the center in the depth direction. The lifetime control region 192 is a region where a recombination center of carriers (electrons or holes) is provided at a higher concentration than the periphery. The recombination center may be a vacancy-type defect such as a vacancy or a compound vacancy, or may be a dislocation, or may be an interstitial atom, or may be a transition metal, etc. The lifetime control region 192 may extend from the diode portion 80 to the intermediate boundary region 190.

[0177] A plurality of guard rings 92, a plurality of field plates 94, and a channel stop region 174 are provided in the edge termination structure portion 90. In the edge termination structure portion 90, a collector region 22 may be provided in a region in contact with the lower surface 23. Each guard ring 92 may be provided in a manner of surrounding the active portion 120 on the upper surface 21. The plurality of guard rings 92 may have a function of extending the depletion layer generated in the active portion 120 to the outside of the semiconductor substrate 10. Thus, it is possible to prevent the electric field from concentrating inside the semiconductor substrate 10, and to improve the withstand voltage of the semiconductor device 100.

[0178] The guard ring 92 of this example is a P+ type semiconductor region formed by ion implantation near the upper surface 21. The bottom of the guard ring 92 may be deeper than the bottoms of the gate trench 40 and the dummy trench 30.

[0179] The upper surface of the guard ring 92 is covered by the interlayer insulating film 38. The field plate 94 is formed of a conductive material such as metal or polysilicon. The field plate 94 can be formed of the same material as the gate metal layer 50 or the emitter 52. The field plate 94 is disposed on the interlayer insulating film 38. The field plate 94 passes through a through hole provided in the interlayer insulating film 38 and is connected to the guard ring 92.

[0180] A protective film 182 is provided on the upper surface 21 side of the semiconductor substrate 10. The protective film 182 may cover the edge terminal structure portion 90, the gate metal layer 50, the boundary portion 72, and a portion of the active portion that contacts the boundary portion 72. The protective film 182 may be an insulating film or an organic film. The protective film 182 in this example is polyimide. A plating layer 184 may be provided on the entire surface of the portion of the emitter 52 that is exposed without forming the protective film 182. The surface of the plating layer 184 may be located closer to the upper surface 21 side than the surface of the protective film 182. The plating layer 184 is connected to the electrode terminal of the power module on which the semiconductor device 100 is mounted.

[0181] The channel stopper region 174 is provided to be exposed on the upper surface 21 and the side surface of the peripheral end 140. The channel stopper region 174 is an N-type region with a higher doping concentration than the drift region 18. The channel stopper region 174 has a function of terminating the depletion layer generated in the active portion 120 at the peripheral end 140 of the semiconductor substrate 10.

[0182] A boundary portion 72 may be provided between the transistor portion 70 and the edge termination structure portion 90. The boundary portion 72 may have a contact region 15, a base region 14, and a dummy groove portion 30 on the upper surface 21 side of the semiconductor substrate 10. The boundary portion 72 may have a P+ type well region 11 having a doping concentration higher than that of the base region 14. The well region 11 is provided in contact with the upper surface 21 of the semiconductor substrate 10. A gate metal layer 50 and a gate flow channel 48 may be provided above the well region 11. The depth of the bottom of the well region 11 may be the same depth as the bottom of the guard ring 92. A portion of the groove portion in the boundary portion 72 may be formed in the well region 11. In the boundary portion 72, a collector region 22 may be provided in an area in contact with the lower surface 23.

[0183] In this example, the second hydrogen concentration peak 141 is disposed between the bottom of the gate trench portion 40 in the Z-axis direction and the upper surface 21 of the semiconductor substrate 10. Fig.12 In the example of FIG. 1 , the second hydrogen concentration peak 141 is disposed at a position deeper than the accumulation region 16, but the second hydrogen concentration peak 141 may be disposed at the same depth as the accumulation region 16, may be disposed at the same depth as the base region 14, or may be disposed at the same depth as the emitter region 12. Fig.12 As shown by the dotted line, the storage region 16 may also be formed in the diode portion 80 .

[0184] Passing region 106 is formed in the range from lower surface 23 of semiconductor substrate 10 to second hydrogen concentration peak 141. In each figure, second hydrogen concentration peak 141 and passing region 106 do not overlap, but passing region 106 is formed to the depth of second hydrogen concentration peak 141.

[0185] In addition, the through region 106 of this example is provided in each of the transistor section 70, the diode section 80, the boundary section 72, and the edge termination structure section 90. In each of the transistor section 70, the diode section 80, the boundary section 72, and the edge termination structure section 90, the depth of the through region 106 can be the same. The through region 106 can be provided in the entire semiconductor substrate 10 when viewed from above. According to this example, the donor concentration can be adjusted over substantially the entire depth direction of the semiconductor substrate 10. In addition, the donor concentration can be adjusted over substantially the entire semiconductor substrate 10 when viewed from above.

[0186] In this example, the region where the through region 106 is not formed is particularly provided at a portion of the edge termination structure portion 90 that is in contact with the upper surface 21. The region where the through region 106 is not formed is a portion where the donor concentration becomes the same as the basic doping concentration Db. The region where the through region 106 is not formed is closer to the upper surface 21 than the depth of the second hydrogen concentration peak 141. That is, the region where the through region 106 is not formed can be a region where the doping concentration basically becomes the basic doping concentration Db. The region where the doping concentration is the basic doping concentration Db is set as a basic doping region 180. In this example, the basic doping region 180 is in contact with the upper surface 21 and is provided at a portion shallower than the well region 11.

[0187] Fig.13 106 is a diagram showing another configuration example of the pass-through area 106. Fig.12 The width of the through area 106 in the depth direction is different. Fig.12 The same as the through area 106 in.

[0188] The second hydrogen concentration peak 141 of this example is arranged between the bottom of the gate groove portion 40 and the lower surface 23 of the semiconductor substrate 10. The thickness of the semiconductor substrate 10 in the depth direction is set to T1, and the distance between the second hydrogen concentration peak 141 and the lower surface 23 of the semiconductor substrate 10 is set to T2. The distance T2 corresponds to the thickness of the through zone 106 in the depth direction. The distance T2 can be greater than 40% and less than 60% of the thickness T1. That is, the through zone 106 can be set from the lower surface 23 of the semiconductor substrate 10 to the approximate center of the semiconductor substrate 10 in the depth direction. However, the distance T2 can be appropriately changed.

[0189] As described above, the base doping region 180 is provided at a position closer to the upper surface 21 than the second hydrogen concentration peak 141. The base doping region 180 of this example is a region extending from the bottom surface of the groove portion to the second hydrogen concentration peak 141, and has a depth of approximately T1-T2. In a plan view, the base doping region 180 of this example is provided on the entire surface of the semiconductor substrate 10.

[0190] Fig.14 106 is a diagram showing another configuration example of the pass-through area 106. Fig.12 The configuration of the through area 106 in the top view is different. The configuration in the depth direction can be different from Fig.12 The same as the through area 106 in.

[0191] In this example, the passing region 106 and the second hydrogen concentration peak 141 are not provided in at least a portion of the edge termination structure portion 90 when viewed from above. Fig.14, an example is shown in which the entire edge termination structure portion 90 is not provided with the through region 106 and the second hydrogen concentration peak 141 when viewed from above. In another example, the through region 106 and the second hydrogen concentration peak 141 can be provided at the end of the edge termination structure portion 90 on the side close to the active portion 120. That is, the through region 106 and the second hydrogen concentration peak 141 are not provided in the area of ​​the semiconductor substrate 10 that contacts the peripheral end 140. In this example, since the second hydrogen concentration peak 141 is not provided near the peripheral end 140, the formation of defects near the peripheral end 140 can be suppressed. Therefore, the increase of leakage current at the peripheral end 140 can be suppressed.

