Semiconductor device and manufacturing method of a semiconductor device

Abstract

Semiconductor device (100) comprising: a semiconductor substrate (10) having an upper surface (21) and a lower surface (23) and containing oxygen; a first peak (133) of a chemical hydrogen concentration located on the lower surface side of the semiconductor substrate (10); and a flat portion (150) disposed on the upper surface side of the semiconductor substrate (10) with respect to the first peak (133), which contains a hydrogen donor and has a substantially flat distribution of the donor concentration in the depth direction of the semiconductor substrate (10), wherein an oxygen contribution fraction, which indicates a fraction of a chemical oxygen concentration that contributes to generating the hydrogen donor in the chemical oxygen concentration of the oxygen, in the range of 1 × 10 -5 up to 7 × 10 -4 lies, a concentration of oxygen contributing to the generation of the hydrogen donor in the flat section (150) is lower than the chemical hydrogen concentration, and a hydrogen donor concentration in the flat section (150) in the range of 2 × 10 12 / cm3 to 5 × 10 14 / cm 3 lies.

Description

TECHNICAL BACKGROUND1. TECHNICAL FIELD

[0001] The present invention relates to a semiconductor device and a manufacturing method. 2. STATE OF THE ART

[0002] Conventionally, a technique for forming a donor is known in which hydrogen is implanted into a predetermined depth of a semiconductor substrate and the hydrogen is diffused to adjust the substrate resistance (e.g., document US 2018 / 0 019 306 A1 TASKS TO BE SOLVED

[0003] The donor concentration of the semiconductor substrate is preferably adjusted with high accuracy. GENERAL REVELATION

[0004] To achieve the above object, a first aspect of the present invention relates to a semiconductor device comprising a semiconductor substrate having an upper surface and a lower surface containing oxygen. The semiconductor device may have a first peak of a chemical hydrogen concentration located on the lower surface side of the semiconductor substrate. The semiconductor device may have a flat portion located on the upper surface side of the semiconductor substrate with respect to the first peak, which contains a hydrogen donor, and which has a substantially flat distribution of the donor concentration in the depth direction of the semiconductor substrate.An oxygen contribution fraction, which indicates a proportion of a chemical oxygen concentration that contributes to generating the hydrogen donor in the chemical oxygen concentration of the oxygen, may range from 1 × 10-5 to 7 × 10-4. A concentration of oxygen that contributes to generating the hydrogen donor in the flat section may be lower than the chemical hydrogen concentration. A hydrogen donor concentration in the flat section may range from 1 × 10- 12 / cm 3 up to 5 × 10 14 / cm 3 lay.

[0005] The oxygen contribution can be 5 × 10 -4 or less.

[0006] The oxygen contribution can be 1 × 10 -4 or more.

[0007] The semiconductor substrate may contain a volume donor.

[0008] A donor concentration of the flat section may be higher than a volume donor concentration.

[0009] The semiconductor device may have a second peak of a chemical concentration of hydrogen or helium located on the upper surface side of the semiconductor substrate. The flat portion may be located closer to the lower surface side of the semiconductor substrate with respect to the second peak. A hydrogen fraction, which indicates a ratio of the chemical hydrogen concentration to the generation of the hydrogen donor in the chemical hydrogen concentration, may range from 0.001 to 0.3. A vacancy concentration in the flat portion may range from 1 × 10 11 / cm 3 up to 1 × 10 14 / cm 3 lay.

[0010] The chemical hydrogen concentration of the first peak may be higher than the chemical hydrogen concentration of the second peak.

[0011] A chemical oxygen concentration in the shallow section can be 1 × 10 17 atoms / cm 3 or more.

[0012] A chemical carbon concentration in the flat section can be in the range of 1 × 10 13 / cm 3 up to 1 × 10 16 / cm 3 lay.

[0013] A sum of a value obtained by multiplying the chemical oxygen concentration in the flat section by the oxygen contribution fraction and a vacancy concentration of the flat section can be defined as a first value of the hydrogen donor concentration.

[0014] A difference resulting from subtracting the volume donor concentration from the donor concentration of the flat section can be defined as a second value of the hydrogen donor concentration. A ratio of the first value of the hydrogen donor concentration to the second value of the hydrogen donor concentration can range from 0.1 to 10.

[0015] A second aspect of the present invention provides a semiconductor device comprising a semiconductor substrate having an upper surface and a lower surface containing oxygen. The semiconductor device may have a first peak of a chemical hydrogen concentration located on the lower surface side of the semiconductor substrate. The semiconductor device may have a flat portion located on the upper surface side of the semiconductor substrate with respect to the first peak, which contains a hydrogen donor and has a substantially flat distribution of the donor concentration in the depth direction of the semiconductor substrate. The semiconductor substrate may contain a bulk donor. A donor concentration of the flat portion may be higher than a bulk donor concentration.A sum of a value obtained by multiplying a chemical oxygen concentration in the flat portion by an oxygen contribution fraction indicating a fraction of a chemical oxygen concentration that contributes to generating the hydrogen donor in the chemical oxygen concentration of the oxygen, and a vacancy concentration of the flat portion can be defined as a first value of the hydrogen donor concentration. A difference obtained by subtracting the volume donor concentration from the donor concentration of the flat portion can be defined as a second value of the hydrogen donor concentration. A ratio of the first value of the hydrogen donor concentration to the second value of the hydrogen donor concentration can be in the range of 0.1 to 10.

[0016] A third aspect of the present invention provides a semiconductor device comprising a semiconductor substrate having an upper surface and a lower surface containing oxygen and carbon. The semiconductor device may have a first peak of a chemical hydrogen concentration located on the lower surface side of the semiconductor substrate. The semiconductor device may have a flat portion located on the upper surface side of the semiconductor substrate with respect to the first peak, which contains a hydrogen donor and has a substantially flat distribution of the donor concentration in the depth direction of the semiconductor substrate. The semiconductor substrate may contain a bulk donor. A donor concentration of the flat portion may be higher than a bulk donor concentration.A sum of a value obtained by multiplying a chemical oxygen concentration in the flat portion by an oxygen contribution fraction indicating a fraction of a chemical oxygen concentration contributing to generating the hydrogen donor in the chemical oxygen concentration of the oxygen, a value obtained by multiplying a chemical carbon concentration in the flat portion by a carbon contribution fraction indicating a fraction of a chemical carbon concentration contributing to generating the hydrogen donor in the chemical carbon concentration of the carbon, and a vacancy concentration of the flat portion can be defined as a third value of a hydrogen donor concentration.A difference resulting from subtracting the volume donor concentration from the donor concentration of the flat section can be defined as a second value of the hydrogen donor concentration. A ratio of the third value of the hydrogen donor concentration to the second value of the hydrogen donor concentration can range from 0.1 to 10.

[0017] A fourth aspect of the present invention relates to a manufacturing method of a semiconductor device. The manufacturing method may include a step of measuring a chemical oxygen concentration of a semiconductor substrate having an upper surface and a lower surface. The manufacturing method may include a step of implanting a charged particle beam from the lower surface of the semiconductor substrate so as to pass through half or more of a thickness of the semiconductor substrate in the depth direction. The manufacturing method may include a step of performing heat treatment of the semiconductor substrate after implanting the charged particle beam. An implantation condition of the charged particle beam in implanting the charged particle beam and / or a heat treatment condition in performing the heat treatment may be adjusted according to the chemical oxygen concentration.

[0018] During measurement, the chemical carbon concentration of the semiconductor substrate can also be measured. During charged particle beam implantation, the charged particle beam implantation conditions can be adjusted according to the chemical oxygen concentration and the chemical carbon concentration.

[0019] At a given depth position of the semiconductor substrate, if a concentration of the hydrogen donor to be generated by the manufacturing process is N VOH1 is a concentration of the actually produced hydrogen donor N VOH2 is a vacancy concentration formed by implanting the charged particle beam, N V is a chemical oxygen concentration C OX is a chemical carbon concentration C Cis an oxygen contribution fraction indicating a fraction of the chemical oxygen concentration that contributes to generating the hydrogen donor in the chemical oxygen concentration, ξ, and a carbon contribution fraction indicating a fraction of the chemical carbon concentration that contributes to generating the hydrogen donor in the chemical carbon concentration, η, the following relationship can be satisfied: NVOH1=NV+ξCOX+ηCc; and 0.1≤NVOH1 / NVOH2≤10.

[0020] During implantation, hydrogen ions can be implanted as the beam of charged particles.

[0021] The manufacturing method may include a step of implanting hydrogen ions into the lower surface side of the semiconductor substrate before performing the heat treatment.

[0022] The manufacturing method may include a step of introducing oxygen into the semiconductor substrate.

[0023] The implantation condition of hydrogen ions when implanting hydrogen can be adjusted based on the chemical oxygen concentration of the semiconductor substrate.

[0024] The implantation condition of hydrogen ions when implanting hydrogen and the heat treatment condition when performing the heat treatment can be adjusted based on an implantation depth of the charged particle beam.

[0025] Introduction of oxygen into the semiconductor substrate may be included.

[0026] The implantation condition of the charged particle beam when implanting the charged particle beam and / or the heat treatment condition when performing the heat treatment can be adjusted based on the volume donor concentration of the semiconductor substrate.

[0027] The manufacturing method may include a step of grinding the semiconductor substrate.

[0028] The manufacturing method may include a step of measuring the thickness of the semiconductor substrate after grinding. The charged particle beam implantation condition for implanting the charged particle beam and / or the heat treatment condition for performing the heat treatment may be adjusted based on the thickness of the semiconductor substrate.

[0029] When implanting the charged particle beam, the implantation condition can be adjusted for each of a variety of semiconductor substrates.

[0030] When performing the heat treatment, the heat treatment condition can be adjusted commonly for the variety of semiconductor substrates.

[0031] The implantation condition of hydrogen ions when implanting hydrogen can be adjusted based on the thickness of the semiconductor substrate.

[0032] When measuring the thickness of the semiconductor substrate, the thickness of the semiconductor substrate can be measured in an edge termination structure portion of the semiconductor substrate.

[0033] A fifth aspect of the present invention relates to a manufacturing method of a semiconductor device. The manufacturing method may include a step of detecting an impurity concentration of a semiconductor substrate having an upper surface and a lower surface. The manufacturing method may include a step of implanting a charged particle beam from the lower surface of the semiconductor substrate so that they pass through half or more of a thickness of the semiconductor substrate in the depth direction. The manufacturing method may include a step of performing heat treatment of the semiconductor substrate after implanting the charged particle beam. An implantation depth of the charged particle beam may be adjusted according to the impurity concentration at the time of implanting the charged particle beam.

[0034] At least one of a volume donor concentration, a chemical oxygen concentration, and a chemical carbon concentration of the semiconductor substrate may be detected when detecting the impurity concentration.

[0035] When implanting the charged particle beam, the implantation depth of the charged particle beam may be further adjusted based on at least an oxygen contribution ratio ξ indicating a ratio of the chemical oxygen concentration that contributes to generating a hydrogen donor in the chemical oxygen concentration and a carbon contribution ratio η indicating a ratio of the chemical carbon concentration that contributes to generating a hydrogen donor in the chemical carbon concentration.

[0036] The summary does not necessarily describe all necessary features of the embodiments of the present invention. The present invention may also be a subcombination of the features described above. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a sectional view showing an example of the semiconductor device 100. Fig. 2 illustrates a depth-directed distribution of a chemical hydrogen concentration C H , a chemical oxygen concentration C OX , a vacancy concentration N V , a contributing hydrogen concentration N H and a contributing oxygen concentration N OX at positions defined by the line AA in Fig. 1 must be specified. Fig. 3 illustrates a depth-directed distribution of the chemical hydrogen concentration C H , the chemical oxygen concentration C OX, the contributing oxygen concentration N OX and a VOH vacancy concentration N VOH after heat treatment. Fig. 4 is a graph showing an example of the distribution of a donor concentration D D after heat treatment. Fig. Figure 5A is a diagram to explain a flat section 150. Fig. 5B is another example of a distribution of a volume donor concentration D0, a hydrogen donor concentration Db, and a donor concentration Dd in the flat section 150. Fig. Figure 5C is another example of a distribution of the volume donor concentration D0, the hydrogen donor concentration Db, and the donor concentration Dd in the flat section 150. Fig. Figure 6 is a graph showing a relationship between an incremental amount of a donor concentration and the chemical oxygen concentration C OX illustrated. Fig. Figure 7 is a graph showing a relationship between the amount of increase of a donor concentration and the chemical oxygen concentration C OX illustrated. Fig. Figure 8 is a graph showing a relationship among the dosage amount of hydrogen ions, an oxygen contribution fraction ξ and a vacancy concentration N V illustrated. Fig. Figure 9 is a graph showing a relationship among the dosage amount of hydrogen ions, the oxygen contribution fraction ξ and the vacancy concentration N V illustrated. Fig. 10 is a plan view illustrating an example of a semiconductor device 100. Fig. 11 is an enlarged view of an area D in Fig. 10. Fig. Figure 12 is a diagram showing an example of a cross section ee in Fig. 11 illustrated. Fig. 13 is a graph showing an example of a distribution of carrier concentration in the depth direction at positions of the line FF in Fig. 12 illustrated. Fig. 14 is a diagram showing an example of a cross section gg in Fig. 10 illustrated. Fig. 15 is a graph showing another example of the cross section gg in Fig. 10 illustrated. Fig. 16 is a diagram illustrating an example of a manufacturing method of the semiconductor device 100. Fig. 17 is a diagram illustrating another example of a calculation method of the implantation amount of a charged particle beam. Fig. 18 is a diagram illustrating another example of the manufacturing method of the semiconductor device 100. Fig. 19 is a diagram illustrating an example of a manufacturing step S1606 of the device. Fig. 20 is a diagram illustrating another example of the device manufacturing step S1606. Fig. 21 is a graph illustrating a relationship between the oxygen contribution fraction ξ and a depth position Z1 at which a second peak 141 is located. Fig. Figure 22 is a graph showing a relationship between the vacancy concentration N V and the depth position Z1 at which the second peak 141 is located. Fig. 23 illustrates a depth-directed distribution of a chemical hydrogen concentration C H , a chemical carbon concentration C C , a contributing carbon concentration N C and a VOH vacancy concentration N VOH after heat treatment at positions defined by line AA in Fig. 1 are specified. Fig. Figure 24 is a graph showing a relationship between the amount of increase in donor concentration and the chemical carbon concentration C C illustrated. Fig. Figure 25 is a graph showing a relationship between the amount of increase of a donor concentration and the chemical oxygen concentration C OX illustrated. Fig. Figure 26 is a graph illustrating a relationship between the dosage of hydrogen ions to the depth position Z1 and a carbon contribution fraction η. Fig. Figure 27 is a graph showing a relationship between the oxygen contribution fraction ξ and a dosage D H of hydrogen ions in a group with a small chemical carbon concentration C C illustrated. Fig. Figure 28 is a graph showing a relationship between the vacancy concentration N V and dosage D Hof hydrogen ions in a group with a small chemical carbon concentration C C illustrated. Fig. 29 is a diagram illustrating an example of a manufacturing method of the semiconductor device 100. Fig. Figure 30 is a graph showing another example of the relationship between the vacancy concentration N V and the depth position Z1. Fig. Figure 31 is a graph illustrating another example of the relationship between the oxygen contribution fraction ξ and the depth position Z1. Fig. Figure 32 is a graph illustrating a relationship between the carbon contribution fraction η and the depth position Z1. Fig. Figure 33A is a graph illustrating a dependence of vacancy concentration on the dosage amount of helium ions on target electrical properties. Fig. Figure 33B is a graph illustrating a dependence of the helium ion dosage amount of the oxygen contribution fraction on the target electrical properties. Fig. Figure 33C is a graph illustrating a dependence of the helium ion dosage amount of the carbon contribution fraction on the target electrical properties. Fig. Figure 34A is a diagram illustrating a dependence of the converted vacancy concentration Nv' on the depth of helium ions in the target electrical properties. Fig. Figure 34B is a diagram illustrating a dependence of the converted oxygen contribution fraction ξ' on the depth of the helium ions in the target electrical properties. Fig. Figure 34C is a diagram illustrating a dependence of the converted carbon contribution fraction η' on the depth of helium ions in the target electrical properties. Fig. 35 is a diagram illustrating another example of the manufacturing method of the semiconductor device 100. Fig. 36 is a diagram illustrating another example of the manufacturing method of the semiconductor device 100. Fig. 37 is a diagram illustrating another example of the manufacturing method of the semiconductor device 100. Fig. Figure 38 is a graph illustrating a relationship between a volume donor concentration and a charged particle implantation depth Z1. Fig. 39 is a diagram illustrating another example of the manufacturing method of the semiconductor device 100. Fig. 40 is a diagram illustrating an example of an equipotential surface 308 in an edge termination structure section 90. DESCRIPTION OF THE EMBODIMENTS

[0037] The present invention is described below using exemplary embodiments of the invention. The following exemplary embodiments do not limit the invention disclosed in the claims. Furthermore, not all combinations of features described in the exemplary embodiments are essential to the inventive solution.

[0038] In this specification, one side in the direction parallel to the depth direction of the semiconductor substrate is referred to as the "upper" side, and the other side is referred to as the "lower" side. One of two main surfaces of a substrate, layer, or other element is referred to as an upper surface, and the other surface is referred to as a lower surface. The "upper" and "lower" directions are not limited to the direction of gravity or the direction at the time of mounting the semiconductor device.

[0039] In this specification, technical matters may be described using orthogonal coordinate axes X, Y, and Z. The orthogonal coordinate axes merely indicate relative positions of components and do not restrict them to a specific direction. For example, the Z axis does not exclusively indicate the height direction relative to the ground. The +Z direction and the -Z direction are opposite directions. Where positive and negative are not specified and are described as the Z axis direction, a direction parallel to the +Z axis and the -Z axis is meant.

[0040] In this specification, orthogonal axes parallel to the upper surface and the lower surface of the semiconductor substrate are defined as an X-axis and a Y-axis. An axis perpendicular to the upper surface and the lower surface of the semiconductor substrate is defined as a Z-axis. In this specification, the Z-axis direction may be referred to as a depth direction. Furthermore, in this specification, a direction parallel to the upper surface and the lower surface of the semiconductor substrate, including the X-axis and the Y-axis, may be referred to as a horizontal direction.

[0041] Furthermore, a region from the center of the semiconductor substrate in the depth direction to the top surface of the semiconductor substrate may be referred to as the top surface side. Similarly, a region from the center of the semiconductor substrate in the depth direction to the bottom surface of the semiconductor substrate may be referred to as the bottom surface side.

[0042] In this specification, the term "same" or "the same" may include a case where an error is included due to a manufacturing variation or the like. The error is, for example, up to 10%.

[0043] In this specification, the conductivity type of the impurity-doped doping region is referred to as P-type or N-type. In this specification, the impurity may specifically refer to an N-type donor or a P-type acceptor and may be described as a dopant. In this specification, doping refers to the introduction of a donor or an acceptor into a semiconductor substrate to form a semiconductor with N-type or P-type conductivity.

[0044] In this specification, the doping concentration refers to the concentration of donors or the concentration of acceptors in the thermal equilibrium state. In this specification, the net doping concentration refers to the net concentration obtained by adding the donor concentration as the concentration of positive ions and the acceptor concentration as the concentration of negative ions, including the polarity of charges. For example, if the donor concentration N D and the acceptor concentration N A is, the net doping concentration at any position N D - N A . In the present specification, the net doping concentration may simply be referred to as a doping concentration.

[0045] The donor functions to provide electrons to the semiconductor. The acceptor functions to receive electrons from the semiconductor. The donor and acceptor are not limited to the impurities themselves. For example, a VOH defect, in which vacancies (V), oxygen (O), and hydrogen (H) are bonded in the semiconductor, can act as a donor that provides electrons. In this specification, the VOH defect may be referred to as a hydrogen donor.

[0046] In this description, the designations P + -like or N + -type that the doping concentration is higher than that of P-type or N-type, and the designations P-type or N - -type means that the doping concentration is lower than that of the designations P-type or N-type. Furthermore, in this description, the designations P ++ -like or N ++-like that the doping concentration is greater than that of P + -like or N + -like. The system of units used in this description is SI unless otherwise stated. The unit of length may be specified in cm, but various calculations can be performed by converting to meters (m).

[0047] In this specification, the chemical concentration refers to the atomic density of impurities, which is measured regardless of the state of electrical activation. The chemical concentration can be measured, for example, by secondary ion mass spectrometry (SIMS). The net doping concentration described above can be measured by a capacitance-voltage (CV) method. Furthermore, the carrier concentration measured by a spreading resistance (SR) measurement method can be a net doping concentration. The carrier concentration measured by the CV method can be a value at thermal equilibrium. In addition, since the donor concentration is sufficiently larger than the acceptor concentration in the N-type region, the carrier concentration in the region can be used as the donor concentration.Similarly, the carrier concentration in the P-type region may be defined as the acceptor concentration. In this specification, the doping concentration of the N-type region may be referred to as the donor concentration, and the doping concentration of the P-type region may be referred to as the acceptor concentration.

[0048] In addition, when the concentration distribution of the donor, acceptor, or net doping has a peak, the peak value may be the concentration of donors, acceptors, or net doping in the region. In a case where the concentration of donors, acceptors, or net doping is substantially (almost) uniform or the like, an average value of the concentration of donors, acceptors, or net doping in the corresponding region may be used as the concentration of donors, acceptors, or net doping. In the present specification, atoms / cm 3 or / cm 3Used to express concentrations per unit volume. This unit is used for the donor or acceptor concentration or the chemical concentration in the semiconductor substrate. The expression of atoms can be omitted. Each concentration in this description can be a value at room temperature. For example, a value of 300 K (Kelvin) (approximately 26.9 °C) can be used as the room temperature.

[0049] The carrier concentration measured by the SR method may be smaller than the concentration of donors or acceptors. In a region where current flows when measuring a spreading resistance, there is a case where the carrier mobility of the semiconductor substrate is lower than the value of the crystal state. The decrease in carrier mobility occurs due to disorder in the crystal structure caused by a lattice defect or the like, which scatters the carriers.

[0050] The concentration of donors or acceptors calculated from the charge carrier concentration measured by the CV or SR method may be lower than the chemical concentration of the elements that constitute the donor or acceptor. For example, the donor concentration of phosphorus or arsenic as a donor, or the acceptor concentration of boron as an acceptor in a silicon semiconductor, is approximately 99% of their chemical concentration. On the other hand, the donor concentration of hydrogen as a donor in a silicon semiconductor is approximately 0.1% to 10% of the chemical concentration of hydrogen.

[0051] Fig. 1 is a cross-sectional view illustrating an example of the semiconductor device 100. The semiconductor device 100 includes a semiconductor substrate 10. The semiconductor substrate 10 is a substrate formed from a semiconductor material. For example, the semiconductor substrate 10 is a silicon substrate.

[0052] At least one transistor device, such as an insulated gate bipolar transistor (IGBT) and / or a diode device, such as a freewheeling diode (FWD), is formed in the semiconductor substrate 10. In Fig. 1, each electrode of the transistor device and the diode device as well as each region in the semiconductor substrate 10 was omitted.

[0053] In the semiconductor substrate 10 of this example, N-type volume donors are distributed throughout. The volume donor is a dopant donor that is approximately uniformly contained in an ingot during the production of the ingot from which the semiconductor substrate 10 is made. The volume donor in this example is an element other than hydrogen. The dopant of the volume donor is, for example, a Group V or Group VI element and is, for example, phosphorus, antimony, arsenic, selenium, or sulfur, without being limited to these examples. The volume donor in this example is phosphorus. The main donor is also contained in the P-type region. The semiconductor substrate 10 may be a wafer cut from a semiconductor ingot or a chip obtained by cutting a wafer into individual pieces.The semiconductor ingot can be manufactured using either a Chokralski process (CZ process), a Chokralski process with applied magnetic field (MCZ process), or a float zone process (FZ process). The ingot in this example is manufactured using the MCZ process.

[0054] For example, the chemical oxygen concentration in the substrate produced by the MCZ process is 1 × 10 17 up to 7 × 10 17 atoms / cm 3 . The chemical oxygen concentration in the substrate produced by the FZ process is, for example, 1 × 10 15 up to 5 × 10 15 atoms / cm 3. The volume donor concentration may use a chemical concentration of the volume donor distributed throughout the semiconductor substrate 10 and may be a value between 90% and 100% of the chemical concentration. In the semiconductor substrate doped with dopants from groups V and VI, such as phosphorus, the volume donor concentration may be in the range of 1 × 10 11 / cm 3 up to 3 × 10 13 / cm 3 The volume donor concentration of the semiconductor substrate doped with dopants from groups V and VI is preferably in the range of 1 × 10 12 / cm 3 up to 1 × 10 13 / cm 3 . A non-doped substrate that does not contain a dopant such as phosphorus can be used as the semiconductor substrate 10. In this case, the volume donor concentration (N BO ) of the non-doped substrate, for example, in the range of 1 × 10 10 / cm 3 up to 5 × 1012 / cm 3 . The volume donor concentration (N BO ) of the non-doped substrate is preferably 1 × 10 11 / cm 3 or more. The volume donor concentration (N BO ) of the non-doped substrate is preferably 5 × 10 12 / cm 3 or less.

[0055] 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 the present specification, an orthogonal axis in a plane parallel to the upper surface 21 and the lower surface 23 is defined as an X-axis and a Y-axis, and an axis perpendicular to the upper surface 21 and the lower surface 23 is defined as a Z-axis.

[0056] A charged particle beam is implanted from the bottom surface 23 into the semiconductor substrate 10 at a predetermined depth position Z1. In the present specification, the distance in the Z-axis direction from the bottom surface 23 may be referred to as a depth direction. The depth position Z1 is a position where the distance in the Z-axis direction from the bottom surface 23 is Z1. The depth position Z1 is located on the bottom surface 21 side of the semiconductor substrate 10. Implanting the charged particle beam at the depth position Z1 means that an average distance (also referred to as a range) of charged particles passing through the interior of the semiconductor substrate 10 is Z1. The charged particles are accelerated by acceleration energy corresponding to the predetermined depth position Z1 and introduced into the semiconductor substrate 10.

[0057] An area where the charged particles have passed through the interior of the semiconductor substrate 10 is defined as a passage area 106. In the example of Fig. 1 is a region from the bottom surface 23 of the semiconductor substrate 10 to the depth position Z1 of the through region 106. The charged particles are particles that can form lattice defects in the through region 106. The charged particles are, for example, hydrogen ions, helium ions, or electrons. The charged particles can be implanted into the entire surface of the semiconductor substrate 10 in the XY plane or can be implanted into only a partial area.

[0058] The semiconductor substrate 10 has a second peak 141 of the charged particle concentration at the depth position Z1. In this example, the charged particles are hydrogen. That is, the semiconductor substrate 10 of this example has the second peak 141 of the chemical hydrogen concentration at the depth position Z1. The second peak 141 is a peak in the distribution of the chemical hydrogen concentration in the depth direction (Z-axis direction). The second peak may be a peak in the distribution of the chemical helium concentration.

[0059] In the passage region 106 through which the charged particles have passed in the semiconductor substrate 10, lattice defects are formed by the passage of charged particles, which mainly include vacancies such as monatomic vacancies (V) and divacancies (VV). Atoms adjacent to the vacancies have dangling bonds. Lattice defects include interstitial atoms, dislocations, and the like, and may include donors and acceptors in a broad sense. However, in the present specification, lattice defects mainly containing vacancies may be referred to as vacancy-type lattice defects, vacancy-type defects, or simply lattice defects. In addition, since a large number of lattice defects are formed due to the implantation of charged particles into the semiconductor substrate 10, the crystallinity of the semiconductor substrate 10 may be severely deteriorated.In the present description, the disturbance of crystallinity can be referred to as disorder.

[0060] In addition, oxygen is contained throughout the semiconductor substrate 10. The oxygen is intentionally or unintentionally introduced during the fabrication of a semiconductor ingot. Hydrogen is contained in at least a portion of the via region 106. The hydrogen can be intentionally implanted into the semiconductor substrate 10.

[0061] In this example, hydrogen ions are implanted from the bottom surface 23 to a depth position Z2. The hydrogen ions in this example are protons. The semiconductor substrate 10 of this example has a first peak 133 of the chemical hydrogen concentration at the depth position Z2. Fig. 1, the second peak 141 and the first peak 133 are schematically indicated by dashed lines. The depth position Z2 may be located in the through region 106. The depth position Z2 in this example is located on the lower surface side 23 of the semiconductor substrate 10. Hydrogen implanted at the depth position Z1 may be diffused into the through region 106, or hydrogen may be introduced into the through region 106 in another way. In these cases, hydrogen ions cannot be implanted into the depth position Z2.

[0062] After the via region 106 is formed in the semiconductor substrate 10 and hydrogen ions are implanted into the semiconductor substrate 10, hydrogen (H), vacancies (V), and oxygen (O) combine in the semiconductor substrate 10, and VOH defects are formed. Furthermore, the heat treatment of the semiconductor substrate 10 causes hydrogen to diffuse, promoting the formation of VOH defects. Furthermore, since hydrogen can be bonded to the vacancies by heat treatment after forming the via region 106, it is possible to suppress the release of hydrogen from the semiconductor substrate 10 to the outside.

[0063] The VOH defect acts as a donor that provides electrons. In the present description, VOH defects may simply be referred to as hydrogen donors. In the semiconductor substrate 10 of this example, a hydrogen donor is formed in the via region 106. The doping concentration of the hydrogen donor at each location is lower than the chemical concentration of hydrogen at each location. The ratio of the chemical hydrogen concentration contributing to the doping concentration of hydrogen donors (VOH defects) to the chemical hydrogen concentration is called the hydrogen contribution ratio. The hydrogen contribution ratio can be viewed as a ratio of the concentration of hydrogen atoms forming the VOH defects to the concentrations of all hydrogen atoms in a given region (e.g., the depth position from the bottom surface or the top surface).The hydrogen contribution fraction can be a value from 0.1% to 30% (i.e., in the range of 0.001 to 0.3). In this example, the hydrogen contribution fraction is 1% to 5%. Note that, unless otherwise stated in this specification, the VOH defects with a distribution similar to the distribution of the chemical hydrogen concentration and the VOH defects with a distribution similar to the vacancies in the via region 106 are both referred to as hydrogen donors or hydrogen as donors.

[0064] By forming a hydrogen donor in the through-region 106 of the semiconductor substrate 10, the donor concentration in the through-region 160 can be set higher than the volume donor concentration. Typically, it is necessary to prepare the semiconductor substrate 10 with a predetermined volume doping concentration according to the properties of a device to be formed in the semiconductor substrate 10, in particular according to a nominal voltage or breakdown voltage. On the other hand, the donor concentration of the semiconductor substrate 10 can be set according to the Fig. 1 by controlling the dosage of charged particles. Thus, the semiconductor device 100 can be manufactured using a semiconductor substrate 10 having a volume donor concentration that does not correspond to the characteristics and the like of the device. The deviation of the volume donor concentration at the time of manufacturing the semiconductor substrate 10 is relatively large, but the dosage of charged particles can be controlled with relatively high accuracy. Therefore, the concentration of lattice defects created by implanting charged particles can be controlled with high accuracy, and the donor concentration in the through region can be controlled with high accuracy.