[0192] That is, the basic doping region 180 of this example is provided in the region of the semiconductor substrate 10 that contacts the peripheral end 140. The basic doping region 180 of this example is provided in at least a portion of the edge termination structure portion 90 when viewed from above. In addition, the basic doping region 180 may be provided in the entire edge termination structure portion 90 and the boundary portion 72 when viewed from above.

[0193] It should be noted that the configuration of the through region 106 of the boundary portion 72 may be the same as that of the edge termination structure portion 90, the transistor portion 70, or the diode portion 80. Fig.14 , an example in which the passing area 106 is not provided in the boundary portion 72 is shown.

[0194] Fig.15 1 is a diagram showing another configuration example of the through area 106. The configuration of the through area 106 in the depth direction of this example is similar to Fig.13 The same as the through area 106 shown, the top view configuration is the same as Fig.14 That is, the edge termination structure 90 does not have a through region 106. In addition, the transistor portion 70 and the diode portion 80 have a through region 106 extending from the lower surface 23 of the semiconductor substrate 10 to the vicinity of the center in the depth direction of the semiconductor substrate 10.

[0195] The base doping region 180 of this example is formed from the upper surface 21 to the buffer region 20 in the boundary portion 72 and the edge termination structure portion 90. In addition, in the active portion, the base doping region 18 is formed at a position of the drift region 18 on the upper surface 21 side relative to the second hydrogen concentration peak 141. At a position on the upper surface 21 side relative to the second hydrogen concentration peak 141, the base doping region 180 of this example is provided on the entire surface of the semiconductor substrate 10 in a plan view.

[0196] Fig.16A 106 is a diagram showing another configuration example of the pass-through area 106. In this example, Figure 12 to Figure 15 The difference between the example in FIG. 1 and FIG. 2 is that the through region 106 and the second hydrogen concentration peak 141 are not provided in at least a portion of the diode portion 80 when viewed from above. Figure 12 to Figure 15The example described in is the same.

[0197] exist Fig.16A , an example is shown in which the through region 106 and the second hydrogen concentration peak 141 are not provided in the entire diode portion 80 when viewed from above. That is, the basic doping region 180 is provided in the entire diode portion 80 when viewed from above. In addition, the basic doping region 180 is also provided in the entire boundary portion 72 and the edge terminal structure portion 90. In another example, the through region 106 and the second hydrogen concentration peak 141 can be provided at the end of the diode portion 80 that contacts the transistor portion 70. By making the configuration of the through region 106 different in the transistor portion 70 and the diode portion 80, the distribution of the doping concentration of the diode portion 80 and the transistor portion 70 can be appropriately different.

[0198] Fig. 16B In the active portion, the through region 106 and the base doping region 180 are formed to be similar to Fig.16A By forming the through region 106 in the diode portion 80, the expansion of the space charge region during reverse recovery is suppressed, and the waveform vibration during reverse recovery is suppressed. On the other hand, by forming the transistor portion 70 as the base doping region 180, the expansion of the space charge region is promoted when a short circuit occurs, and the injection of holes is promoted, thereby suppressing short circuit damage.

[0199] Fig.17A 1 is a diagram showing another configuration example of the through area 106. The configuration of the through area 106 in the depth direction of this example is similar to Fig.13 The same as the through area 106 shown, the top view configuration is the same as Fig.16A That is, the diode portion 80 does not have the through region 106. In the transistor portion 70, the through region 106 is provided from the lower surface 23 of the semiconductor substrate 10 to the center of the semiconductor substrate 10 in the depth direction. Fig.17A In the example, the second hydrogen concentration peak 141 is retreated from the bottom surface of the groove portion toward the lower surface 23 , and a base doping region 180 is formed on the entire surface in a plan view at a position closer to the upper surface 21 than the second hydrogen concentration peak 141 .

[0200] Fig. 17B FIG. 1 is a diagram showing another configuration example of the pass-through area 106 . Fig. 17B An example is that in the active portion, the through region 106 and the base doping region 180 are formed to be similar to Fig.17A The opposite composition. Fig. 17B In the example, it also plays a role in Fig. 16B Same effect.

[0201] In order to form Figure 14 to Figure 17BIn the second implantation step S1902 described later, hydrogen ions are selectively implanted in the through region 106 as viewed from above. Figure 14 to Figure 17B The photoresist film 200 shown is capable of selective hydrogen ion implantation.

[0202] At this time, before the second implantation step S1902, a photoresist film 200 having a predetermined thickness is selectively formed on a portion of the lower surface 23 of the semiconductor substrate 10. The thickness of the photoresist film 200 is a thickness capable of shielding hydrogen ions.

[0203] After the photoresist film 200 is formed, the second implantation step S1902 is performed. The region where the photoresist film 200 is formed shields the hydrogen ions by the photoresist film 200. Therefore, the region of the semiconductor substrate 10 covered by the photoresist film 200 does not invade the hydrogen ions. The region where the photoresist film 200 is not formed is implanted with hydrogen ions at the second depth position Z2 according to the acceleration energy. It should be noted that Fig.14 , Fig.15 , Fig.16A , Fig. 16B , Fig.17A and Fig. 17B In each example, the photoresist film 200 is formed so as to be in contact with the lower surface 23 of the semiconductor substrate 10. In the step of forming the photoresist film 200, the collector electrode 54 is not provided on the lower surface 23.

[0204] Fig. 17C 1 is a diagram for explaining the minimum film thickness M of the photoresist film 200 for preventing hydrogen ions from invading the semiconductor substrate 10 . Fig. 17C Indicates the film thickness M relative to the range Rp of hydrogen ions.

[0205] The hydrogen ions in this example are injected from the accelerator into the semiconductor substrate 10 without interposing an absorber other than the photoresist film 200. The range Rp of the hydrogen ions is uniquely determined by the acceleration energy of the accelerator.

[0206] In addition, the minimum film thickness M of the photoresist film 200 that can shield hydrogen ions is determined by the acceleration energy of hydrogen ions. Therefore, the minimum film thickness M of the photoresist film 200 can be represented by the range Rp of hydrogen ions. Fig. 17C This is a graph obtained by measuring the relationship between the range Rp of hydrogen ions and the film thickness M at three points and approximating it with a straight line. The relationship between the film thickness M (μm) and the range Rp (μm) can be expressed by the following equation.

[0207] M=1.76×Rp+12.32

[0208] The thickness of the photoresist film 200 is preferably the same as or greater than the minimum film thickness M shown by the above formula.

[0209] Fig.18A 1 is a diagram showing another configuration example of the through region 106. In this example, the through region 10 provided in the edge termination structure portion 90 has a width T5 in the depth direction shorter than the through region 106 provided in the active portion 120 (the diode portion 80 in this example) has a width T4 in the depth direction.

[0210] In the diode portion 80, the second hydrogen concentration peak 141 may be disposed between the bottom of the dummy trench portion 30 and the upper surface 21 of the semiconductor substrate 10. In the edge termination structure portion 90, the second hydrogen concentration peak 141 may be disposed between the guard ring 92 and the lower surface 23 of the semiconductor substrate 10. The width T5 of the through region 106 in the edge termination structure portion 90 may be greater than half the thickness T of the semiconductor substrate 10.

[0211] In addition, the width T3 of the through region 106 provided in the transistor portion 70 in the depth direction may be shorter than the width T4 of the through region 106 provided in the diode portion 80 in the depth direction. That is, in the transistor portion 70, the base doping region 180 is formed deeper than the depth of the groove portion. That is, the second hydrogen concentration peak 141 of the transistor portion 70 is located at a position closer to the lower surface 23 than the bottom surface of the groove portion. In the transistor portion 70, the second hydrogen concentration peak 141 may be arranged between the bottom of the gate groove portion 40 and the lower surface 23 of the semiconductor substrate 10. The width T3 may be the same as the width T5, may be larger than the width T5, or may be smaller than the width T5. The width T3 of the through region 106 in the transistor portion 70 may be larger than half the thickness T of the semiconductor substrate 10. Thus, the base region 14 forming the channel in the transistor portion 70 can be arranged separately from the second hydrogen concentration peak 141. Therefore, the increase of defects near the channel can be suppressed.