[0065] The depth position Z1 may be in a range of half or less of the thickness of the semiconductor substrate 10 with respect to the upper surface 21, or may be in a range of 1 / 4 or less of the thickness of the semiconductor substrate 10. The depth position Z2 may be in a range of half or less of the thickness of the semiconductor substrate 10 with respect to the lower surface 23, or may be in a range of 1 / 4 or less of the thickness of the semiconductor substrate 10. However, the first depth position Z1 and the second depth position Z2 are not limited to these ranges.

[0066] Fig. 2 illustrates the depth-directed distribution of the chemical hydrogen concentration C H , the chemical oxygen concentration C OX , the vacancy concentration N V , the contributing hydrogen concentration N H and the contributing oxygen concentration N OX at positions defined by the line AA in Fig. 1 must be specified. Fig. Figure 2 illustrates each distribution immediately after the charged particles and hydrogen ions are implanted. That is, Fig. Figure 2 illustrates each distribution after charged particles and hydrogen ions are implanted into the semiconductor substrate 10 and before heat treatment is performed at a temperature above room temperature 25 °C.

[0067] In Fig. 2, the horizontal axis represents the depth position from the bottom surface 23 and the vertical axis represents each concentration per unit volume on a logarithmic axis. The chemical concentration in Fig. 2 is measured, for example, by a SIMS method. In Fig. 2, the volume donor concentration N BO indicated by a dashed line. The volume donor concentration N BOcan be uniform throughout the semiconductor substrate 10. A central depth position in the depth direction of the semiconductor substrate 10 is defined as Zc.

[0068] The distribution of the chemical hydrogen concentration C H has the second peak 141 at depth position Z1 and the first peak 133 at depth position Z2. The chemical hydrogen concentration C H has a local maximum value at the depth positions Z1 and Z2. The second peak 141 and the first peak 133 of the chemical hydrogen concentration C H are 1000 times or more than the minimum value of the chemical hydrogen concentration C H . The first peak 133 may be larger than the second peak 141. The first peak 133 may be 10 times or more or 100 times or more larger than the second peak 141.

[0069] The distribution of the chemical hydrogen concentration C Hhas an upper skirt 143 in which the chemical hydrogen concentration C H from the second peak 141 to the upper surface 21, and a lower skirt 142 in which the chemical hydrogen concentration C H from the second peak 141 to the lower surface 23. If hydrogen ions are implanted from the lower surface 23, the lower skirt 142 is gentler than the upper skirt 143. In the present specification, the term "the skirt is gentle" means that a position that is a half value of the corresponding peak value is farther away from the corresponding peak position.

[0070] The contributing hydrogen concentration N H is the concentration of hydrogen that forms VOH defects. Since VOH defects contain vacancies and oxygen in addition to hydrogen, the contributing hydrogen concentration N Hdepending on the concentration of vacancies and oxygen. The contributing hydrogen concentration N H can be between 0.1% and 30% of the chemical hydrogen concentration C H lay.

[0071] The distribution of the contributing hydrogen concentration N H resembles the distribution of the chemical hydrogen concentration C H . The contributing hydrogen concentration N H has a first contributing concentration peak 161 at or near the depth position Z1 and a second contributing concentration peak 151 at or near the depth position Z2.

[0072] Oxygen is often introduced during the production of an ingot and is often evenly distributed within the semiconductor substrate 10. The chemical oxygen concentration C OX may be uniform throughout the semiconductor substrate 10. In another example, the chemical oxygen concentration COX from the lower surface 23 to the upper surface 21 of the semiconductor substrate 10 may increase or decrease monotonically. In addition, oxygen near the upper surface 21 or the lower surface 23 of the semiconductor substrate 10 may be released from the semiconductor substrate 10 to the outside. The chemical oxygen concentration C OX may decrease monotonically toward the upper surface 21 and the lower surface 23 in the vicinity of the upper surface 21 and the lower surface 23. The vicinity of the upper surface 21 and the lower surface 23 is, for example, a region within a distance of 1 µm from the upper surface 21 or the lower surface 23, but is not limited thereto. Apart from the vicinity of the upper surface 21 and the lower surface 23, the chemical oxygen concentration C OX be uniform as described above and can increase or decrease monotonically.

[0073] The chemical oxygen concentration C OX can be in the range of 3 × 10 15 atoms / cm 3 up to 2 × 10 18 Atoms / cm 3 If in this description the chemical oxygen concentration C OX of the semiconductor substrate 10, everything between the second peak 141 and the first peak 133 satisfies the definition of chemical oxygen concentration, unless otherwise stated. Everything between the second peak 141 and the bottom surface 23 can satisfy the definition of oxygen concentration, and the entire semiconductor substrate 10 can satisfy the definition of oxygen concentration. The chemical oxygen concentration C OX can 1 × 10 16 atoms / cm 3 or more or 1 × 10 17 atoms / cm 3 or more. The chemical oxygen concentration C OX can 1 × 10 18 atoms / cm 3 or less or 1 × 10 17 atoms / cm3 or less.

[0074] The contributing oxygen concentration N OX is the concentration of oxygen that forms VOH defects. Since VOH defects contain vacancies and hydrogen in addition to oxygen, the contributing oxygen concentration N OX vary depending on the concentration of vacancies and hydrogen. In this description, the relationship between the contributing oxygen concentration N OX and the chemical oxygen concentration C OX defined as an oxygen contribution fraction ξ. That is, ξ = N OX / C OXThe oxygen contribution fraction ξ can be considered as a ratio of the concentration of oxygen atoms forming the VOH defect to the concentrations of all oxygen atoms in a given region (e.g., the lower surface or the depth position from the upper surface). ξ ranges from 0 to 1. The unit of the oxygen contribution fraction is a dimensionless quantity.

[0075] The distribution of the contributing oxygen concentration N OX can be similar to the distribution of the chemical oxygen concentration C OX For example, the contributing oxygen concentration N OX in the depth direction of the semiconductor substrate 10 and may monotonically increase or decrease. Alternatively, the contributing oxygen concentration N OX have a concentration distribution with a peak at a given depth position.

[0076] The vacancy concentration N Vhas a vacancy peak 171 at a depth position Zd. The depth position Zd may be the same as the depth position Z1 and may be slightly closer to the lower surface 23 than the depth position Z1. The contributing hydrogen concentration N H can find the first contributing concentration peak 161 at the same depth position as the vacancy concentration N V have.

[0077] If charged particles are implanted into the semiconductor substrate 10, damage is introduced into a region from the implantation surface of the semiconductor substrate 10 to a range portion of the charged particles. The damage refers to a disturbance of a crystal lattice and may be in an amorphous state in addition to vacancies and dislocations. The vacancy concentration N V can also show a peak at Z2. The vacancy concentration N Vcan be essentially (almost) uniformly monotonically increasing or monotonically decreasing between the peaks. In a case where the charged particles are electrons, the vacancy concentration N V from the top surface to the bottom surface of the semiconductor substrate 10 may be substantially (almost) uniform, may increase monotonically, may decrease monotonically, or may have a smooth distribution with a peak at a given position. The vacancy concentration N Vcan be calculated, for example, using software known as "Transport of Ions in Matter" (TRIM) (see, for example, http: / / www.srim.org / . A TRIM manual is disclosed at http: / / srim.org / SRIM / SRIM%2008.pdf and http: / / srim.org / SRIM / SRIM%2009.pdf. Part 2 of the manual describes a method for calculating the vacancy concentration). It is assumed that most vacancies are not occupied by hydrogen before heat treatment.

[0078] Fig. 3 illustrates the depth-directed distributions of the chemical hydrogen concentration C H , the chemical oxygen concentration C OX , the contributing oxygen concentration N OX and the VOH vacancy concentration N VOHafter heat treatment. Through the heat treatment, hydrogen diffuses from the second peak 141 and the first peak 133 to the upper surface side 21 and the lower surface side 23. Even after heat treatment, the relationship of the magnitude, the ratio, the value, and the like of the concentration of each peak and the chemical oxygen concentration can be the same as those before heat treatment in Fig. 2. As a result, the chemical hydrogen concentration C H between the second peak 141 and the first peak 133. Since in this example the first peak 133 is provided with a high concentration, more hydrogen is diffused from the first peak 133. Thus, the chemical hydrogen concentration C H in the range from depth position Z2 to depth position Z1 decreases monotonically over half or more of the length. The chemical hydrogen concentration C Hcan decrease monotonically from the depth position Z2 to the upper surface side 21 of the depth position Zc.

[0079] During heat treatment, hydrogen occupies the dangling bonds in the vacancies. As a result, donors of VOH defects (terminating dangling bonds) are formed. The concentration of VOH defects N VOH depends on the contributing hydrogen concentration N H , the contributing oxygen concentration N OX and the vacancy concentration N V The concentration of VOH defects N VOH In this example, a first VOH peak 191 is located near depth position Z1 and a second VOH peak 181 is located near depth position Z2. The first VOH peak 191 may be located at depth position Zd. In this example, the second VOH peak 181 has a higher concentration than the first VOH peak 191.

[0080] In addition, the semiconductor substrate 10 has a flat portion 150 between the depth position Z1 and the depth position Z2. The flat portion 150 is a region where the concentration distribution of the VOH defects N VOH is substantially (almost) flat. The flat portion 150 may be provided over half or more of the length between the depth position Z1 and the depth position Z2, or over a length of 75% or more.

[0081] In the shallow section 150, the contributing oxygen concentration N OX smaller than the chemical hydrogen concentration C OX In this case, the chemical hydrogen concentration C OX a minimum value in the shallow section 150 can be used. The contributing oxygen concentration N OX may contain 10% or less of the chemical hydrogen concentration C OX be.

[0082] Fig. Figure 4 is a graph showing an example of the distribution of donor concentration D D after heat treatment. In Fig. 4 are the chemical hydrogen concentration C H , the chemical oxygen concentration C OX and the volume donor concentration N BO together. The chemical hydrogen concentration C H and the chemical oxygen concentration C OX are the same as in the example of Fig. 3. The volume donor concentration N BO is identical to the one in the example of Fig. 2.

[0083] The donor concentration D D in this example is a concentration obtained by adding the concentration of VOH defects N VOH to the volume donor concentration N BO The donor concentration N BOhas a first donor peak 121 at depth position Zd and a second donor peak 111 at depth position Z2. In this example, the second donor peak 111 has a higher concentration than the first donor peak 121. In addition, the donor concentration D D in the flat section essentially (almost) flat. The donor concentration D D of the flat section 150 is higher than the volume donor concentration N BO .

[0084] Fig. 5A is a diagram for explaining the flat portion 150. The flat portion 150 is a portion where a region where the donor concentration D Dis continuous in the depth direction between a specified maximum value (max) and a specified minimum value (min). The maximum value (max) can be the maximum value of the donor concentration in the region. The minimum value (min) can be 50%, 70%, or 90% of the maximum value (max).

[0085] Alternatively, with respect to the average concentration of the donor concentration distribution in a predetermined range in the depth direction, a 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 described above, the concentration of the VOH defect N VOH in the flat section 150 also essentially (almost) flat, as the donor concentration D D .

[0086] Fig. Figure 5B is another example of a distribution of the volume donor concentration DO, the hydrogen donor concentration Db and the donor concentration Dd in the flat section 150. This example differs from the example of Fig. 5A in that the flat portion 150 has a slope in the depth direction. The thickness of the semiconductor substrate 10 in this example is 120 µm. The vertical axis in this drawing is a linear scale. The depth of 20 µm to 80 µm from the hydrogen ion-implanted surface is defined as a predetermined region. The predetermined region is a region through which hydrogen ions penetrate and in which there is no local peak of the donor concentration Dd. The thickness of the predetermined region in this example is 50% of the thickness of the semiconductor substrate 10. The volume donor concentration DO in this example is 3.1 × 10 13 / cm 3, which corresponds to 150 Ωcm. The sum of the volume donor concentration DO at each depth and the value of the terminal dangling bond Db is the donor concentration Dd.

[0087] A distribution obtained by connecting the concentrations at both ends of the specified range with a straight line may be an approximate linear distribution. The approximate linear distribution may be a straight line obtained by fitting the concentration in the specified range with a linear function. In addition, the approximate linear distribution may be a straight line obtained by fitting a distribution without local peaks of each concentration distribution with a linear function. Furthermore, a band-shaped region with a width of 30% of the approximate linear distribution value around the approximate linear distribution is called a band-shaped region.A monotonically increasing or decreasing concentration distribution within a given range refers to a state where the concentration values ​​at both ends of the given range are different, and the concentration distribution is contained within the band-shaped region described above. The band-shaped region can have a width of 20% or 10% of the value of the approximate linear distribution.

[0088] The linear approximate distribution 214 of the donor concentration Dd is a distribution in which the concentration increases with increasing distance from the implantation surface. The concentration of vacancies formed in a given region through which hydrogen ions penetrate has a distribution in which the concentration increases with increasing distance from the implantation surface. Diffused hydrogen occupies the dangling bond in the formed vacancies, thereby forming a distribution of the donor concentration of hydrogen according to the concentration distribution of the vacancies. In this example, the value of the donor concentration Db in a given region through which hydrogen ions penetrate fluctuates by approximately ± 7% with respect to the linear approximate distribution 214. The deviation of the donor concentration Dd is defined as a band-shaped region 216.That is, the width of the band-shaped region 216 in this example is ±7% of the value of the approximate linear distribution 214. In a predetermined region with a thickness of 30% or more of the thickness of the semiconductor substrate 10, when the distribution of the donor concentration Db lies within the band-shaped region 216, the distribution of the donor concentration Db may be a flat distribution. That is, this predetermined region may be a flat region with a hydrogen donor. The approximate linear distribution 214 of the donor concentration Dd may be a distribution in which the concentration decreases with increasing distance from the implantation surface.

[0089] Fig. Figure 5C is another example of the distribution of the volume donor concentration DO, the hydrogen donor concentration Db and the donor concentration Dd. This example differs from the example of the Fig. 5B in that the slope of the flat portion 150 is greater. In this example, a region in the depth of 10 µm to 70 µm from the hydrogen ion implanted surface is defined as a predetermined region. The thickness of the predetermined region with respect to the thickness (120 µm) of the semiconductor substrate 10 in this example is 50% the same thickness as in the example of Fig. 5B.

[0090] The linear approximate distribution 214 of the donor concentration Dd is a distribution in which the concentration increases with increasing distance from the implantation surface. However, the linear approximate distribution 214 of this example has a larger slope of increase than the linear approximate distribution 214 of the Fig. 5B. Furthermore, the value of the donor concentration Db fluctuates within the predetermined range by approximately ±17% with respect to the approximate linear distribution 214. The deviation of the donor concentration Dd is defined as a band-shaped region 216. The width of the band-shaped region 216 is ±17% of the value of the approximate linear distribution 214. Therefore, in a predetermined region with a thickness of 30% or more of the thickness of the semiconductor substrate 10, when the distribution of the donor concentration Db lies within the band-shaped region 216, the distribution of the donor concentration Db can be a flat distribution. That is, this predetermined region can be a flat hydrogen donor region.

[0091] The flat region with hydrogen donor can be provided in a range of 20% to 80% of the thickness of the semiconductor substrate. The magnitude of the slope of the linear approximate distribution 214 in the flat region with hydrogen donors can be in the range of 0 / (cm 3 µm) to 2 × 10 12 / (cm 3 ·µm) and can be greater than 0 / (cm 3 µm) and 1 × 10 12 / (cm 3 ·µm) or more in terms of depth (µm). In addition, the magnitude of the slope of the linear approximate distribution 214 in the shallow region with hydrogen donors can be in the range of 1 × 10 10 / (cm 3 µm) to 1 × 10 12 / (cm 3 ·µm) and can be in the range of 1 × 10 10 / (cm 3 µm) to 5 × 10 11 / (cm 3 ·µm) in relation to the depth (µm). Here, 5 × 10 11 / (cm 3 ·µm) the same gradient (equivalent) as 5 × 10 15 / cm 4 .