[0212] The through region 106 in the boundary portion 72 may have the same structure as the through region 106 in the edge termination structure portion 90, the same structure as the through region 106 in the transistor portion 70, or the same structure as the through region 106 in the diode portion 80. Fig.18A In the example of FIG. 1 , the through region 106 may not be provided in the transistor portion 70 . The through region 106 may not be provided in the diode portion 80 . The through region 106 may not be provided in the edge termination structure portion 90 . The through region 106 may not be provided in the boundary portion 72 .

[0213] Fig.18B 1 is a diagram showing another configuration example of the through region 106. In this example, the through region 106 provided in the edge termination structure portion 90 has a width T5 in the depth direction shorter than the through region 106 provided in the active portion 120 (the transistor portion 70 in this example).

[0214] In the transistor portion 70, the second hydrogen concentration peak 141 may be disposed between the bottom of the gate trench portion 40 and the upper surface 21 of the semiconductor substrate 10. The through region 106 and the second hydrogen concentration peak 141 in the edge termination structure portion 90 are configured similarly to Fig.18A Same as the example.

[0215] The width T4 of the through region 106 provided in the diode portion 80 in the depth direction may be shorter than the width T3 of the through region 106 provided in the transistor portion 70 in the depth direction. That is, in the diode portion 80, the base doping region 180 is formed deeper than the depth of the groove portion. That is, the second hydrogen concentration peak 141 of the diode portion 80 is located at a position closer to the lower surface 23 than the bottom surface of the groove portion. In the diode portion 80, the second hydrogen concentration peak 141 may be arranged between the bottom of the dummy groove portion 30 and the lower surface 23 of the semiconductor substrate 10. The width T4 may be the same as the width T5, may be larger than the width T5, or may be smaller than the width T5. The width T4 of the through region 106 in the diode portion 80 may be larger than half the thickness T of the semiconductor substrate 10.

[0216] The through region 106 in the boundary portion 72 may have the same structure as the through region 106 in the edge termination structure portion 90, the same structure as the through region 106 in the transistor portion 70, or the same structure as the through region 106 in the diode portion 80. Fig.18B In the example of FIG. 1 , the through region 106 may not be provided in the transistor portion 70 . The through region 106 may not be provided in the diode portion 80 . The through region 106 may not be provided in the edge termination structure portion 90 . The through region 106 may not be provided in the boundary portion 72 .

[0217] like Figure 12 to Figure 18B As shown, by adjusting the composition of the through region 106, the donor concentration distribution in the transistor portion 70, the diode portion 80 and the edge termination structure portion 90 can be easily adjusted. The composition of the through region 106 is not limited to Figure 12 to Figure 18B Example shown.

[0218] Fig.19 1 is a diagram showing a process of forming a through region 106 in a method of manufacturing a semiconductor device 100. When forming the through region 106, in a first implantation step S1900, hydrogen ions are implanted from the lower surface 23 of the semiconductor substrate 10 to a first depth Z1. In addition, in a second implantation step S1902, the through region 106 is formed by implanting hydrogen ions from the lower surface 23 of the semiconductor substrate 10 to a second depth Z2. Either the first implantation step S1900 or the second implantation step S1902 may be performed first.

[0219] It should be noted that, in the case where the first implantation step S1900 is performed first, if heat treatment is performed between the first implantation step S1900 and the second implantation step S1902, the donor concentration of the through region 106 may not be increased. That is, if heat treatment is performed before the through region 106 is formed, the hydrogen implanted in the first implantation step S1900 may not be combined with the crystal defects of the through region 106 and escape to the outside of the semiconductor substrate 10. Therefore, it is preferred that heat treatment is not performed between the first implantation step S1900 and the second implantation step S1902. Heat treatment is, for example, a process of heating the semiconductor substrate 10 at 300° C. or above.

[0220] After the first implantation step S1900 and the second implantation step S1902, a diffusion step S1904 is performed. In the diffusion step S1904, the semiconductor substrate 10 is heat-treated so that the hydrogen implanted to the first depth Z1 diffuses toward the through-zone 106. In the diffusion step S1904, the semiconductor substrate 10 may be heated to 300° C. or higher. The heating temperature may be 350° C. or higher. In the diffusion step S1904, the semiconductor substrate 10 may be heated for more than 1 hour, or for more than 3 hours.

[0221] By diffusing hydrogen in the diffusion step S1904, the crystal defects in the through region 106 are combined with hydrogen to become donors. As a result, the donor concentration in the through region 106 can be increased. In the diffusion step S1904, it is preferred that the minimum value of the donor concentration in the through region 106 is higher than the donor concentration (basic doping concentration) of the semiconductor substrate 10 before the first implantation step S1900 and the second implantation step S1902. That is, it is preferred that the donor concentration in the entire through region 106 is higher than the basic doping concentration.

[0222] In order to increase the donor concentration in the entire through-region 106, it is preferred that the hydrogen implanted at the first depth position Z1 diffuses to the vicinity of the second depth position Z2. In the first implantation step S1900, by adjusting the dose of hydrogen implanted at the first depth position Z1, sufficient hydrogen can be diffused to the vicinity of the second depth position Z2. In the first implantation step S1900, it is preferred that the dose of hydrogen be determined in such a way that the minimum value of the donor concentration in the through-region 106 is higher than the basic doping concentration.

[0223] In the buffer zone 20 Fig. 9In the case of the peak 24 of multiple hydrogen donors shown in the figure, in addition to the hydrogen ion implantation in the first implantation step S1900 and the second implantation step S1902, the hydrogen ion implantation may be performed multiple times. The multiple hydrogen ion implantations for forming the peak 24 may be performed in the first implantation step S1900 or in the second implantation step S1902. That is, the first implantation step S1900 may be performed multiple times, and the second implantation step S1902 may be performed multiple times. In addition, the first implantation step S1900 and the second implantation step S1902 may be performed multiple times, respectively.

[0224] When hydrogen ion implantation is performed multiple times in the second implantation step S1902, a photoresist film 200 having a thickness corresponding to the range Rp of each hydrogen ion can be formed (see Fig.14 Alternatively, a minimum film thickness M corresponding to the deepest range Rpmax in multiple hydrogen ion implantations may be formed (see Fig. 17C ) is much thicker than the photoresist film 200. The photoresist film 200 with a film thickness M can be used to perform hydrogen ion implantation at an acceleration energy corresponding to each range Rp. The thickness of the photoresist film 200 can be at least twice the minimum film thickness M corresponding to the deepest range Rpmax. Thus, the durability of the photoresist film 200 against multiple hydrogen ion implantations can be ensured, and the number of times the photoresist film 200 is formed can be reduced.

[0225] Figure 20 to Figure 26 1 is a diagram illustrating a method for determining a dose of hydrogen ions to be implanted into a first depth Z1 (referred to as a first dose). In the first implantation step S1900 of this example, hydrogen is implanted at a dose greater than a minimum dose determined based on the diffusion coefficient of hydrogen in the semiconductor substrate 10 and the position of the second depth Z2 (i.e., the distance that the hydrogen implanted into the first depth position Z1 should diffuse).

[0226] Fig. 20 19 is a diagram showing an example of carrier concentration distribution in the semiconductor substrate 10 after the diffusion step S1904. Fig. 20 The carrier concentration distribution can be obtained, for example, by spread resistance profiling. Figure 20 to Figure 26 In each of the figures, the lower surface 23 of the semiconductor substrate 10 is set to a reference position (0 μm) of the depth (μm). In addition, the first depth Z1 is 10 μm or less. The first depth Z1 can be set to 0 μm.

[0227] exist Fig. 20, three types of carrier concentration distributions of semiconductor substrates 10 are shown. The first example 161 and the second example 162 are examples of hydrogen ion implantation into the first depth Z1 and the second depth Z2, and the third example 163 is an example of hydrogen ion implantation only into the second depth Z2. In each example, the dose of hydrogen ion implantation into the second depth Z2 (referred to as the second dose) is set to 1×10 13 / cm 2 In addition, the range of hydrogen ion implantation to the second depth Z2 is set to 100 μm, and the acceleration energy is set to 3.1 MeV. The range of hydrogen ions can be adjusted by the acceleration energy, and can also be adjusted by an aluminum absorber or the like.

[0228] In the first example 161 and the second example 162, the acceleration energy of the hydrogen ions implanted into the first depth Z1 is set to 400 keV. In the first example 161, the first dose is set to 1×10 15 / cm 2 In the second example 162, the first dose is set to 3×10 14 / cm 2 .

[0229] After the hydrogen ion implantation, the semiconductor substrate 10 of each example was annealed at 370° C. for 5 hours in the same annealing furnace. Fig. 20 : is the carrier concentration distribution after annealing. In each example, crystal defects are formed in the through region 106 (the range from the lower surface 23 of the semiconductor substrate 10 to the second depth position Z2) before annealing. Therefore, the carrier concentration of the through region 106 is reduced.

[0230] After annealing, hydrogen combines with crystal defects to form donors, thereby increasing the carrier concentration. However, in the third example 163, since hydrogen is not injected into the first depth Z1, the carrier concentration does not increase substantially. As shown in the first example 161 and the second example 162, the more the first dose increases, the more the carrier concentration in the through region 106 increases.

[0231] It should be noted that in the second example 162, a valley of carrier concentration is generated at a position shallower than the second depth Z2 (near the depth of 80 μm). This is considered to be because when the hydrogen injected into the first depth Z1 diffuses in the depth direction, the concentration of the diffused hydrogen is insufficient, and hydrogen with a sufficient concentration does not diffuse to a position deeper than x2. On the other hand, in the first example 161, the carrier concentration is lower than the basic doping concentration (in this example, 1.0×10 14 / cm 3 This is considered to be because when the hydrogen implanted to the first depth Z1 diffuses in the depth direction, the concentration of the diffused hydrogen is sufficiently high, and a sufficient concentration of hydrogen diffuses to a position deeper than x1.

[0232] If the first dose is set to Q, the diffusion depth of hydrogen from the first depth Z1 is set to x(x1, x2) (cm), and the diffusion coefficient of hydrogen is set to D (cm 2 / s), the diffusion time is t(s), and the basic doping concentration of the semiconductor substrate 10 is C0 (atoms / cm 3 ), then their relationship is expressed by the following formula (1).

[0233] [Mathematical formula 1]

[0234]

[0235] Formula (1) is a value calculated from the solution of the diffusion equation. When the diffusion equation is solved with the boundary condition that the total amount of hydrogen is constant, the solution is Gaussian distributed. In the solution of the Gaussian distribution obtained, x when the concentration C(x, t) is consistent with the basic doping concentration C0 is Formula (1).

[0236] According to equation (1), the diffusion coefficient D of hydrogen in the semiconductor substrate 10 can be numerically calculated. The diffusion depth x in the second example 162 can be calculated according to Fig. 20 The diffusion depth x can be determined by the shape of the curve. For example, the diffusion depth x can be the distance from the first depth position Z1 to the initial inflection point of the valley portion of the carrier concentration. Alternatively, the diffusion depth x can be the distance from the first depth position Z1 to the position where the carrier concentration initially falls below the base doping concentration. Fig. 20 In the first example 161, Q=1×10 15 / cm 2 , t = 5h (= 1.8 × 10 3 s), x1=85μm(=8.5×10 -3 cm), C0=1×10 14 / cm 3 In the second example 162, Q = 3 × 10 14 / cm 2 , t = 5h (= 1.8 × 10 3 s), x2 = 75 μm (= 7.5 × 10 -3 cm), C0=1×10 14 / cm 3 .

[0237] It should be noted that the crystal defects formed when hydrogen ions are injected into the second depth Z2 are various defects such as point defects and dislocations. In particular, vacancy defects such as vacancies and compound vacancies in point defects are formed. At this time, the concentration of crystal defects has a peak at a position slightly closer to the ion injection surface (the lower surface 23 of the semiconductor substrate 10) than the second depth Z2. Since the hydrogen ions injected into the first depth Z1 are combined with the crystal defects, the diffusion depth x1 of hydrogen in the first example 161 becomes a value slightly smaller than the second depth Z2 (100μm) (85μm in this example).

[0238] Fig.21 is a graph showing the relationship between the diffusion coefficient D of hydrogen and the first dose Q. Fig.21 In the Fig. 20 The first example 161 and the second example 162 are shown. The diffusion coefficient D is relatively not dependent on the first dose Q. On the other hand, if the first dose Q increases, the diffusion coefficient D increases slightly. The hydrogen injected into the first depth Z1 will diffuse to the second depth Z2 through the dangling bond termination edge of the region 106. It is believed that if the first dose Q increases, the proportion of hydrogen diffused in the region where the dangling bonds are terminated increases, so hydrogen is easy to diffuse. It should be noted that the value of the diffusion coefficient D will change depending on the experimental conditions, etc. Fig.21 The diffusion coefficient D shown can be allowed to have an error within at least ±50%, and an error within ±100% can also be allowed.

[0239] Fig. 22 It is a graph showing the relationship between the diffusion coefficient D and the annealing temperature T. Fig. 22 is obtained for multiple annealing temperatures T Fig. 20 and Fig.21 In this example, the first dose is set to Q = 1 × 10 15 / cm 2 .

[0240] The diffusion coefficient D is represented by D = D0exp(-Ea / kBT). D0 is a constant, Ea is the activation energy, and kB is the Boltzmann constant. Fig. 22 The chart shows that D0 = 0.33237 (cm 2 / s), Ea=1.204 (eV). From this, the diffusion coefficient of hydrogen in the semiconductor substrate 10 can be calculated.

[0241] Fig.23 is a graph showing the relationship between the diffusion depth of hydrogen and the first dose. Fig.23 In the middle, the black circle is used to depict Fig. 20 The first example 161 and the second example 162 are shown. Fig. 20As described in the first example 161, the first dose is Q = 1 × 10 15 / cm 2 , the diffusion depth is x1=85μm. In addition, the first dose in the second example 162 is Q=3×10 14 / cm 2 , the diffusion depth is x2=75μm.

[0242] like Fig.23 As shown, by connecting the curves of the first example 161 and the second example 162 with a straight line, the first dose for each diffusion depth x can be determined. That is, the straight line represents the minimum first dose for each diffusion depth x. In the first injection step S1900, by setting a first dose greater than the straight line, the donor concentration of the entire through region 106 can be made greater than the basic doping concentration. As an example, when the second depth Z2 is set to the diffusion depth x (μm), the first dose Q (ions / cm 2 ) can satisfy the following formula.

[0243] Q≥2.6186×10 10 ×e 0.12412x

[0244] In this formula, it should be noted that the unit of x is (μm).

[0245] As described above, the peak of the crystal defect concentration when hydrogen ions are implanted into the second depth Z2 is located at a position slightly shallower than the second depth Z2. Fig.23 The horizontal axis corresponds to the peak position of the crystal defect concentration. Fig.23 In the case of the through zone 106 having a length X0 corresponding to the horizontal axis of , hydrogen ions are implanted to the second depth Z2 according to the range Rp of the following formula in consideration of the standard deviation (Straggling) ΔRp in ion implantation.

[0246] Rp≥X0+ΔRp

[0247] By setting the concentration peak position of the crystal defect at a position closer to the upper surface 21 than the bottom of the groove portion provided on the upper surface 21 of the semiconductor substrate 10 (a position at a length X0 from the lower surface 23 ), a through region 106 can be formed over substantially the entire depth direction of the semiconductor substrate 10 .

[0248] It should be noted that the minimum dose may also be calculated based on the following formula (2) which is a modification of the formula (1).