[0092] A semi-logarithmic slope can be used as a further index of the slope of the linear approximate distribution 214. The position of one end of the given range is defined as x1 (cm) and the position of the other end is defined as x2 (cm). The concentration at x1 is denoted by N1 ( / cm 3 ) and the concentration at x2 is denoted by N2 ( / cm 3 ). A semi-logarithmic slope η ( / cm) in a given range is defined as η = (log 10 (N2) - log 10 (N1)) / (x2 - x1). The magnitude of the semi-logarithmic slope η of the linear approximate distribution 214 in the flat region with a hydrogen donor can be in the range from 0 / cm to 50 / cm or in the range from 0 / cm to 30 / cm. Furthermore, the magnitude of the semi-logarithmic slope η of the linear approximate distribution 214 in the flat region with a hydrogen donor can be in the range from 0 / cm to 20 / cm or in the range from 0 / cm to 10 / cm.

[0093] It is assumed that in the passage region through which the hydrogen ions have passed, the vacancies (V, VV, etc.) generated by the passage of hydrogen are generated with a substantially (almost) uniform concentration in the depth direction. In addition, it is assumed that oxygen (O) or the like implanted at the time of manufacturing the semiconductor substrate 10 is also uniformly distributed in the depth direction. On the other hand, during the manufacture of the semiconductor device 100, oxygen may diffuse from the upper surface 21 or the lower surface 23 of the semiconductor substrate 10 to the outside of the semiconductor substrate 10 when the high-temperature treatment is performed at 1100°C or higher. As a result, the oxygen concentration toward the upper surface 21 or the lower surface 23 of the semiconductor substrate 10 may decrease.

[0094] According to the regulation with reference to Fig. 1 to Fig. 5C, the vacancy concentration N V be controlled by the dosage of charged particles and the chemical hydrogen concentration C H can be controlled by the dosage of hydrogen ions. Therefore, the donor concentration D D in the flat portion 150 can be easily controlled. By adjusting the charged particle implantation position Z1, the area in which the flat portion 150 is formed can be easily controlled.

[0095] Next, the range and the like of the oxygen contribution ratio ξ in the semiconductor substrate 10 will be described. A final doping concentration at any position from the depth position Z1 to the depth position Z2 in a state where the semiconductor device 100 is completed is referred to as N F defined. The doping concentration N F is described by expression (1). NF=NB0+NVOH

[0096] Here the concentration of VOH defects is N VOH the sum of the concentration of vacancies occupied by hydrogen and the contributing oxygen concentration N OX (that is, ξC OX ). The reason for this is that the concentration of VOH defects in the shallow section 150 is limited by the hydrogen-occupied vacancy concentration and the contributing oxygen concentration. In this example, essentially all vacancies from depth position Z1 to depth position Z2 are occupied by hydrogen, since the chemical hydrogen concentration C H is sufficiently high. This means that the vacancy concentration can exceed the vacancy concentration N V which is occupied by hydrogen. This gives expression (2). NVOH=NV+ξCOX

[0097] Expression (3) results from expression (1) and expression (2). NF=NB0+NV+ξCOX

[0098] Here, a case is considered where the semiconductor device 100 is formed using two semiconductor substrates having the same conditions for hydrogen implantation and heat treatment, the same volume donor concentration N BO and various chemical oxygen concentrations C OX The final doping concentration in the first semiconductor substrate is N F1 and the final doping concentration in the second semiconductor substrate is N F2 . In addition, the chemical oxygen concentration in the first semiconductor substrate is C OX , and the chemical oxygen concentration in the second semiconductor substrate is C OX2 .

[0099] Since the conditions for hydrogen implantation are identical, the vacancy concentrations N Vin the respective semiconductor substrates. Thus, the oxygen contribution fraction ξ is also identical between the two semiconductor substrates. Expressions (4) and (5) follow from expression (3). NF1=NB0+NV+ξCOX1 NF2=NB0+NV+ξCOX2Here NF2>NF1.

[0100] The difference in the final doping concentration is the difference ΔN VOH in the concentration of VOH defects N VOH . Thus, expression (6) results from expressions (4) and (5). NF2−NF1=ΔNVOH =ξ(COX2−COX1)ξ=ΔNVOH / (COX2−COX1) Furthermore, expression (7) results from expressions (2) and (6). NV=NVOH−ξCOX=NVOH−(ΔNVOH / (COX2−COX1))COX

[0101] If the oxygen contribution fraction ξ is specified, the vacancy concentration N V for any concentration of VOH defects N VOH and chemical oxygen concentration C OX can be calculated from expression (7).

[0102] Fig. Figure 6 is a graph showing a relationship between the amount of increase of a donor concentration and the chemical oxygen concentration C OX In this example, for two substrates—a semiconductor substrate before implanting charged particles and hydrogen ions, and a semiconductor substrate after implanting charged particles and hydrogen ions and after heat treatment—each carrier concentration at the depth position Zc is measured using the SR method, and the difference between them is considered the increase in the donor concentration. The increase in the donor concentration corresponds to the concentration of VOH defects N. VOH . In addition, the chemical oxygen concentration C OX of the semiconductor substrate of this example is uniformly distributed in the depth direction.

[0103] In Fig. 6, the depth position Z1 is set to 100 µm and the hydrogen ions are implanted at the depth position Z1. In this example, the dosage of hydrogen ions to the depth position Z1 is measured for three types of 3 × 10 12 ions / cm 2 , 1 × 10 13 ions / cm 2 and 3 × 10 13 ions / cm 2 As shown in Fig. 6, the donor concentration increases linearly proportional to the chemical oxygen concentration C OX .

[0104] In the example of Fig. 6, an approximate expression is calculated in which the relationship between the chemical oxygen concentration C OX and the amount of increase in donor concentration (ie, N VOH ) is approximated by a straight line. In Fig. 6 shows an example where the dosage of hydrogen ions is 3 × 10 13 ions / cm 2is approximated by a straight line 601, an example where the Do of the hydrogen ions is 1 × 10 13 ions / cm 2 is approximated by a straight line 602 and an example where the dosage of hydrogen ions is 3 × 10 12 ions / cm 2 is approximated by a straight line 603.

[0105] Each line is described by expression (8). NVOH=a×COX+b

[0106] At this time, the slope a and intercept b of each line are as follows by least squares fitting. Line 601: a=2.96303×10−4, b=2.18399×1013 Line 602: a=1.87895×10−4, b=1.47920×1013 Line 603: a=7.58824×10−5, b=6.38380×1012

[0107] Note that comparing expression (2) and expression (8), we get a = ξ and b = N V .

[0108] Fig. Figure 7 is a graph showing a relationship between the amount of increase of a donor concentration and the chemical oxygen concentration C OX In this example, the depth position Z1 is 50 µm. Other conditions are identical to those in Fig. 6. In this example, the increase in donor concentration is also proportional to the chemical oxygen concentration C OX increases linearly.

[0109] Fig. Figure 8 is a graph showing a relationship among the dosage amount of hydrogen ions, the oxygen contribution fraction ξ and the vacancy concentration N V illustrated. In Fig. 8 are the ones in the example of Fig. 6 and approximated by a curve. In this example, the oxygen contribution fraction ξ and the vacancy concentration N V by power functions with respect to the dosage D Hof the hydrogen ions to the depth position Z1. The reason for this is that the oxygen contribution ξ is assumed to be 0 when the dosage D H of the hydrogen ions tends to 0. If the oxygen contribution ξ is approximated by a logarithmic function, the dosage of the hydrogen ions D H a finite value greater than 0 when the oxygen contribution fraction ξ becomes 0. If the dosage of hydrogen ions D H continues to decrease, the oxygen contribution ξ becomes a negative value. The same applies to the vacancy concentration N V .

[0110] In Fig. 8, the relationship between the oxygen contribution fraction ξ and the dosage of hydrogen ions is approximated by a curve 801 and the relationship between the vacancy concentration N Vand the dosage of hydrogen ions is approximated by a curve 802. Curve 801 is described by expression (9), and curve 802 is described by expression (10). The coefficients c to f are determined by least squares fitting as follows. ξ=c×(DH)d where c=3.11503×10−12,d=5.94169×10−1 NV=e×(DH)f where e=1.36398×106 and f=5.36782×10−1

[0111] If the dosage of hydrogen ions D H is too small, the concentration of VOH defects N VOH In this case, it is difficult to determine the amount of increase in the donor concentration that causes the fluctuation of the volume donor concentration N BO absorb. Therefore, the dosage of hydrogen ions is preferably 1 × 10 11 ions / cm 2 or more. At this time, the oxygen contribution fraction ξ is 1 × 10 -5or more according to expression (9). If, in addition, the dosage of hydrogen ions D H is too large, the vacancy concentration N V compared to the concentration of VOH defects that can be formed by the oxygen and hydrogen concentrations. Thus, the number of vacancies that do not become VOH defects increases. The dosage of hydrogen ions is preferably 1.2 × 10 14 ions / cm 2 or less. At this time, the oxygen contribution fraction ξ is 7 × 10 -4 or less according to expression (9). That is, the oxygen contribution fraction ξ can be in the range of 1 × 10 -5 up to 7 × 10 -4 The oxygen contribution fraction ξ can be 1 × 10 -4 or more. The oxygen contribution fraction ξ can be 5 × 10 -4 or less. Similarly, the vacancy concentration N V in a range of 1 × 10 11 / cm 3 up to 1 × 1014 / cm 3 The vacancy concentration N V can 1 × 10 12 / cm 3 or more or 1 × 10 13 / cm 3 or more. The vacancy concentration N V can 3 × 10 13 / cm 3 or less. The vacancy concentration N V can be calculated from a difference (N VOH - N OX ) between the concentration of VOH defects N VOH and the contributing oxygen concentration N OX The concentration of VOH defects N VOH can be calculated from a difference (N F - N BO ) between the final doping concentration N F and the volume donor concentration N BO be calculated.

[0112] As the chemical oxygen concentration C OX a general value of the MCZ substrate is used. That is, C OX is 1 × 10 17 up to 7 × 10 17 / cm3 . Since N OX = ξ × C OX the contributing oxygen concentration is N OX 1 × 10 12 / cm 3 up to 5 × 10 14 / cm 3 From expression (2) we get N VOH = N V + N OX Similarly, the concentration of VOH defects N VOH in a range of 2 × 10 12 / cm 3 up to 6 × 10 14 / cm 3 The concentration of VOH defects N VOH can 1 × 10 13 / cm 3 or more. The concentration of VOH defects N VOH can 5 × 10 14 / cm 3 or less.

[0113] Fig. Figure 9 is a graph showing a relationship among the dosage amount of hydrogen ions, the oxygen contribution fraction ξ and the vacancy concentration N V illustrated. Fig. 9 corresponds to the example of Fig. 7 (i.e., Z1 = 50 µm). In Fig. 9, the relationship between the oxygen contribution fraction ξ and the dosage of hydrogen ions is approximated by a curve 901 and the relationship between the vacancy concentration N V and the dosage of hydrogen ions is approximated by a curve 902. Curve 901 is described by expression (12), and curve 902 is described by expression (11). The coefficients c to f are determined by least squares fitting as follows. ξ=c×(DH)d where c=1.53343×10−12,d=6.25800×10−1 NV=e×(DH)f where e=3.11098×103 and f=7.41056×10−1

[0114] Also in this example the oxygen contribution ξ can be in the same range as in the example of the Fig. 8. The vacancy concentration N V can be in the same range as in the example of Fig. 8. The contributing oxygen concentration N OXcan also be in the same range as in the example of Fig. 8. The concentration of VOH defects can also be in the same range as in the example of Fig. 8.

[0115] In the flat section 150 the relationship between a maximum value O max and a minimum value O min the chemical oxygen concentration C OX defined as v. That is, v = O min / O max . The ratio v can be in the range from 0.1 to 1. If the ratio v is small, the fluctuation of the concentration of VOH defects N VOH in the depth direction, and the breakdown voltage of the semiconductor substrate 10 may deteriorate. The ratio v may be 0.3 or more, or 0.5 or more. The ratio v may be 0.95 or less, or 0.9 or less.

[0116] The flat section 150 may contain carbon. The chemical carbon concentration in the flat section 150 may be in the range of 1 × 10 13 / cm 3 up to 1 × 10 16 / cm 3 The chemical carbon concentration can be 1 × 10 14 atoms / cm 3 or more. The chemical carbon concentration can be 5 × 10 15 atoms / cm 3 or less or 2 × 10 15 Atoms / cm 3 or less. The contribution of the chemical carbon concentration to the concentration of VOH defects N VOH will be described later.

[0117] Fig. 10 is a plan view illustrating an example of the semiconductor device 100. Fig. 10 illustrates a position where each element is projected onto the upper surface of the semiconductor substrate 10. In Fig. 10, only some elements of the semiconductor device 100 are shown and other elements have been omitted.

[0118] The semiconductor device 100 includes the semiconductor substrate 10. The semiconductor substrate 10 can support the distribution of each Fig. 1 to Fig. 9. However, the semiconductor substrate 10 may further have another concentration peak different from each Fig. 1 to Fig. 9. As in a buffer region 20 described later, hydrogen ions may be implanted to form an N-type region in the semiconductor substrate 10. In this case, the distribution of the chemical hydrogen concentration may include a local peak of the hydrogen concentration in addition to the concentration in Fig. 2 and the like. In addition, as in the emitter region 12 described later, an N-type impurity other than hydrogen, such as phosphorus, may be implanted to form an N-type region in the semiconductor substrate 10. In this case, the donor concentration distribution may include a local peak of the donor concentration in addition to the donor concentration shown in Fig. 4 described distribution of donor concentration.

[0119] The semiconductor substrate 10 has an end face 162 in plan view. When "plan view" is used in this description, it means a view from the upper surface of the semiconductor substrate 10. The semiconductor substrate 10 of this example includes two sets of end faces 162 that are opposite each other in plan view. Fig. 10, the X-axis and the Y-axis are arranged parallel to one of the end faces 162. The Z-axis is perpendicular to the upper surface of the semiconductor substrate 10.

[0120] The semiconductor substrate 10 has an active portion 160. The active portion 160 is a region through which a main current flows in the depth direction between the top surface and the bottom surface of the semiconductor substrate 10 when the semiconductor device 100 is operated. An emitter electrode is arranged above the active portion 160, but has been Fig. 10 omitted.

[0121] In the active section 160, at least one transistor section 70 comprising a transistor device such as an IGBT or a diode device 80 comprising a diode element such as a freewheeling diode (FWD) is arranged. In the example of Fig. 10, the transistor portion 70 and the diode portion 80 are alternately arranged along a predetermined arrangement direction (in the present example, the X-axis direction) on the upper surface of the semiconductor substrate 10. In another example, only one of the transistor portion 70 and the diode portion 80 may be arranged in the active portion 160.

[0122] In Fig. 10, a region where the transistor section 70 is arranged is indicated by a symbol "I" and a region where the diode section 80 is arranged is indicated by a symbol "F". In the present description, a direction perpendicular to the arrangement direction in plan view may be referred to as a propagation direction (Y-axis direction in Fig. 10). Each of the transistor section 70 and the diode section 80 may have a longitudinal length in the propagation direction. That is, the length of the transistor section 70 in the Y-axis direction is greater than its width in the X-axis direction. Similarly, the length of the diode section 80 in the Y-axis direction is greater than its width in the X-axis direction. The propagation direction of the transistor section 70 and the diode section 80 may be the same as the longitudinal direction of each trench section, which will be described below.