[0249] [Mathematical formula 2]

[0250]

[0251] The diffusion coefficient D is given by Fig. 22Calculated using the method described in . Fig.23 In FIG. 1 , the minimum dose calculated according to equation (2) is depicted by a white circle. It should be noted that since the diffusion coefficient D has the dimension of the square of the distance, the diffusion coefficient D can be expressed as a function of the diffusion depth x of hydrogen.

[0252] Fig.24 is a graph showing the relationship between the diffusion coefficient D and the diffusion depth x. Fig.24 In, depict Fig. 20 The first example is 161, and the second example is 162. Fig.24 As shown, if the diffusion depth x increases, the diffusion coefficient D also increases.

[0253] As the diffusion depth x increases, the distance from the first depth Z1 to the concentration peak of the crystal defects increases. Therefore, the proportion of the region with relatively few crystal defects in the through-zone 106 increases. If there are few crystal defects, the diffusion coefficient becomes larger, so as the diffusion depth x increases, the average diffusion coefficient of the through-zone 106 also becomes larger.

[0254] It should be noted that Figure 20 to Figure 24 The second dose was set to 1×10 13 / cm 2 . However, even if the second dose is changed, the minimum dose of the first dose can be determined by the same method. In addition, by adjusting the second dose, the donor concentration of the through region 106 can also be adjusted. By adjusting the second dose, the concentration of crystal defects formed in the through region 106 can be adjusted. In addition, although the annealing temperature is set to 370°C, even if the annealing temperature is changed, the minimum dose can be determined according to formula (2).

[0255] Fig.25 is a graph showing a straight line that specifies the minimum dose for each annealing temperature. In this example, the diffusion coefficient D is made constant regardless of the diffusion depth. In the first implantation step S1900, as long as the ratio is Fig.25 The minimum dose represented by the straight line can be used to implant hydrogen ions into the first depth Z1 with a dose larger than the minimum dose.

[0256] Fig.26 is a graph showing the relationship between the second dose and the minimum dose of the first dose. In this example, for each diffusion depth x, the relationship is shown. In the first implantation step S1900, as long as the ratio is Fig.26 The curve represents the minimum dose and a larger dose can be used to implant hydrogen ions into the first depth Z1.

[0257] Fig. 27 is a diagram illustrating an example of the first depth Z1. Fig. 27, the donor concentration distribution and the hydrogen chemical concentration distribution in the depth direction of the semiconductor substrate 10 are shown. The hydrogen chemical concentration distribution only schematically shows the vicinity of the peak. Fig. 27 In the figure, the distribution near the upper surface 21 of the semiconductor substrate 10 (the region 100 μm or more from the lower surface 23) is omitted. In addition, the carrier concentration distribution in the N-type region of the semiconductor substrate 10 may be used as the donor concentration distribution.

[0258] In this example, the first depth Z1 of the first hydrogen concentration peak 131 is included in the range of 5 μm or less in the depth direction from the lower surface 23 of the semiconductor substrate 10. In the semiconductor device 100, the configuration other than the first depth Z1 is the same as Figure 1 to Figure 26 The donor concentration distribution of the semiconductor device 100 of this example is the same as Fig.10 Same as the example.

[0259] By arranging the first depth Z1 near the lower surface 23, the distance between the first depth Z1 and the second depth Z2 can be extended. Therefore, the donor concentration can be adjusted with high precision in a wider range of the semiconductor substrate 10. The first depth Z1 may be within a range of 4 μm from the lower surface 23, or within a range of 3 μm.

[0260] In order to diffuse hydrogen into a wider range, it is preferred to further increase the dose of hydrogen implanted into the first depth Z1. In this example, the dose of hydrogen implanted into the first depth Z1 may be 1×10 15 atoms / cm 2 Above, it can also be 1×10 16 atoms / cm 2 Above, it can also be 1×10 17 atoms / cm 2 Above, it can also be 1×10 18 atoms / cm 2 At the first depth Z1, a first donor concentration peak 111 formed of hydrogen donors may be formed. The donor concentration of the first donor concentration peak 111 may be 1×10 15 / cm 3 Above, it can also be 1×10 16 / cm 3 The donor concentration of the first donor concentration peak 111 may be 1×10 17 / cm 3 the following.

[0261] The injection of hydrogen into the first depth Z1 can be performed by plasma doping. In plasma doping, a raw gas containing a gas for plasma excitation and hydrogen is supplied to a storage container containing the semiconductor substrate 10. The excitation gas may contain an inert element such as argon. The raw gas may use phosphine (PH3) or the like. By using these gases to generate plasma in the storage container, the lower surface 23 of the semiconductor substrate 10 is exposed to the plasma, so that a high concentration of hydrogen can be easily injected into a shallow position from the lower surface 23. In addition, by using plasma doping to inject hydrogen into a shallow position near the lower surface 23, the generation of crystal defects in the semiconductor substrate 10 can be suppressed. In addition, since there are fewer crystal defects, the annealing temperature can be reduced, and the yield in the manufacture of the semiconductor device 100 can be improved. However, the method of injecting hydrogen into the first depth Z1 is not limited to plasma doping.

[0262] The injection of hydrogen into the second depth Z2 of the second hydrogen concentration peak 141 may be performed by a method other than plasma doping. Hydrogen ions may be accelerated by an electric field or the like to inject hydrogen into the second depth Z2. The second depth position Z2 may be separated from the lower surface 23 by more than 80 μm in the depth direction. The second depth position Z2 may be separated from the lower surface 23 by more than 90 μm, or by more than 100 μm. The distance in the depth direction between the first depth position and the second depth position may be more than 50% of the thickness of the semiconductor substrate 10 in the depth direction, or more than 65%, or more than 80%.

[0263] Fig.28 Another example of the donor concentration distribution and hydrogen chemical concentration distribution in the depth direction of the semiconductor substrate 10 is shown. Fig. 27 The second depth Z2 may be arranged on the upper surface 21 side of the semiconductor substrate 10. The upper surface 21 side refers to a region from the center of the semiconductor substrate 10 in the depth direction to the upper surface 21.

[0264] exist Fig. 27 and Fig.28 In any example, the hydrogen chemical concentration distribution may have one or more hydrogen concentration peaks 194 between the first depth Z1 and the second depth Z2. The hydrogen concentration peaks 194 may be arranged at Figure 6 The first hydrogen concentration peak 131 may be disposed in the buffer region 20 , or may be disposed between the buffer region 20 and the lower surface 23 .

[0265] Fig.291 is a diagram showing an example of the hydrogen chemical concentration distribution and the argon chemical concentration distribution near the first hydrogen concentration peak 131. In this example, the first hydrogen concentration peak 131 is a peak corresponding to hydrogen injected by plasma doping, and the hydrogen concentration peak 194 is a peak corresponding to hydrogen injected by a method other than plasma doping.

[0266] If hydrogen is injected into the first depth position Z1 by plasma doping, impurities other than hydrogen may be injected near the first depth position Z1. For example, when argon gas is used in plasma excitation, argon may be injected near the first depth position Z1. Fig.29 In the graph, an argon concentration peak 196 appears at a depth position Z0.

[0267] The depth position Z0 may be disposed between the lower surface 23 and the first depth position Z1. Since argon is an element heavier than hydrogen, the argon concentration peak 196 is likely to be formed at a position shallower than the first hydrogen concentration peak 131.

[0268] In this example, no peak of argon chemical concentration is provided between the first depth position Z1 and the second depth position Z2. Since the hydrogen concentration peak 194 at the second depth position Z2 is formed by a method other than plasma doping, argon is not injected near the hydrogen concentration peak 194. The argon chemical concentration between the first depth position Z1 and the second depth position Z2 is smaller than the argon concentration peak 196. The maximum value of the argon chemical concentration between the first depth position Z1 and the second depth position Z2 may be less than the minimum value of the argon chemical concentration between the lower surface 23 and the first depth position Z1.