[0123] The diode section 80 comprises an N +-like cathode region in a region connected to the lower surface of the semiconductor substrate 10. In the present description, a region in which the cathode region is arranged is referred to as a diode portion 80. That is, the diode portion 80 is a region that overlaps with the cathode region in plan view. On the lower surface of the semiconductor substrate 10, a P + -type collector region 22 may be disposed in a region different from the cathode region. In the present description, the diode section 80 may also include an extension region 81 formed by extending the diode section 80 in the Y-axis direction to a gate rotor described later. A collector region is disposed on a lower surface of the extension region 81.

[0124] The transistor section 70 comprises a P +-type collector region in a region connected to the lower surface of the semiconductor substrate 10. In the transistor section 70, a gate structure including an N-type emitter region, a P-type base region, a gate wiring portion, and a gate insulating film is periodically arranged on the upper surface side of the semiconductor substrate 10.

[0125] The semiconductor device 100 may include one or more pads above the semiconductor substrate 10. The semiconductor device 100 of this example includes a gate pad 164. The semiconductor device 100 may include pads such as an anode pad, a cathode pad, and a current detection pad. Each pad is located near the end face 162. The vicinity of the end face 162 refers to a region located between the end face 162 and the emitter electrode in plan view. At the time of assembling the semiconductor device 100, each pad may be connected to an external circuit via wiring such as a wire.

[0126] A gate potential is applied to the gate pad 164. The gate pad 164 is electrically connected to the conduction portion of the gate trench portion of the active portion 160. The semiconductor device 100 includes a gate runner connecting the gate pad 164 and the gate trench portion. Fig. 10 the gate runner is hatched with diagonal lines.

[0127] The gate runner of this example includes an outer circumferential gate runner 130 and an active-side gate runner 131. The outer circumferential gate runner 130 is disposed between the active portion 160 and the end face 162 of the semiconductor substrate 10 in plan view. The outer circumferential gate runner 130 of this example surrounds the active portion 160 in plan view. A region surrounding the outer circumferential gate runner 130 in plan view may be the active portion 160. The outer circumferential gate runner 130 is connected to the gate pad 164. The outer circumferential gate runner 130 is disposed above the semiconductor substrate 10. The outer circumferential gate runner 130 may be a metal wiring including aluminum or the like.

[0128] The active-side gate slider 131 is arranged in the active section 160. By providing the active-side gate slider 131 in the active section 160, it is possible to reduce a deviation of the wiring length from the gate pad 164 in any region of the semiconductor substrate 10.

[0129] The active-side gate runner 131 is connected to the gate trench portion of the active section 160. The active-side gate runner 131 is disposed above the semiconductor substrate 10. The active-side gate runner 131 may be a wiring formed from a semiconductor material such as polysilicon doped with impurities.

[0130] The active-side gate runner 131 may be connected to the outer orbiting gate runner 130. The active-side gate runner 131 of this example is arranged to extend in the X-axis direction from one outer orbiting gate runner 130 up to the other outer orbiting gate runner 130, so that it traverses the active section 160 substantially in the middle of the Y-axis direction. In a case where the active section 160 is divided by the active-side gate runner 131, the transistor section 70 and the diode section 80 may be arranged alternately in the X-axis direction in each divided region.

[0131] In addition, the semiconductor device 100 may include a temperature measuring section (not shown) that is a PN junction diode formed of polysilicon or the like, and a current measuring section (not shown) that simulates the operation of the transistor section arranged in the active section 160.

[0132] The semiconductor device 100 of this example includes an edge termination structure portion 90 between the active portion 160 and the end face 102 in plan view. The edge termination structure portion 90 of this example is arranged between the outer circumferential gate slider 130 and the end face 162. The edge termination structure portion 90 reduces an electric field strength on the upper surface side of the semiconductor substrate 10. The edge termination structure portion 90 may include at least a guard ring, a field plate, and a RESURF circularly surrounding the active portion 160.

[0133] Fig. 11 is an enlarged view of an area D in Fig. 10. Region D is a region including the transistor portion 70, the diode portion 80, and the active-side gate slider 131. The semiconductor device 100 of this example includes a gate trench portion 40, a dummy trench portion 30, a well region 11, an emitter region 12, a base region 14, and a contact region 15 disposed in the upper surface side of the semiconductor substrate 10. Both the gate trench portion 40 and the dummy trench portion 30 are each examples of a trench portion. Furthermore, the semiconductor device 100 of this example includes an emitter electrode 52 and the active-side gate slider 131 disposed above the upper surface of the semiconductor substrate 10. The emitter electrode 52 and the active-side gate slider 131 are spaced apart from each other.

[0134] An interlayer dielectric film is arranged between the emitter electrode 52 and the active-side gate mover 131 and the upper surface of the semiconductor substrate 10, but the illustration thereof has been omitted in Fig. 11 is omitted. In the interlayer dielectric film of this example, a contact hole 54 is arranged through the interlayer dielectric film. In Fig. 11, each contact hole 54 is hatched with oblique lines.

[0135] The emitter electrode 52 is arranged above the gate trench portion 40, the dummy trench portion 30, the well region 11, the emitter region 12, the base region 14, and the contact region 15. The emitter electrode 52 is connected to the emitter region 12, the contact region 15, and the base region 14 on the upper surface of the semiconductor substrate 10 through the contact hole 54. The emitter electrode 52 is connected to a dummy line portion in the dummy trench portion 30 through a contact hole arranged in the interlayer dielectric film. The emitter electrode 52 may be connected to the dummy line portion of the dummy trench portion 30 at the edge of the dummy trench portion 30 in the Y-axis direction.

[0136] The active-side gate mover 131 is connected to the dummy trench portion 40 through a contact hole disposed in the interlayer dielectric film. The active-side gate mover 131 may be connected to the gate line portion of the gate trench portion 40 at an edge portion of the gate trench portion 40 in the Y-axis direction. The active-side gate mover 131 is not connected to the dummy line portion in the dummy trench portion 30.

[0137] The emitter electrode 52 is formed of a material containing metal. Fig. 11 illustrates a region in which the emitter electrode 52 is arranged. For example, at least a portion of the emitter electrode 52 is formed from aluminum or an aluminum-silicon alloy, e.g., a metal alloy such as AlSi or AlSiCu. The emitter electrode 52 may have a metal barrier made of titanium or a titanium composite in a layer of a region made of aluminum or the like. Furthermore, the contact hole may contain a plug with tungsten or the like buried therein, which plug is connected to the metal barrier, aluminum, or the like.

[0138] The drain region 11 is arranged overlapping with the active-side gate runner 131. The drain region 11 extends with a predetermined width even in a region where the active-side gate runner 131 does not overlap. The drain region 11 of this example is arranged away from the end of the contact hole 54 in the Y-axis direction from the active-side gate runner 131. The drain region 11 is a region of a second conductivity type having a higher doping concentration than the base region 14. The base region 14 in this example is P - -like and the sink area 11 is P + -like.

[0139] The transistor section 70 and the diode section 80 each include a plurality of trench sections arranged in the arrangement direction. In the transistor section 70 of this example, one or more gate trench sections 40 and one or more dummy trench sections 30 are alternately arranged along the arrangement direction. In the diode section 80 of this example, a plurality of dummy trench sections 30 are arranged along the arrangement direction. The diode section 80 of this example does not have a gate trench section 40.

[0140] The gate trench portion 40 of this example may include two straight portions 39 (portions of the trenches that are straight along the extension direction) extending along the extension direction perpendicular to the arrangement direction, and the edge portion 41 for connecting the two linear portions 39. The extension direction in Fig. 11 is the Y-axis direction.

[0141] At least a portion of the edge portion 41 is preferably arranged in a curved shape in plan view. By connecting the ends of the two straight portions 39 in the Y-axis direction to each other through the edge portion 41, the electric field strength at the end portion of the straight portion 39 can be reduced.

[0142] In the transistor section 70, the dummy trench section 30 is arranged between the straight sections 39 of the gate trench section 40. One dummy trench section 30 may be arranged between the straight sections 39, and a plurality of dummy trench sections 30 may be provided. The dummy trench section 30 may have a straight shape extending in the extension direction and may include a straight section 29 and an edge section 31 similar to the gate trench section 40. Fig. The semiconductor device 100 shown in Fig. 11 includes both the straight dummy trench portion 30 without the edge portion 31 and the dummy trench portion 30 with the edge portion 31.

[0143] The diffusion depth of the well region 11 may be deeper than the depths of the gate trench portion 40 and the dummy trench portion 30. The end portions of the gate trench portion 40 and the dummy trench portion 30 in the Y-axis direction include the well region 11 in plan view. That is, the bottom portion of each trench portion in the depth direction is aligned with the well region 11 at the end portion of each trench portion in the Y-axis direction. As a result, the electric field strength at the bottom portion of each trench portion can be reduced.

[0144] A mesa portion is arranged between the trench portions in the arrangement direction. The mesa portion refers to a region sandwiched between the trench portions in the semiconductor substrate 10. For example, the top end of the mesa portion is the upper surface of the semiconductor substrate 10. The depth position of the bottom end of the mesa portion is the same as the depth position of the bottom end of the trench portion. The mesa portion of this example is arranged to extend in the extension direction (Y-axis direction) along the trench in the upper surface of the semiconductor substrate 10. In this example, the transistor portion 70 includes a mesa portion 60, and the diode portion 80 includes a mesa portion 61. If simply a mesa portion is mentioned in this specification, the mesa portion refers to the mesa portion 60 and the mesa portion 61, respectively.

[0145] The base region 14 is arranged in each mesa section. In the base region 14 exposed to the upper surface of the semiconductor substrate 10 in the mesa section, a region arranged closest to the active-side gate slider 131 is defined as a base region 14-e. Fig. 11, the base region 14-e is shown arranged at one end portion of each mesa portion in the extension direction, but the base region 14-e is also arranged at the other end portion of each mesa portion. In each mesa portion, at least one of the emitter region 12 of the first conductivity type and the contact region 15 of the second conductivity type may be arranged in a region enclosed between the base regions 14-e in plan view. The emitter region 12 in this example is N + -like and the contact area 15 is P +-like. The emitter region 12 and the contact region 15 can be arranged between the base region 14 and the upper surface of the semiconductor substrate 10 in the depth direction.

[0146] The mesa portion 60 of the transistor portion 70 includes the emitter region 12 exposed to the top surface of the semiconductor substrate 10. The emitter region 12 is connected to the gate trench portion 40. The mesa portion 60 connected to the gate trench portion 40 may include the contact region 15 exposed to the top surface of the semiconductor substrate 10.

[0147] The contact region 15 and the emitter region 12 in the mesa section 60 are each arranged from one trench section to the other trench section in the X-axis direction. For example, the contact regions 15 and the emitter regions 12 of the mesa section 60 are arranged alternately along the extension direction (Y-axis direction) of the trench section.

[0148] In another example, the contact region 15 and the emitter region 12 of the mesa portion 60 may be arranged in a stripe shape along the extension direction (Y-axis direction) of the trench portion. For example, the emitter region 12 is arranged in a region connected to the trench portion, and the contact region 15 is arranged in a region enclosed by the emitter region 12.

[0149] The emitter region 12 is not disposed in the mesa section 61 of the diode section 80. The base region 14 and the contact region 15 may be disposed on the upper surface of the mesa section 61. The contact region 15 may be connected to each of the base regions 14-e and disposed in a region sandwiched between the base regions 14-e in the upper surface of the mesa section 61. The base region 14 may be disposed in a region sandwiched between the contact regions 15 in the upper surface of the mesa section 61. The base region 14 may be provided in the entire region sandwiched by the contact region 15.

[0150] The contact hole 54 is arranged above each mesa portion. The contact hole 54 is arranged in a region sandwiched between the base regions 14. The contact hole 54 of this example is provided above each region of the contact region 15, the base region 14, and the emitter region 12, respectively. The contact hole 54 is not arranged in a region corresponding to the base region 14-e and the drain region 11. The contact hole 54 may be arranged at the center in the arrangement direction (X-axis direction) of the mesa portion 60.

[0151] In the diode section 80 there is an n + -like cathode region 82 is arranged in a region adjacent to the lower surface of the semiconductor substrate 10. In the lower surface of the semiconductor substrate 10, a P +-like collector region 22 may be arranged in a region where the cathode region 82 is not arranged. The cathode region 82 and the collector region 22 are arranged between the lower surface 23 of the semiconductor substrate 10 and the buffer region 20. In Fig. 11, the boundary between the cathode region 82 and the collector region 22 is shown with a dotted line.

[0152] The cathode region 82 is spaced apart from the drain region 11 in the Y-axis direction. As a result, a distance is ensured between the P-type region (drain region 11) with a relatively high doping concentration, which is formed to a deep position, and the cathode region 82, and the breakdown voltage can be improved. In the Y-axis direction, the end portion of the cathode region 82 of this example is located farther from the drain region 11 than the end portion of the contact hole 54 in the Y-axis direction. In another example, the end portion of the cathode region 82 may be located between the drain region 11 and the contact hole 54 in the Y-axis direction.

[0153] Fig. Figure 12 is a diagram showing an example of a cross section ee in Fig. 11. The cross section ee is an XZ plane passing through the emitter region 12 and the cathode region 82. The semiconductor device 100 of this example includes, in cross section, the semiconductor substrate 10, an interlayer dielectric film 38, the emitter electrode 52, and a collector electrode 24.

[0154] The interlayer dielectric film 38 is disposed in the upper surface of the semiconductor substrate 10. The interlayer dielectric film 38 is a film containing at least one of an insulating film, such as silicate glass, to which an impurity such as boron or phosphorus is added, a thermal oxide film, and other insulating films. The interlayer dielectric film 38 is in contact with the Fig. 11 described contact hole 54.

[0155] The emitter electrode 52 is disposed above the interlayer dielectric film 38. The emitter electrode 52 is connected to the upper surface 21 of the semiconductor substrate 10 through the contact hole 54 of the interlayer dielectric film 38. The collector electrode 24 is disposed on the lower surface 23 of the semiconductor substrate 10. The emitter electrode 52 and the collector electrode 24 are formed of a metal such as aluminum. In this specification, a direction (Z-axis direction) connecting the emitter electrode 52 and the collector electrode 24 is referred to as the depth direction.

[0156] The semiconductor substrate 10 has an N-type or N-type drift region 18. The drift region 18 is arranged in the transistor section 70 and the diode section 80, respectively.

[0157] In the mesa section 60 of the transistor section 70, an N +-type emitter region 12 and a P-type base region 14 are arranged in sequence from the upper surface 21 of the semiconductor substrate 10. The drift region 18 is arranged below the base region 14. The mesa section 60 may have an N + -like accumulation region 16. The accumulation region 16 is arranged between the base region 14 and the drift region 18.

[0158] The emitter region 12 is exposed to the upper surface 21 of the semiconductor substrate 10 and is connected to the gate trench section 40. The emitter region 12 can be connected to the trench sections on either side of the mesa section 60. The emitter region 12 has a higher doping concentration than the drift region 18.

[0159] The base region 14 is disposed below the emitter region 12. The base region 14 of this example is connected to the emitter region 12. The base region 14 may be connected to the trench portions on either side of the mesa portion 60.

[0160] The accumulation area 16 is arranged below the base area 14. The accumulation area 16 is an N + -like region having a higher doping concentration than the drift region 18. By providing the high-concentration accumulation region 16 between the drift region 18 and the base region 14, the carrier implantation amplification effect (IE effect) can be enhanced and the ON voltage can be reduced. The accumulation region 16 can be arranged covering the entire lower surface of the base region 14 in each mesa section 60.

[0161] The mesa section 61 of the diode section 80 has a P-type base region 14 connected to the upper surface 21 of the semiconductor substrate 10. The drift region 18 is arranged below the base region 14. In the mesa section 61, the accumulation region 16 can be arranged below the base region 14.

[0162] In each of the transistor section 70 and the diode section 80, an N +-like buffer region 20 may be arranged below the drift region 18. The doping concentration of the buffer region 20 is higher than the doping concentration of the drift region 18. The buffer region 20 includes a peak 25 having a larger doping concentration than the drift region 18. The doping concentration of the peak 25 indicates a doping concentration at the local maximum value of the peak 25. An average value of the doping concentrations in a region where the doping concentration distribution is substantially (almost) flat can be used as the doping concentration of the drift region 18.

[0163] The buffer region 20 of this example has three or more peaks 25 in the depth direction (Z-axis direction) of the semiconductor substrate 10. The peak 25 of the buffer region 20 is, for example, a concentration peak of hydrogen (proton) or phosphorus. The buffer region 20 can serve as a field stop layer that prevents a depletion layer spreading from a lower end of the base region 14 from affecting the P + -like collector area 22 and the N + -like cathode region 82. In the present description, the depth position of the upper end of the buffer region 20 is defined as Zf. The depth position Zf may be a position where the doping concentration is higher than the doping concentration of the drift region 18.

[0164] In transistor section 70, the P +-type collector region 22 is arranged below the buffer region 20. The acceptor concentration of the collector region 22 is higher than the acceptor concentration of the base region 14. The collector region 22 may contain the same acceptor as the base region 14 and may contain a different acceptor. The acceptor of the collector region 22 is, for example, boron.

[0165] In the diode section 80, the N +-like cathode region 82 is arranged below the buffer region 20. The donor concentration of the cathode region 82 is higher than the donor concentration of the drift region 18. The donor of the cathode region 82 is, for example, hydrogen or phosphorus. Elements that become donors and acceptors in each region are not limited to the examples described above. The collector region 22 and the cathode region 82 are exposed on the lower surface 23 of the semiconductor substrate 10 and are connected to the collector electrode 24. The collector electrode 24 may be in contact with the entire lower surface 23 of the semiconductor substrate 10. The emitter electrode 52 and the collector electrode 24 are formed of a metal such as aluminum.

[0166] One or more gate trench portions 40 and one or more dummy trench portions 30 are arranged on the upper surface 21 side of the semiconductor substrate 10. Each trench portion penetrates from the upper surface 21 of the semiconductor substrate 10 into the base region 14 and reaches the drift region 18. In the region where at least one of the emitter region 12, the contact region 15, and the accumulation region 16 is arranged, each trench portion also penetrates these impurity regions and reaches the drift region 18. The trench portion penetrating the impurity region is not limited to those formed in the order of forming the impurity region and then forming the trench portion. A case where an impurity region is formed between the trench portions after the trench portion is formed is also included in a case where the trench portion penetrates the impurity region.

[0167] As described above, the transistor section 70 includes the gate trench section 40 and the dummy trench section 30. The diode section 80 includes the dummy trench section 30 and no gate trench section 40. In this example, the boundary between the diode section 80 and the transistor section 70 in the X-axis direction is the boundary between the cathode region 82 and the collector region 22.

[0168] The gate trench portion 40 includes a gate trench, a dielectric insulating film 42, and a gate conductive portion 44 disposed on the upper surface 21 of the semiconductor substrate 10. The gate insulating film 42 is disposed covering an inner wall of the gate trench. The gate insulating film 42 can be formed by oxidizing or nitriding the semiconductor on the inner wall of the gate trench. The gate conductive portion 44 is disposed on an inner side of the gate insulating film 42 in the gate trench. 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.

[0169] The gate line portion 44 may be provided longer in the depth direction than the base region 14. The gate trench portion 40 in cross section is covered by the interlayer dielectric film 38 on the upper surface 21 of the semiconductor substrate 10. The gate line portion 44 is electrically connected to the gate rotor 15. If a predetermined gate voltage is applied to the gate line portion 44, a channel is formed by an inversion layer of electrons in a surface layer of the interface in contact with the gate trench portion 40 of the base region 14.

[0170] The dummy trench portion 30 may have the same cross-sectional structure as the gate trench portion 40. The dummy trench portion 30 includes a dummy trench on the upper surface 21 of the semiconductor substrate 10, a dummy insulating film 32, and a dummy wiring portion 34. The dummy wiring portion 34 is electrically connected to the emitter electrode 52. The dummy insulating film 32 covers the inner wall of the dummy trench. The dummy wiring portion 34 is disposed in the dummy trench and on the inner side of the dummy insulating film 32. The dummy insulating film 32 insulates the dummy wiring portion 34 from the semiconductor substrate 10. The dummy wiring portion 34 may be formed of the same material as the gate wiring portion 44. For example, the dummy line portion 34 is formed from a conductive material such as polysilicon. The dummy line portion 34 may have the same length in the depth direction as the gate line portion 44.

[0171] The gate trench portion 40 and the dummy trench portion 30 of this example are covered by the interlayer dielectric film 38 on the upper surface 21 of the semiconductor substrate 10. Note that the bottom portions of the dummy trench portion 30 and the gate trench portion 40 may have a curved surface shape (a curved shape in cross section) that protrudes downward. In the present specification, the depth position of the bottom end of the gate trench portion 40 is defined as Zt.

[0172] The drift area 18 can Fig. 4 and the like. That is, the drift region 18 has a donor concentration that is mainly determined by a volume donor concentration and a hydrogen donor (VOH defect) concentration. In the drift region 18, the second peak 141 of the chemical hydrogen concentration C HA dopant is implanted locally in a different region than the drift region 18. Therefore, the doping concentration in these regions differs from that in Fig. 4 and the like described donor concentration D D .

[0173] Fig. Figure 13 is a graph showing an example of the distribution of carrier concentration in the depth direction at positions of the line FF in Fig. 12 illustrated. In Fig. 13 is also part of the distribution of the chemical hydrogen concentration C H The vertical axis in Fig. 13 is a logarithmic axis.

[0174] The carrier concentration distribution in the buffer region 20 of this example has a plurality of peaks 25 at different positions in the depth direction. Peak 25 is a donor concentration peak. Peak 25 may contain hydrogen as an impurity. By arranging the plurality of peaks 25, the depletion layer can be further suppressed from reaching the collector region 22. The second donor peak 111 may function as peak 25 in the buffer region 20.

[0175] For example, the second donor peak 111 may function as a peak that is farthest from the bottom surface 23 of the semiconductor substrate 10 among the plurality of peaks 25 of the buffer region 20. The flat portion 150 is disposed between the second donor peak 111 in the buffer region 20 and the first donor peak 121.

[0176] The second donor peak 111 may have a higher donor concentration than a peak 25 of the plurality of peaks 25 in the buffer region 20 that is furthest from the bottom surface 23 after the second donor peak 111. By increasing the concentration of the second donor peak 111, the flat portion 150 is easily formed. The distribution of the chemical hydrogen concentration C H may have one or more hydrogen concentration peaks 194 between depth position Z2 and the bottom surface 23. The hydrogen concentration peak 194 may be located in the buffer region 20. The hydrogen concentration peak 194 may be located at the same depth position as peak 25.

[0177] The accumulation region 16 of this example has a plurality of peaks 26. The peak 26 is a peak of the donor concentration. The first donor peak 121 of this example is located closer to the bottom surface 23 than the accumulation region 16. A region 180 with a doping concentration lower than that of the flat portion 150 may be arranged between the first donor peak 121 and the accumulation region 16. The doping concentration of the region 180 may be the bulk donor concentration N BO be.

[0178] Furthermore, the semiconductor device 100 may use a non-doped substrate in which the entire ingot has not been doped with a dopant such as phosphorus (P) during the production of the semiconductor ingot as the semiconductor substrate 10. In this case, a base doping concentration Dn of the region 180 is lower than the volume donor concentration N BO. The base doping concentration Dn is, for example, in the range of 1 × 10 10 atoms / cm 3 up to 5 × 10 12 atoms / cm 3 The base doping concentration Dn can be 1×10 11 atoms / cm 3 or more. The base doping concentration Dn can be 5×10 12 atoms / cm 3 or less.

[0179] Fig. 14 is a diagram showing an example of a cross section gg in Fig. 10. The Fig. The cross section shown in Figure 14 is an XZ plane including the edge termination structure portion 90 and the transistor portion 70. The outer circumferential gate slider 130 is disposed above the semiconductor substrate 10 between the edge termination structure portion 90 and the transistor portion 70. The outer circumferential gate slider 130 is disposed separately from the emitter electrode 52. The well region 11 is disposed in the upper surface 21 of the semiconductor substrate 10 between the edge termination structure portion 90 and the transistor portion 70. The well region 11 of this example is disposed in a region shallower than the trench portion. The structure of the transistor portion 70 is similar to that of the Fig. 10 to Fig. 12 described transistor section 70.

[0180] The edge termination structure portion 90 includes a plurality of guard rings 92, a plurality of field plates 94, and a channel stopper 174. In the edge termination structure portion 90, a collector region 22 may be disposed in a region connected to the lower surface 23. Each guard ring 92 may be disposed to encircle the active portion 160 in the upper surface 21. The plurality of guard rings 92 may serve to spread the depletion layer generated in the active portion 160 to the outside of the semiconductor substrate 10. As a result, an electric field strength within the semiconductor substrate 10 can be prevented, and the breakdown voltage of the semiconductor device 100 can be improved.

[0181] The guard ring 92 of this example is a P-type semiconductor region formed by ion implantation near the top surface 21. The depth of the lower portion of the guard ring 92 may be shallower than the bottom portions of the gate trench portion 40 and the dummy trench portion 30.

[0182] The upper surface of the guard ring 92 is covered with an interlayer dielectric film 38. The field plate 94 is formed of a conductive material such as metal or polysilicon. The field plate 94 may be formed of the same material as the emitter electrode 52. The field plate 94 is disposed on the interlayer dielectric film 38. The field plate 94 is connected to the guard ring 92 through a through-hole in the interlayer dielectric film 38.

[0183] The channel stopper 174 is arranged to be exposed to the top surface 21 and the side surface of the end face 162. The channel stopper 174 is an N-type region with a higher doping concentration than the drift region 18. The channel stopper 174 functions to terminate the depletion layer generated in the active portion 160 at the end face 162 of the semiconductor substrate 10.

[0184] In this example, the second peak 141 of the chemical hydrogen concentration is located between the bottom portion of the well region 11 and the lower surface 23 of the semiconductor substrate 10. The second peak 141 may also be located in the edge termination structure portion 90. The second peak 141 may also be located between the edge termination structure portion 90 and the transistor portion 70. The second peak 141 may be located in the entire XY plane of the semiconductor substrate 10. The second peak 141 may be located in the Fig. 1 and the like is formed from the lower surface 23 of the semiconductor substrate 10 to the second peak 141.

[0185] Fig. 15 is a graph showing another example of the cross section gg in Fig. 10. In this example, the sink region 11 is located lower than the lower ends of the gate trench section 40 and the dummy trench section 30. Other features may be similar to those of the example of Fig. 14. At least one trench section can be arranged in cross-section within the depression area 11. In the example of Fig. 15, a trench portion located closest to the edge termination structure portion 90 is disposed within the depression region 11.

[0186] The second peak 141 of this example is located between the lower end of the well region 11 and the upper surface 21 of the semiconductor substrate 10. In another example, the second peak 141 may be located between the lower end of the well region 11 and the lower surface 23 of the semiconductor substrate 10.

[0187] Fig. 16 is a diagram illustrating an example of a manufacturing method of the semiconductor device 100. The manufacturing method of this example includes a substrate preparation step S1600, a device manufacturing step S1606, a concentration measurement step S1602, and an implantation amount calculation step S1604.

[0188] In the substrate preparation step S1600, the semiconductor substrate 10 is prepared. The semiconductor substrate 10 is, for example, an MVZ substrate. In the concentration measurement step S1602, the chemical oxygen concentration C OXof the semiconductor substrate 10. In the concentration measurement step S1602, the chemical oxygen concentration can be measured using an FTIR (infrared absorption spectroscopy) method. In the concentration measurement step S1602, a substrate resistance value (Ω cm) of the semiconductor substrate 10 can also be measured.

[0189] In the implantation amount calculation step S1604, the implantation amount of the charged particle beam to be implanted at the depth position Z1 is calculated based on the chemical oxygen concentration measured in step S1602. As described above, the concentration of VOH defects to be formed can be controlled by the implantation amount of the charged particle beam. In the implantation amount calculation step S1604, the implantation amount of the charged particle beam can be calculated so that the substrate resistance value of the charged particle beam in Fig. 4 and the like, assumes a predetermined target resistance value. The target resistance value can be set by a manufacturer of the semiconductor device 100. The substrate resistance value of the flat portion 150 corresponds to the donor concentration of the flat portion 150 on a one-to-one basis. Therefore, the amount of increase in the donor concentration is determined so that the substrate resistance value of the flat portion 150 becomes the target resistance value. The relationship between the implantation amount of the charged particle beam, the amount of increase in the donor concentration, and the chemical oxygen concentration can be determined as shown in Fig. 6 and Fig. 7 can be experimentally acquired in advance. In implantation amount calculation step S1604, the implantation amount of the charged particle beam can be calculated based on the pre-acquired relationship.

[0190] By substituting the above-described expressions (11) and (12) into expression (2), we obtain expression (13). NVOH=c×(DH)d+e×(DH)f×COX

[0191] In addition, expressions (1) to (13) are represented by expression (14). NF−NB0=c×(DH)d+e×(DH)f×COX

[0192] The final doping concentration N F is a set value and the volume donor concentration N BO is known from the measured value or determination value of the semiconductor wafer. The chemical oxygen concentration C OX is known from the measurement in step S1602. The parameters c, d, e, and f can be determined experimentally in advance as described above. Thus, the variable of expression (14) is only the implantation amount of the charged particles (dosage D Hof hydrogen ions in this example). By numerically solving expression (14), the implantation amount of charged particles can be calculated. The implantation amount D obtained from expression (14) H of the charged particles may have a width (error) that reflects a fluctuation in the values ​​of the data when fitting expressions (8), (10), and (11). That is, if the implantation amount D H of the charged particles is, for example, in the range of ±50% with respect to the value obtained from Expression (13) or Expression (14), the implantation amount can be regarded as a value obtained from Expression (13) or Expression (14).

[0193] The device manufacturing step S1606 of this example includes a particle implantation step S1608, a hydrogen implantation step S1610, and a heat treatment step S1612. The device manufacturing step S1606 includes a step of forming each Fig. 10 to Fig. 12 described structure, was in Fig. 16, however, was omitted.

[0194] In the particle implantation step S1608, the charged particle beam is implanted from the lower surface 23 of the semiconductor substrate 10 so that it passes through half or more of the thickness of the semiconductor substrate 10 in the depth direction. In step S1608, hydrogen ions such as protons can be implanted as the charged particle beam. As a result, the Fig. 2, the second peak 141 is formed. In addition, the Fig. 2 distribution of the vacancy concentration N V The implantation amount of the charged particle beam in step S1608 is set to the implantation amount calculated in implantation amount calculation step S1604. As a result, it is possible to form vacancies having a concentration such that the flat portion 150 has a target resistance value.

[0195] In a case where the charged particle beam is an electron beam, the second peak 141 is not formed. Also in this case, vacancies are formed with a concentration corresponding to the implantation amount of the electron beam. Therefore, by adjusting the implantation amount of the electron beam, it is possible to form vacancies having a concentration such that the flat portion 150 has a target resistance value.

[0196] In the hydrogen implantation step S1610, hydrogen ions are implanted from the lower surface 23 of the semiconductor substrate 10. In step S1608, hydrogen ions are implanted into the Fig. 1 and the like. In the hydrogen implantation step S1610, hydrogen ions can be implanted at a concentration that can sufficiently occupy the vacancies formed in step S1608. The particle implantation step S1608 and the hydrogen implantation step S1610 are performed before the heat treatment step S1612.