[0269] It should be noted that, depending on the composition of the gas used in plasma doping, other impurities may be injected into the semiconductor substrate 10 instead of argon. When PH3 gas is used in plasma doping, a phosphorus concentration peak may be set between the lower surface 23 and the first depth position Z1. When BF3 gas is used in plasma doping, a fluorine concentration peak may be set between the lower surface 23 and the first depth position Z1, or a boron concentration peak may be set. The concentration value of the concentration peak of argon, phosphorus, fluorine or boron may be less than the concentration value of the first hydrogen concentration peak 131. The concentration value of the concentration peak of argon, phosphorus, fluorine or boron may be less than half of the concentration value of the first hydrogen concentration peak 131, or may be less than 1 / 10.

[0270] Fig.30 1 is a diagram showing another structural example of the semiconductor device 100. The semiconductor device 100 of this example is Fig.11 The example shown in FIG. 1 also includes a transistor unit 70 and a diode unit 80. The structure of the transistor unit 70 is similar to that of FIG. Figure 6 The transistor portion 70 and the diode portion 80 are provided adjacent to each other in the X-axis direction.

[0271] The diode portion 80 of this example differs from the transistor portion 70 in that the diode portion 80 has a dummy trench portion 30 instead of the gate trench portion 40, a cathode region 82 instead of the collector region 22, and no emitter region 12. The other structures are the same as those of the transistor portion 70.

[0272] The dummy groove portion 30 may have the same structure as the gate groove portion 40. The dummy groove portion 30 includes a dummy insulating film 32 and a dummy conductive portion 34. The dummy insulating film 32 and the dummy conductive portion 34 may have the same structure and material as the gate insulating film 42 and the gate conductive portion 44. However, while the gate conductive portion 44 is electrically connected to the gate electrode, the dummy conductive portion 34 is electrically connected to the emitter 52. It should be noted that the dummy groove portion 30 may also be provided in the transistor portion 70. That is, a portion of the gate groove portion 40 in the transistor portion 70 may be replaced by the dummy groove portion 30.

[0273] The cathode region 82 is exposed on the lower surface 23 of the semiconductor substrate 10 similarly to the collector region 22. The cathode region 82 is connected to the collector electrode 54 on the lower surface 23. The cathode region 82 is an N+ type region doped with N type impurities such as phosphorus. A buffer region 20 may be provided between the cathode region 82 and the drift region 18.

[0274] In addition, in the diode portion 80, the base region 14 may be exposed on the upper surface 21. The base region 14 of the diode portion 80 is electrically connected to the emitter electrode 52. With such a configuration, the diode portion 80 functions as a diode.

[0275] In this example, hydrogen is also implanted at the first depth position Z1 and the second depth position Z2 in the diode section 80. In addition, the diode section 80 also has a through region similar to that of the transistor section 70. The concentration distributions in the transistor section 70 may be the same as those in the transistor section 70. Figure 1 to Figure 29 The chemical hydrogen concentration distribution of the diode portion 80 in the depth direction may be the same as the chemical hydrogen concentration distribution of the transistor portion 70 in the depth direction.

[0276] Fig.31 express Fig.30 An example of carrier concentration distribution, hydrogen chemical concentration distribution, and boron chemical concentration distribution at the DD line of the embodiment. The DD line passes through the collector region 22 and a portion of the buffer region 20 in the transistor portion 70. The collector region 22 of this embodiment can be formed by implanting boron. The boron of the collector region 22 of this embodiment is implanted in a different process from the hydrogen of the first hydrogen concentration peak 131. At least a portion of the boron of the collector region 22 can be implanted in the plasma doping used to implant the hydrogen of the first hydrogen concentration peak 131.

[0277] exist Figure 7In the example of , the first hydrogen concentration peak 131 is arranged in the buffer region 20. The first hydrogen concentration peak 131 in this example is arranged in the cathode region 82 and the collector region 22. Since the doping concentrations of the cathode region 82 and the collector region 22 are very high, by setting the first hydrogen concentration peak 131 in the cathode region 82 and the collector region 22, even when a high concentration of hydrogen donors is generated by the first hydrogen concentration peak 131, the change in the shape of the carrier concentration distribution can be suppressed. Therefore, it is easy to suppress the influence on the characteristics of the semiconductor device 100.

[0278] The concentration of the first hydrogen concentration peak 131 is set such that, at the first depth position Z1, the hydrogen donor concentration is much smaller than the carrier concentration. The activation rate of hydrogen is about 1%. At the first depth position Z1, 1% of the hydrogen chemical concentration may be smaller than the boron chemical concentration.

[0279] In addition, the peak position of the carrier concentration distribution in the collector region 22 is arranged closer to the peak of the boron chemical concentration than the first hydrogen concentration peak 131. Fig.31 In the example, the peak of the boron chemical concentration is arranged at the lower surface 23. The peak position of the carrier concentration distribution in the collector region 22 can be the same as the peak position of the boron chemical concentration. The first hydrogen concentration peak 131 can be arranged between the peak of the carrier concentration distribution in the collector region 22 and the buffer region 20. In this example, the peak position of the carrier concentration distribution in the collector region 22 is consistent with the lower surface 23. It should be noted that in the buffer region 20, the depth position of the hydrogen concentration peak 194 can be consistent with the depth position of the peak 24 of the carrier concentration distribution.

[0280] Fig.32 express Fig.30 FIG. 1 is an example of carrier concentration distribution, hydrogen chemical concentration distribution, and phosphorus chemical concentration distribution at the EE line of FIG. 1 . The EE line passes through the cathode region 82 and a portion of the buffer region 20 in the diode portion 80. The cathode region 82 of this example is formed by implanting phosphorus. The phosphorus in the cathode region 82 is implanted in a different process from the hydrogen in the first hydrogen concentration peak 131. At least a portion of the phosphorus in the cathode region 82 can be implanted in the plasma doping used to implant the hydrogen in the first hydrogen concentration peak 131.

[0281] The first hydrogen concentration peak 131 in this example is configured in the cathode region 82 and the collector region 22. The concentration of the first hydrogen concentration peak 131 is set such that at the first depth Z1, the hydrogen donor concentration is much smaller than the carrier concentration. The activation rate of hydrogen is about 1%. At the first depth Z1, 1% of the hydrogen chemical concentration can be smaller than the phosphorus chemical concentration.

[0282] In addition, the peak position of the carrier concentration distribution in the cathode region 82 can be arranged closer to the peak of the phosphorus chemical concentration than the first hydrogen concentration peak 131. Fig.32In the example, the peak of the phosphorus chemical concentration is arranged at the lower surface 23. The peak position of the carrier concentration distribution in the cathode region 82 can be the same as the peak position of the phosphorus chemical concentration. The first hydrogen concentration peak 131 can be arranged between the peak of the carrier concentration distribution in the cathode region 82 and the buffer region 20. In this example, the peak position of the carrier concentration distribution in the cathode region 82 is consistent with the lower surface 23. It should be noted that in the buffer region 20, the depth position of the hydrogen concentration peak 194 can be consistent with the depth position of the peak 24 of the carrier concentration distribution.

[0283] Fig.33 FIG. 1 is a diagram showing a part of the steps of a method for manufacturing the semiconductor device 100. Fig.33 Prior to the steps shown, structures on the upper surface 21 side, such as the groove portions, the emitter region 12 , the base region 14 , and the accumulation region 16 , may be formed.

[0284] In this example, in the implantation step S3300, hydrogen ions are implanted from the lower surface 23 of the semiconductor substrate 10 to the second depth Z2. The implantation step S3300 may be performed in the same manner as described above. Fig.19 The same as the second injection step S1902 in the example.

[0285] In addition, in the implantation step S3302, hydrogen ions are implanted from the lower surface 23 of the semiconductor substrate 10 to the first depth Z1. In the implantation step S3302, hydrogen ions are implanted by plasma doping. The dose of hydrogen in the implantation step S3302 may be Fig.19 The dosage of hydrogen in the first implantation step S1900 of the example is the same. Either the implantation step S3300 or the implantation step S3302 may be performed first.