[0197] In the heat treatment step S1612, the semiconductor substrate 10 is subjected to a heat treatment. The heat treatment temperature of the heat treatment step S1612 may be from 350°C to 380°C.

[0241] By such a method, the resistance value of the flat portion 150 can be adjusted to a target value even if the chemical oxygen concentration of the semiconductor substrate 10 fluctuates.

[0198] Before the particle implantation step S1608, a structure may be formed on the upper surface side 21 of the semiconductor substrate 10 in the Fig. 12 can be formed. The structure of the upper surface side 21 can include each trench portion, the emitter region 12, the base region 14, the accumulation region 16, the interlayer dielectric film 38, and the emitter electrode 52. The cathode region 82 and the collector region 22 can be formed before the particle implantation step S1608. In the particle implantation step S1608, the charged particles can be implanted to the depth position Z1 closer to the lower surface side 23 than the lower end of the gate trench portion 40. As a result, it is possible to suppress the influence of damage due to the implantation of the charged particle beam on the gate insulating film 42.

[0199] Fig. Figure 17 is a diagram illustrating another example of a calculation method for the implantation amount of the charged particle beam. The charged particle beam in this example consists of hydrogen ions. Fig. Figure 17 includes an upper diagram illustrating a relationship between the chemical oxygen concentration and the amount of increase in the donor concentration, and a lower diagram illustrating a relationship between the dosage of hydrogen ions and the inverse of the chemical oxygen concentration. The upper diagram is an enlarged view of a portion of the diagram in Fig. 6.

[0200] In the upper diagram, the target value of the increase in the donor concentration is indicated by a chain line. The target increase is an increase amount for setting the resistance value of the flat portion 150 to a target value. From the upper diagram, the chemical oxygen concentration corresponding to the target increase amount is obtained on lines 602 and 603. The lower diagram plots the relationship between the inverse of the chemical oxygen concentration obtained from the upper diagram and the dosage of hydrogen ions on lines 602 and 603. In the lower diagram, the relationship between the inverse of the chemical oxygen concentration and the dosage of hydrogen ions is approximated by a curve.

[0201] If the relationship shown in the diagram below is acquired in advance, the dosage of hydrogen ions to be implanted can be calculated from the chemical oxygen concentration of the semiconductor substrate 10 used to manufacture the semiconductor device 100. For example, in a case where the chemical oxygen concentration of the semiconductor substrate 10 is 3.7 × 10 17 atoms / cm 3 is, the reciprocal of the chemical oxygen concentration is 2.8 × 10 -18 From the relationship in the lower diagram, the dosage of hydrogen ions, the 2.8 × 10 -18 corresponds to 4.2 × 10 12 ions / cm 2The relationship in the lower diagram changes according to the target increase in the donor concentration. Thus, if the relationship in the lower diagram is determined in advance, the dosage of hydrogen ions to be implanted can be calculated from the target increase in the donor concentration and the chemical oxygen concentration.

[0202] Fig. Fig. 18 is a diagram illustrating another example of the manufacturing method of the semiconductor device 100. The manufacturing method of this example includes, compared to that in Fig. 16 further comprises an oxygen introduction step S1802. The other steps are identical to those of the example of Fig. 16. The oxygen introduction step S1802 is performed before the particle implantation step S1608.

[0203] In oxygen introduction step S1802, oxygen is introduced into the semiconductor substrate 10. In the oxygen introduction step S1802, oxygen can be introduced into the semiconductor substrate 10 by heat-treating the semiconductor substrate 10 in an atmosphere containing oxygen. In the oxygen introduction step S1802, oxygen can be introduced so that the chemical oxygen concentration of the semiconductor substrate 10 falls within a predetermined range. By introducing oxygen into the semiconductor substrate 10, the donor concentration of the flat portion 150 can be easily adjusted. For example, the chemical oxygen concentration can be increased by introducing oxygen into the semiconductor substrate 10, even in a case where the chemical oxygen concentration in the semiconductor substrate 10 prepared in step S1600 is low and the VOH defects are not sufficiently formed.

[0204] The chemical oxygen concentration introduced in the oxygen introduction step S1892 (referred to as an introduction concentration) may be greater than the chemical oxygen concentration (referred to as an original concentration) of the semiconductor substrate 10 before the oxygen introduction step S1802. Since the introduction concentration can be precisely controlled by the above-described conditions such as the temperature and duration of the heat treatment and the oxygen concentration of the atmosphere, the fluctuation of the chemical oxygen concentration of the semiconductor substrate 10 can be reduced by making the ratio of the introduction concentration greater than the original concentration. The introduction concentration may be 2 times or more, 5 times or more, or 10 times or more of the original concentration.

[0205] In the oxygen introduction step S1802, oxygen may be introduced into the semiconductor substrate 10 according to the chemical oxygen concentration measured in the concentration measurement step S1802. For example, oxygen may be introduced so that the sum of the chemical oxygen concentration introduced in the oxygen introduction step S1802 and the chemical oxygen concentration measured in the concentration measurement step S1602 becomes a predetermined target value. In the concentration measurement step S1602, the chemical oxygen concentration of the semiconductor substrate 10 is measured after performing the oxygen introduction step S1802. In this case, the implantation amount of the charged particle beam can be calculated more accurately.

[0206] Fig. 19 is a diagram illustrating an example of the device manufacturing step S1606. The device manufacturing step S1606 of this example includes an upper surface side process S1902 and a lower surface side process S1904. The upper surface side process S1902 is a step of forming a structure on the upper surface side 21 of the semiconductor substrate 10. The structure on the upper surface side 21 includes, for example, a trench portion, the emitter region 12, the base region 14, the accumulation region 16, the well region 11, the emitter electrode 52, the gate mover, the guard ring 92, the field plate 94, the channel stopper 174, and the interlayer dielectric film 38. The lower surface side process S1904 is a step of forming a structure on the lower surface side 23 of the semiconductor substrate 10.The structure on the lower surface side 23 includes, for example, the cathode region 82, collector region 22, buffer region 20 and collector electrode 24.

[0207] In this example, in the upper surface side process S1902, the oxygen introduction step S1802 is performed. The oxygen introduction step S1802 may also serve as a heat treatment step performed to form a structure on the upper surface side 21. For example, the oxygen introduction step S1802 may be a heat treatment step performed after a dopant is implanted into the emitter region 12, the base region 14, or the accumulation region 16.

[0208] In this example, in the lower surface side process S1904, a particle implantation step S1608, a hydrogen implantation step S1610, and a heat treatment step S1612 are performed. The hydrogen implantation step S1610 and the heat treatment step S1612 may be part of the process for forming the buffer region 20. That is, in the hydrogen implantation step S1610, any peak 25 of the buffer region 20 may be formed. The heat treatment step S1612 may be performed after hydrogen ions are implanted into the positions of the plurality of peaks 25 in the buffer region 20.

[0209] Fig. Fig. 20 is a diagram illustrating another example of the device manufacturing step S1606. The device manufacturing step S1606 of this example differs from the example of Fig. 19, the oxygen introduction step S1802 is performed in the lower surface side process S1904. The other steps are identical to those of the example of Fig. 19. The oxygen introduction step S1802 may also serve as a heat treatment step performed to form a structure on the lower surface side 23. For example, the oxygen introduction step S1802 may be a heat treatment step performed after a dopant is implanted into the collector region 22 or the cathode region 82. In this example, the oxygen introduction step S1802 is also performed before the hydrogen implantation step S1610. The oxygen introduction step S1802 may be performed before the particle implantation step S1608.

[0210] Fig. 21 is a graph illustrating a relationship between the oxygen contribution ratio ξ and the depth position Z1 at which the second peak 141 is located. Fig. Figure 21 illustrates the relationship between the oxygen contribution fraction ξ and the depth position Z1 for each of the cases where the hydrogen ion dosage is 3 × 10 14 ions / cm 2 , 1 × 10 14 ions / cm 2 , 3 × 10 13 ions / cm 2 , 1 × 10 13 ions / cm 2 , 3 × 10 12 ions / cm 2 or 1 × 10 12 ions / cm 2 amounts.

[0211] As in Fig. 8 and Fig. As shown in Figure 9, the oxygen contribution fraction ξ changes depending on the depth position Z1. Each value represented by a square in Fig. 8 plotted point is at a depth position of Z1 = 100 µm in Fig. 21. Each by a square in Fig. 9 plotted point is at a depth position of Z1 = 50 µm in Fig. 21. In addition, a plot at a depth position of Z1 = 150 µm is plotted in Fig. 21. The plots in Fig. 21 are used for the dosages of hydrogen ions of 3 × 10 14 ions / cm 2 , 1 × 10 14 ions / cm 2 and 1 × 10 12 ions / cm 2 omitted. A line obtained by linear approximation (the horizontal axis is linear and the vertical axis is a conventional logarithm) of these plots for each dosage of hydrogen ions by the least squares method is shown in Fig. 21 is indicated by a thick line. The oxygen contribution fraction ξ can decrease exponentially with respect to the depth position Z1.

[0212] The oxygen contribution fraction ξ can be calculated from the Fig. 21 shown relationship, the dosage D Hof hydrogen ion with respect to the depth position Z1 and the depth position Z1 can be detected. The depth position Z1 can be measured from a peak position in the distribution of the chemical hydrogen concentration of the semiconductor device 100. The dosage D H can be measured by integrating the chemical hydrogen concentration in the depth direction with respect to a mountain-shaped peak of the chemical hydrogen concentration with a vertex at the depth direction Z1. The integration range can, for example, be the full width (FW 10% M) at 10% of the peak value of the chemical hydrogen concentration. Alternatively, the dosage D H be a value obtained by multiplying the peak chemical hydrogen concentration by the full width at half maximum (FWHM).

[0213] For example, in Fig. 21 a case where the depth position Z1 is 120 µm and the dosage D H 5 × 1012 ions / cm 2 indicated by a black circle. In this case, the oxygen contribution fraction ξ is approximately ξ = 1.2 × 10 -4 . The value is defined as ξ1.

[0214] Fig. Figure 22 is a graph showing a relationship between the vacancy concentration N V and the depth position Z1 at which the second peak 141 is located. Fig. Figure 21 illustrates the relationship between the oxygen contribution fraction ξ and the depth position Z1 for each of the cases where the hydrogen ion dosage is 3 × 10 14 ions / cm 2 , 1 × 10 14 ions / cm 2 , 3 × 10 13 ions / cm 2 , 1 × 10 13 ions / cm 2 , 3 × 10 12 ions / cm 2 or 1 × 10 12 ions / cm 2 As in Fig. 8 and Fig. 9, the vacancy concentration N Valso depends on the depth position Z1, similar to the oxygen contribution fraction ξ. The diagram of the Fig. 22 is generated in the same way as the diagram of the Fig. 21. The vacancy concentration N V can decrease exponentially with respect to the depth position Z1.

[0215] The vacancy concentration N V can be found in Fig. 22 shown relationship, the dosage D H of hydrogen ion with respect to the depth position Z1 and the depth position Z1. The depth position Z1 and the dosage D H can, as in Fig. 21 described from the semiconductor device 100. For example, in Fig. 22 a case where the depth position Z1 is 120 µm and the dosage D H 5 × 10 12 ions / cm 2 indicated by a black circle. In this case, the vacancy concentration is N V approximately N v= 7 × 10 12 ions / cm 3 . The value is defined as N V 1.

[0216] From the calculated oxygen contribution fraction ξ1 and the vacancy concentration N V 1 and the chemical oxygen concentration C OX can be a concentration of VOH defects N VOH 1 (a first value of the hydrogen donor concentration) can be calculated using expression (2). That is, the first value N VOH 1 is the sum of the value obtained by multiplying the chemical oxygen concentration in the flat section 150 by the oxygen contribution fraction and the vacancy concentration of the flat section 150. The chemical oxygen concentration C OX can be obtained by measuring the chemical oxygen concentration in the semiconductor device 100. For example, in the case where the chemical oxygen concentration C OX 2 × 10 17 atoms / cm 3is, the concentration of VOH defects N VOH 1 as follows. NVOH1=7×1012+1.2×10−4×2×1017=3.1×1013(atoms / cm3)

[0217] On the other hand, the concentration of VOH defects N VOH can be measured from the properties of the semiconductor device 100. For example, the concentration of VOH defects N VOH from the difference (N F - N BO ) between the final doping concentration N F and the volume donor concentration N BO The final doping concentration N F and the volume donor concentration N BO can be measured from the semiconductor device 100. The measured concentration of VOH defects N VOH is called N VOH 2 (a second value of the hydrogen donor concentration). That is, the second value N VOH2 is a difference obtained by subtracting the bulk donor concentration from the donor concentration of the flat portion 150. For example, in a case where the final doping concentration N F 7 × 10 13 (Atoms / cm 3 ) and the volume donor concentration N BO 2 × 10 12 (Atoms / cm 3 ), the concentration of VOH defects is N VOH 2 as follows. NVOH2=7×1013−2×1012=6.8×1013(atoms / cm3)

[0218] If the calculated value N VOH 1 sufficient with the actually measured value N VOH 2, it can be determined that the calculated oxygen contribution fraction ξ1 and the calculated vacancy concentration N V 1 are essentially correct. That is, if the ratio N VOH 1 / N VOH 2 of the first value N VOH 1 of the hydrogen donor concentration to the second value NVOH 2 of the hydrogen donor concentration 0.1 ≤ N VOH 1 / N VOH 2 ≤ 10, it can be assumed that they are sufficiently consistent. In the case of the above example, N VOH 1 / N VOH 2 = 3.1 × 10 13 / 6. 8 × 10 13 ≈ 0.46, and the calculated oxygen contribution fraction ξ1 and the vacancy concentration N V 1 are correct.

[0219] If the calculated value N VOH 1 sufficient with the actually measured value N VOH 2 or the calculated oxygen contribution ratio ξ1 is within a predetermined range, it can be determined that the oxygen contribution ratio ξ of the semiconductor device 100 is within the range. In addition, if the calculated value N VOH 1 sufficient with the actually measured value N VOH 2 or the calculated vacancy concentration N V1 falls within a given range, it can be determined that the vacancy concentration N V 1 of the semiconductor device 100 falls within the range.

[0220] In case that 0.2 ≤ N VOH 1 / N VOH 2 ≤ 5, it can be determined that the calculated value N VOH 1 sufficient with the actually measured value N VOH 2. In the case that 0.3 ≤ N VOH 1 / N VOH 2 ≤ 3, it can be determined that the calculated value N VOH 1 sufficient with the actually measured value N VOH 2. In the case that 0.5 ≤ N VOH 1 / N VOH 2 ≤ 2, it can be determined that the calculated value N VOH 1 sufficient with the actually measured value N VOH 2matches.

[0221] In Expression (1), the contribution of oxygen in generating hydrogen donors (VOH defects) was considered. However, as described later, the contribution of carbon in hydrogen donor formation cannot be ignored if the depth position Z1 is shallow, if the hydrogen ion dosage is high, or if the chemical carbon concentration is high.

[0222] The ratio of the chemical carbon concentration contributing to the doping concentration of hydrogen donors to the chemical carbon concentration is called the carbon contribution fraction η. The carbon contribution fraction η can be regarded as a ratio of a chemical concentration of carbon atoms contributing to the formation of a hydrogen donor among chemical concentrations of all carbon atoms in a given range (e.g., a depth position from the bottom surface or the top surface). The carbon contribution fraction η can be a ratio of the increased amount of the concentration of hydrogen donors to the increased amount of the chemical carbon concentration in a case where the chemical carbon concentration is increased. The carbon contribution fraction can be a value from 0.01% to 10% (i.e., from 0.0001 to 0.1).