[0286] After the injection step S3300 and the injection step S3302, a diffusion step S3304 is performed. Fig.19 The diffusion step S1904 in the example of FIG. 3304 is the same. By diffusing hydrogen in the diffusion step S3304, the crystal defects in the through region 106 are combined with hydrogen to become donors. Thus, the donor concentration in the through region 106 can be increased.

[0287] After the diffusion step S3304, a grinding step S3306 is performed. In the grinding step S3306, the lower surface 23 side of the semiconductor substrate 10 is ground by chemical mechanical polishing (CMP) or the like. In the grinding step S3306, grinding can be performed in a range shallower than the first depth Z1, or in a range deeper than the first depth Z1. Thus, since grinding can be performed on a region where hydrogen is distributed at a high concentration, the amount of hydrogen near the lower surface 23 can be reduced.

[0288] After the grinding step S3306, in the lower surface side structure forming step S3308, structures on the lower surface 23 side such as the collector region 22, the cathode region 82 and the buffer region 20 are formed. In the lower surface side structure forming step S3308, after dopants are injected into the cathode region 82 and the buffer region 20, laser annealing can be performed near the lower surface 23. Thus, the vicinity of the lower surface 23 of the semiconductor substrate 10 can be locally heat-treated at a high temperature. In addition, after the laser annealing, dopants such as hydrogen can be injected into the buffer region 20. After the dopants are injected into the buffer region 20, the entire semiconductor substrate 10 can be heat-treated using an annealing furnace.

[0289] Fig.34 The diagram shows a part of the steps of the method for manufacturing the semiconductor device 100. The manufacturing method of this example includes a laser annealing step S3307 instead of Fig.33 The other steps are different from Fig.33 Same as the example.

[0290] In the laser annealing step S3307, laser annealing is performed on the lower surface 23 of the semiconductor substrate 10. In the laser annealing step S3307, laser can be irradiated near the first depth Z1. As a result, at least a portion of the hydrogen near the first depth Z1 can be released to the outside of the semiconductor substrate 10. As a result, the hydrogen chemical concentration near the first depth Z1 can be reduced. In the laser annealing step S3307, the laser can be irradiated in a manner that the first hydrogen concentration peak 131 remains, or in a manner that the first hydrogen concentration peak 131 does not remain. It should be noted that even if laser irradiation is performed, heavy elements such as argon are more likely to remain on the semiconductor substrate 10 than hydrogen. Therefore, even if the laser annealing step S3307 is performed, sometimes there is a hydrogen concentration peak 131 on the semiconductor substrate 10. Fig.29 The concentration peaks of impurities such as argon are shown.

[0291] In addition, in the laser annealing step S3307, the laser may be irradiated to the transistor portion 70, but not to the diode portion 80. Even if a high concentration of hydrogen donors remains on the lower surface 23 of the diode portion 80, the influence on the characteristics is relatively small. At this time, the hydrogen chemical concentration at the first depth Z1 of the diode portion 80 is higher than the hydrogen chemical concentration at the first depth Z1 of the transistor portion 70.

[0292] exist Fig.19 , Fig.33 , Fig.34 In the example shown, the hydrogen ion implantation and heat treatment to the first depth Z1 and the second depth Z2 may be performed before, after, or during the formation of the lower surface side structure.

[0293] Fig.35 An example of a process of implanting hydrogen ions into the first depth Z1 and the second depth Z2 in the lower surface side structure forming step is shown. In this example, the implantation of hydrogen ions into the first depth Z1 and the second depth Z2 and the heat treatment are performed in the buffer zone forming step S3504 of forming the buffer zone 20.

[0294] The lower surface side structure forming step can be performed after the upper surface side structure forming step S3500 such as the groove portion. The lower surface side structure forming step includes a collector region forming step S3502 and a buffer region forming step S3504. Fig.35 In the step of forming the lower surface side structure, other steps are omitted. In this example, in the buffer zone forming step S3504, hydrogen ions are injected into a plurality of locations including the first depth Z1 and the second depth Z2. In S3504, after the hydrogen ions are injected, the semiconductor substrate 10 is heat treated to diffuse the hydrogen ions. After the lower surface side structure forming step, the collector forming step S3506 can be performed.

[0295] Fig.36 Another example of the process of injecting hydrogen ions into the first depth Z1 and the second depth Z2 in the lower surface side structure forming step is shown. In this example, the injection of hydrogen ions into the first depth Z1 and the heat treatment are performed in the cathode region forming step S3503. In addition, the injection of hydrogen ions into the second depth Z2 and the heat treatment are performed in the buffer region forming step S3506.

[0296] The lower surface side structure forming step can be performed after the upper surface side structure forming step S3500 such as the groove portion. The lower surface side structure forming step of this example includes the cathode region forming step S3503 and the buffer region forming step S3504. Fig.36 In the process, other steps of the lower surface side structure formation step are omitted.

[0297] The semiconductor device 100 of this example may include a transistor portion 70 and a diode portion 80. At this time, after forming the cathode region 82 on the entire lower surface 23, a P-type dopant is selectively injected, so that a P-type collector region 22 can be formed in a part of the cathode region 82. In the cathode region forming step S3503, a raw material gas such as PH3 containing an N-type dopant such as phosphorus can be used. In the cathode region forming step S3503, hydrogen ions are injected into the first depth Z1 on the entire lower surface 23.

[0298] In addition, the collector region 22 may be selectively formed on the lower surface 23, and then N-type dopants and hydrogen ions may be injected into the entire lower surface 23 to form the cathode region 82. At this time, a high concentration of P-type dopants may be injected into the collector region 22 in advance so as not to invert the conductivity type to the N-type. Through such a process, the first depth Z1 is configured in the collector region 22 and the cathode region 82.

[0299] It should be noted that when the semiconductor device 100 does not include the transistor portion 70 but includes the diode portion 80, the step of forming the collector region 22 can be omitted. In addition, hydrogen injection into the second depth Z2 can be performed in the step of forming the buffer region 20. After the lower surface side structure formation step, the collector formation step S3506 can be performed.

[0300] The present invention has been described above using the embodiments, but the technical scope of the present invention is not limited to the scope described in the above embodiments. It is clear to those skilled in the art that various changes or improvements can be made to the above embodiments. According to the description of the claims, it can be known that the various changes or improvements obtained can also be included in the technical scope of the present invention.

[0301] It should be noted that as long as the execution order of each process such as work, sequence, step and stage in the device, system, program and method shown in the claims, description and drawings is not specifically specified as "before...", "prior to...", etc., and as long as the result of the previous process is not used in the subsequent process, it can be implemented in any order. Even if the workflow in the claims, description and drawings is described using "first", "next", etc. for convenience, it does not mean that it must be executed in this order.

Claims

1. A semiconductor device, characterized in that: A semiconductor substrate is provided, wherein the semiconductor substrate has an upper surface and a lower surface, In a depth direction connecting the upper surface and the lower surface of the semiconductor substrate, a hydrogen concentration distribution has a first hydrogen concentration peak and a second hydrogen concentration peak, and a donor concentration distribution has a first donor concentration peak and a second donor concentration peak, The first hydrogen concentration peak and the first donor concentration peak are arranged at a first depth, the second hydrogen concentration peak and the second donor concentration peak are arranged at a second depth, and the second depth is deeper than the first depth when the lower surface is used as a reference, Each concentration peak has an upward slope with a concentration value increasing from the lower surface toward the upper surface, A value obtained by normalizing the slope of the upward slope of the second donor concentration peak with the slope of the upward slope of the second hydrogen concentration peak is smaller than a value obtained by normalizing the slope of the upward slope of the first donor concentration peak with the slope of the upward slope of the first hydrogen concentration peak.

2. The semiconductor device according to claim 1, wherein: Each concentration peak has a downward slope in which the concentration value decreases from the lower surface toward the upper surface, The upward slope of the second hydrogen concentration peak has a smaller slope than the downward slope.

3. The semiconductor device according to claim 1, wherein: Each concentration peak has a downward slope in which the concentration value decreases from the lower surface toward the upper surface, The upward slope of the second donor concentration peak has a smaller slope than the downward slope.

4. The semiconductor device according to any one of claims 1 to 3, characterized in that The donor concentration distribution between the first depth and the second depth has a flat region with a substantially constant donor concentration, A length of the flat region in the depth direction is equal to or greater than 10% of a thickness of the semiconductor substrate in the depth direction.

5. The semiconductor device according to any one of claims 1 to 3, characterized in that The donor concentration distribution between the first depth and the second depth has a flat region with a substantially constant donor concentration, The length of the flat region in the depth direction is 10 μm or more.

6. The semiconductor device according to claim 4, wherein: A minimum value of the donor concentration of the flat region is higher than a donor concentration of the semiconductor substrate.

7. The semiconductor device according to claim 5, wherein: A minimum value of the donor concentration of the flat region is higher than a donor concentration of the semiconductor substrate.

8. The semiconductor device according to claim 6, wherein: A minimum value of the donor concentration between the first depth and the second depth is higher than a donor concentration of the semiconductor substrate.

9. The semiconductor device according to claim 7, wherein: A minimum value of the donor concentration between the first depth and the second depth is higher than a donor concentration of the semiconductor substrate.

10. The semiconductor device according to any one of claims 1 to 3, wherein: The concentration value of the second hydrogen concentration peak is smaller than the concentration value of the first hydrogen concentration peak.

11. The semiconductor device according to any one of claims 1 to 3, wherein: The semiconductor device further comprises: An N-type drift region, which is disposed on the semiconductor substrate; an emitter region, which is provided in the semiconductor substrate in contact with the upper surface and has a donor concentration higher than that of the drift region; A P-type base region, which is arranged between the emitter region and the drift region; A P-type collector region, which is arranged in the semiconductor substrate in contact with the lower surface; An N-type buffer region is disposed between the collector region and the drift region and has one or more donor concentration peaks having a higher donor concentration than that of the drift region. The first donor concentration peak is the donor concentration peak in the buffer zone.

12. The semiconductor device according to claim 11, wherein: The semiconductor device further includes an accumulation region, the accumulation region being provided between the base region and the drift region and having one or more donor concentration peaks having a higher donor concentration than the donor concentration of the drift region. The second donor concentration peak is the donor concentration peak of the accumulation region.

13. The semiconductor device according to claim 12, wherein: The accumulation region has the donor concentration peak formed of a donor other than hydrogen in addition to the second donor concentration peak.

14. The semiconductor device according to claim 11, wherein: The semiconductor device further includes an accumulation region, the accumulation region being provided between the base region and the drift region and having one or more donor concentration peaks having a higher donor concentration than the donor concentration of the drift region. The second donor concentration peak is arranged between the buffer area and the accumulation area.

15. The semiconductor device according to any one of claims 1 to 3, wherein: The semiconductor device further includes a gate trench portion, wherein the gate trench portion is provided on the upper surface of the semiconductor substrate. The second donor concentration peak is arranged between the bottom of the gate trench portion and the upper surface of the semiconductor substrate.

16. The semiconductor device according to any one of claims 1 to 3, wherein: The semiconductor device further comprises: an active portion disposed on the semiconductor substrate; and an edge terminal structure portion, which is arranged in a manner of surrounding the active portion when the semiconductor substrate is viewed from above, The semiconductor substrate has a passage region through which hydrogen injected into the position of the second hydrogen concentration peak passes, In the depth direction, the passing region provided in the edge termination structure portion is shorter than the passing region provided in the active portion, or the passing region is not provided in the edge termination structure portion.

17. The semiconductor device according to any one of claims 1 to 3, characterized in that The semiconductor device further includes a transistor portion and a diode portion provided on the semiconductor substrate. The semiconductor substrate has a passage region through which hydrogen injected into the position of the second hydrogen concentration peak passes, In the depth direction, the passing region provided in the diode portion is shorter than the passing region provided in the transistor portion, or the passing region is not provided in the diode portion.

18. The semiconductor device according to any one of claims 1 to 3, wherein: The semiconductor device further includes a transistor portion and a diode portion provided on the semiconductor substrate. The semiconductor substrate has a passage region through which hydrogen injected into the position of the second hydrogen concentration peak passes, In the depth direction, the through region provided in the transistor portion is shorter than the through region provided in the diode portion, or the through region is not provided in the transistor portion.

19. The semiconductor device according to any one of claims 1 to 3, wherein: The first depth is included in a range of 5 μm or less from the lower surface along the depth direction.

20. The semiconductor device according to claim 19, wherein: The donor concentration at the first hydrogen concentration peak is 1×10 15 / cm 3 Above and 1×10 17 / cm 3 the following.

21. A semiconductor device, characterized in that: A semiconductor substrate is provided, wherein the semiconductor substrate has an upper surface and a lower surface, The hydrogen concentration distribution in the depth direction connecting the upper surface and the lower surface of the semiconductor substrate has a first hydrogen concentration peak and a second hydrogen concentration peak, the first hydrogen concentration peak is arranged in a range from the lower surface along the depth direction to less than 5 μm, and the second hydrogen concentration peak is arranged in a range closer to the upper surface than the first hydrogen concentration peak and separated from the lower surface by more than 80 μm in the depth direction.

22. The semiconductor device according to claim 21, wherein: The semiconductor substrate has an impurity concentration peak between the lower surface and the first hydrogen concentration peak, and an impurity of the impurity concentration peak is argon or fluorine.

23. A manufacturing method, characterized in that: A method for manufacturing a semiconductor device according to any one of claims 1 to 20, wherein the semiconductor device comprises a semiconductor substrate having an upper surface and a lower surface, comprising: A first implantation step of implanting hydrogen from the lower surface of the semiconductor substrate to a first depth; a second implantation step of implanting hydrogen from the lower surface of the semiconductor substrate to a second depth to form a passing region through which the hydrogen passes; and a diffusion step of heat-treating the semiconductor substrate to diffuse the hydrogen implanted to the first depth into the through region; determining a dose of hydrogen in the first implantation step so that, in the semiconductor substrate subjected to the heat treatment in the diffusion step, a minimum value of the donor concentration in the through region becomes higher than the donor concentration in the semiconductor substrate before the hydrogen implantation, In a depth direction connecting the upper surface and the lower surface of the semiconductor substrate, a hydrogen concentration distribution has a first hydrogen concentration peak and a second hydrogen concentration peak, and a donor concentration distribution has a first donor concentration peak and a second donor concentration peak, The first hydrogen concentration peak and the first donor concentration peak are arranged at the first depth, and the second hydrogen concentration peak and the second donor concentration peak are arranged at the second depth, and the second depth is deeper than the first depth when the lower surface is used as a reference.

24. The manufacturing method according to claim 23, characterized in that: In the first implantation step, hydrogen is implanted at a dose greater than a minimum dose, the minimum dose being determined based on a diffusion coefficient of hydrogen in the semiconductor substrate and the second depth.

25. The manufacturing method according to claim 23 or 24, characterized in that: The semiconductor substrate is a silicon substrate. When the second depth from the lower surface is set to x, the hydrogen dose Q in the first implantation step satisfies the following formula: 2 , Q≥2.6186×10 10 ×e 0.12412x 。 26. The manufacturing method according to claim 23 or 24, characterized in that: In the first implantation step, hydrogen is implanted into the first depth by plasma doping.

27. The manufacturing method according to claim 26, characterized in that: After the plasma doping, the lower surface of the semiconductor substrate is ground.

28. The manufacturing method according to claim 26, characterized in that: After the plasma doping, laser annealing is performed on the lower surface of the semiconductor substrate.

Images