[0223] Fig. 23 illustrates the distribution in depth direction after heat treatment of the chemical hydrogen concentration C H , the chemical oxygen concentration C OX , the chemical carbon concentration C C , the contributing oxygen concentration N OH , the contributing carbon concentration N C and the concentration of VOH defects at the positions that are Fig. 1 by the line AA. The structure and manufacturing method of the semiconductor device 100 of this example except for the chemical carbon concentration C C and the contributing carbon concentration N C are similar to those in Fig. 1 to Fig. 22 described semiconductor device 100.

[0224] Carbon is often introduced during the manufacture of an ingot and is often evenly distributed within the semiconductor substrate 10. The chemical carbon concentration CC may be uniform throughout the semiconductor substrate 10. In another example, the chemical carbon concentration C C from the lower surface 23 to the upper surface 21 of the semiconductor substrate 10 monotonically increase or decrease. In addition, carbon near the upper surface 21 or the lower surface 23 of the semiconductor substrate 10 can be released from the semiconductor substrate 10 to the outside. Similar to the chemical oxygen concentration C OX the chemical carbon concentration C C monotonically decrease towards the upper surface 21 and the lower surface 23 in the vicinity of the upper surface 21 and the lower surface 23. Apart from the vicinity of the upper surface 21 and the lower surface 23, the chemical carbon concentration C C be uniform as described above and can increase or decrease monotonically.

[0225] The chemical carbon concentration C C in the flat section 150 can be in the range of 1 × 10 13 / cm 3 up to 1 × 10 16 / cm 3 The chemical carbon concentration C C can 1 × 10 14 atoms / cm 3 or more. The chemical carbon concentration C C can 5 × 10 15 atoms / cm 3 or less or 2 × 10 15 atoms / cm 3 or less. The chemical carbon concentration C C may be smaller than the chemical oxygen concentration C OX . The chemical carbon concentration C C may be 1 / 100 or less or 1 / 1000 or less of the chemical oxygen concentration C OX In a case where the chemical carbon concentration C Cof the semiconductor substrate 10 in the present specification, everything between the second peak 141 and the first peak 133 satisfies the definition of chemical carbon concentration unless otherwise stated. Everything between the second peak 141 and the bottom surface 23 may satisfy the definition of chemical carbon concentration, and the entire semiconductor substrate 10 may satisfy the definition of chemical carbon concentration.

[0226] The contributing carbon concentration N C refers to the concentration of carbon that contributes to the formation of VOH defects. It has been experimentally confirmed that the concentration of VOH defects can also change if the chemical carbon concentration C C In this description, the relationship between the contributing carbon concentration N C and the chemical carbon concentration C Cdefined as a carbon contribution fraction η. That is, η = N C / C C . The carbon contribution fraction η lies in the range from 0 to 1. The unit of the carbon contribution fraction η is a dimensionless quantity.

[0227] The distribution of the contributing carbon concentration N C can be similar to the distribution of the chemical carbon concentration C C For example, the contributing carbon concentration N C in the depth direction of the semiconductor substrate 10 and can increase or decrease monotonically. The contributing carbon concentration N C may have a concentration distribution with a peak at a given depth position.

[0228] The range and the like of the carbon contribution ratio η in the semiconductor substrate 10 are described below. The concentration of VOH defects N VOHis defined by expression (2a) in a case when the carbon contribution fraction η is taken into account. NVOH=NV+ξCOX+ηCc

[0229] The concentration of VOH defects N VOH in expression (2a) is obtained by adding the product of the chemical carbon concentration C C and the carbon contribution fraction η with the concentration of VOH defects N described in expression (2) VOH This means that the hydrogen donor concentration increased by the presence of carbon becomes the concentration of VOH defects N VOH in expression (2). Incidentally, the hydrogen donor formed by the contribution of carbon is not limited to the VOH defect, but in expression (2a), the hydrogen donor concentration increased by the presence of carbon is in the concentration of the VOH defects N VOH contain.

[0230] Fig. Figure 24 is a graph showing a relationship between the amount of increase in donor concentration and the chemical carbon concentration C C The method for measuring the increase in donor concentration in this example is similar to that in the example of Fig. 6. In addition, the chemical carbon concentration C C of the semiconductor substrate 10 of this example is uniformly distributed in the depth direction.

[0231] The depth position Z1 in Fig. 24 is 50 µm. In this example, the dosage of hydrogen ions to the depth position Z1 is calculated for three types of 3 × 10 12 ions / cm 2 , 1 × 10 13 ions / cm 2 and 3 × 10 13 ions / cm 2 shown. In Fig. 6 shows an example where the dosage of hydrogen ions is 3 × 10 13 ions / cm 2is approximated by a straight line 621, an example where the Do of the hydrogen ions is 1 × 10 13 ions / cm 2 is approximated by a straight line 622 and an example where the dosage of hydrogen ions is 3 × 10 12 ions / cm 2 is approximated by a straight line 623. The size of each plot indicates the magnitude of the chemical oxygen concentration C OX As in Fig. 24, the increase in donor concentration is essentially (almost) proportional to the chemical carbon concentration C C increases linearly.

[0232] In the example of Fig. 24, an approximate expression is calculated in which the relationship between the chemical carbon concentration C C and the increase in the donor concentration is approximated by a straight line. In Fig. 24 shows an example where the dosage of hydrogen ions is 3 × 10 13 ions / cm 2is approximated by a straight line 611, an example where the Do of the hydrogen ions is 1 × 10 13 ions / cm 2 is approximated by a straight line 612 and an example where the dosage of hydrogen ions is 3 × 10 12 ions / cm 2 is approximated by a straight line 613.

[0233] Each line is described by expression (8a). NVOH=a×Cc+b

[0234] At this time, the slope a and intercept b of each line calculated by least squares fitting are as follows. Line 611: a=5.00851×10−2, b=6.46656×1013 Line 612: a=2.35891×10−2, b=4.14509×1013 Line 613: a=7.13212×10−3, b=2.26076×1013

[0235] As in Fig. 24, there is not necessarily a strong correlation between the chemical carbon concentration C Cand the chemical oxygen concentration C OX . When comparing the semiconductor substrate 10 with the same chemical carbon concentration C C of the same order of magnitude as in a case where the chemical carbon concentration C C 6 × 10 14 atoms / cm 3 and a case where the chemical carbon concentration C C 8 × 10 14 atoms / cm 3 the chemical oxygen concentration is C OX (the size of the plot) for example 9 × 10 15 atoms / cm 3 and 2.4 × 10 17 atoms / cm 3 , and a difference of 10 times or more. On the other hand, the chemical carbon concentration C C in the semiconductor substrate 10 with a chemical carbon concentration C C of 2.5 × 10 15 atoms / cm 3about three times higher than in the example above. However, the chemical oxygen concentration is C OX of the semiconductor substrate 10 4 × 10 17 Atoms / cm 3 and the difference in chemical oxygen concentration C OX is not that big.

[0236] Fig. Figure 25 is a graph showing a relationship between the amount of increase of a donor concentration and the chemical oxygen concentration C OX In this example, each Fig. 24 plot shown in a group with small chemical carbon concentration C C (1 × 10 15 atoms / cm 3 or less) and a group with a large chemical carbon concentration C C (2 × 10 15 atoms / cm 3 or more), and the relationship between the amount of increase in donor concentration and the chemical oxygen concentration C OX is approximated by a straight line.

[0237] In Fig. 5, for a plot where the dosage of hydrogen ions is 3 × 10 13 ions / cm 2 is the group with high chemical carbon concentration C C indicated by a straight line 631, and the group with small chemical carbon concentration C C is indicated by a straight line 641. In a plot where the dosage of hydrogen ions is 1 × 10 13 ions / cm 2 is the group with high chemical carbon concentration C C indicated by a straight line 632, and the group with small chemical carbon concentration C C is indicated by a straight line 642. In a plot where the dosage of hydrogen ions is 3 × 10 12 ions / cm 2 is the group with high chemical carbon concentration C C indicated by a straight line 633, and the group with small chemical carbon concentration CC is indicated by a straight line 643.

[0238] In a case that each line is described by expression (8), the slope a and the intercept b of each line are as follows. Line 631: a=3.64419×10−4, b=4.15739×1013 Line 641: a=2.80673×10−4, b=4.15739×1013 Line 632: a=2.04534×10−4, b=2.21483×1013 Straight line 642: a=1.67965×10−4, b=2.21483×1013 Line 633: a=8.60908×10−5, b=8.32518×1013 Line 643: a=8.05915×10−5, b=8.32518×1013

[0239] In a case where the dosage of hydrogen ions is equal, the vacancy concentration N V than the intercept essentially (almost) the same value. As in the case of oxygen, it is obvious to consider that the vacancy concentration N V approaches a constant value if the chemical carbon concentration approaches 0.

[0240] The slope of the straight line of the group with high chemical carbon concentration C C is greater than the slope of the straight line of the group with small chemical carbon concentration C C . That is, with increasing chemical carbon concentration C C the increase in the donor concentration increases. Furthermore, the increase in the donor concentration increases with increasing hydrogen ion dosage. That is, the ratio (the slope of the straight line) of the increase in the concentration of VOH defects N VOH to the amount of increase in the chemical oxygen concentration C OX is big.

[0241] The slope a of each line is the oxygen contribution fraction ξ. It is assumed that the oxygen contribution fraction ξ increases α-fold when the chemical carbon concentration C C increases. Note that α ≥ 1. The concentration of VOH defects N VOHis described by expression (2b). NVOH=NV+αξCOX

[0242] Expression (15) is obtained by comparing expression (2a) with expression (2b). ξCOX+ηCc=αξCOXη=(α−1)(COX / Cc)ξ

[0243] As shown in expression (15), the carbon contribution fraction η is given by the product of the chemical oxygen concentration C OX / C C per unit of chemical carbon concentration and (α - 1)ξ, which is an increment of ξ. That is, the carbon contribution fraction η depends on the chemical oxygen concentration C OX and the oxygen contribution fraction ξ.

[0244] As in Fig. 25, if the dosage of hydrogen ions is the same and the chemical carbon concentration C C increases, the ratio (the slope ξ of the straight line) of the increase in the concentration of VOH defects N VOHto the amount of increase in the chemical oxygen concentration C OX by α-fold and the contributing carbon concentration N C = ηC C increases. That is, the contributing carbon concentration N C not only increases the chemical carbon concentration C C , but also the effect of the chemical oxygen concentration C OX and the concentration of VOH defects N VOH decreases by the increased contributing carbon concentration N C This suggests that the donor enhanced by the carbon contribution may be a donor separate from the VOH defect. In the present description, the donor enhanced by the carbon contribution may be referred to as a VOH-C defect.

[0245] Fig. Figure 26 is a graph illustrating a relationship between the hydrogen ion dosage to the depth position Z1 and the carbon contribution fraction η. In Fig. 26 shows three types of properties of the semiconductor substrate 10, where the depth position Z1 is 50 µm, 100 µm, and 150 µm. In this example, the carbon contribution ratio η is similar to the oxygen contribution ratio ξ in Fig. 8 by a power function with respect to the dosage D H of the hydrogen ions to the depth position Z1.

[0246] In Fig. 26, the relationship between the dosage of hydrogen ions and the carbon contribution ratio η in the case of Z1 = 50 μm is approximated by a curve 811, the relationship between the dosage of hydrogen ions and the carbon contribution ratio η in the case of Z1 = 100 μm is approximated by a curve 812, and the relationship between the dosage of hydrogen ions and the carbon contribution ratio η in the case of Z1 = 150 μm is approximated by a curve 813. Each curve 801 is described by Expression (15). Here, the coefficients g and h of each curve are as follows. η=g×(DH)h Curve 811:g=2.57839×10−13,h=7.95528×10−1 Curve 812:g=1.35314×10−21,h=1.38598 Curve 813:g=3.49381×10−31,h=2.07102

[0247] Fig. Figure 27 is a graph showing a relationship between the oxygen contribution fraction ξ and the dosage D Hof the hydrogen ions in the group with low chemical carbon concentration C C illustrated. The group with low chemical carbon concentration C C is a group with a chemical carbon concentration C C of 1 × 10 15 atoms / cm 3 or less, as in Fig. 25. Since the oxygen contribution fraction ξ is a value in a case where there is no influence of the chemical carbon concentration, the value of the group with small chemical carbon concentration C C used. In Fig. 27, three types of characteristics of the semiconductor substrate 10 with the depth positions Z1 of 50 µm, 100 µm and 150 µm are respectively represented by a curve 821, a curve 822 and a curve 823. Each curve is a curve obtained by approximating each plot by a power function similar to the example of Fig. 8 results.

[0248] Fig. Figure 28 is a graph showing a relationship between the vacancy concentration N V and the dosage DH of hydrogen ions in the group with low chemical carbon concentration C C illustrated. Since the vacancy concentration N V is also a value in a case where there is no influence of the chemical carbon concentration, a value of the group with small chemical carbon concentration C C used. In Fig. 28, three types of properties of the semiconductor substrate 10 with the depth positions Z1 of 50 µm, 100 µm and 150 µm are respectively represented by a curve 831, a curve 832 and a curve 833. Each curve is a curve obtained by approximating each plot by a power function similar to the example of Fig. 8 results.

[0249] Each in Fig. 26 to Fig. The curve described in Figure 28 is expressed as a power function. Thus, expression (2a) becomes expression (13a). NVOH=c×(DH)d+e×(DH)f×COX+g×(DH)h×Cc

[0250] In addition, expressions (1) to (13a) are represented by expression (14). NF−NB0=c×(DH)d+e×(DH)f×COX+g×(DH)h×Cc

[0251] The final doping concentration N F is a set value and the volume donor concentration N BO is known from the measured value or determination value of the semiconductor wafer. The chemical oxygen concentration C OX and the chemical carbon concentration C C are obtained by measuring each concentration in the semiconductor substrate 10 by a SIMS method or the like. The parameters c, d, e, f, g, and h can be determined experimentally in advance. Therefore, the variables of expression (14a) are only the implantation amount of the charged particles (the dosage D Hof hydrogen ions in this example) and the right-hand side of expression (14a) is a constant that does not change with respect to the implantation amount.

[0252] By numerically solving expression (14a), the dosage of the charged particles to be implanted into the semiconductor substrate 10 can be determined with respect to the control value of the final doping concentration N F The implantation quantity D obtained from expression (14a) H of the charged particles may have a width (error) that may cause a fluctuation in the values ​​of the data when fitting in Fig. 26 to Fig. 28. That is, if the implantation quantity D H of the charged particles is, for example, within a range of ±50% with respect to the value obtained from Expression (13a) or Expression (14a), it can be regarded as a value obtained from Expression (13a) or Expression (14a).

[0253] Fig. 29 is a diagram illustrating an example of a manufacturing method of the semiconductor device 100. The manufacturing method of this example is different from that in Fig. 18 by the processes in the concentration measuring step S1602 and the implantation amount calculating step S1604. The processes except for the concentration measuring step S1602 and the implantation amount calculating step S1604 are similar to the example of Fig. 18.

[0254] The concentration measurement step S1602 of this example differs from the concentration measurement step S1602 in Fig. 18, that in addition to the chemical oxygen concentration C OX furthermore the chemical carbon concentration C C of the semiconductor substrate 10. The other points are similar to the example, which in connection with Fig. 18. In the concentration measurement step S1602, each concentration can be measured by an FTIR (infrared absorption spectroscopy) method.

[0255] In the implantation amount calculation step S1604, the implantation amount of the charged particle beam to be implanted at the depth position Z1 is calculated based on the chemical oxygen concentration and chemical carbon concentration measured in step S1602. As described above, the concentration of VOH defects to be formed can be controlled by the implantation amount of the charged particle beam. In the implantation amount calculation step S1604, the implantation amount can be calculated based on Expression (13a) or Expression (14a). In step S1604, the implantation amount of the charged particle beam can be calculated so that the Fig. 4 described substrate resistance value of the flat portion 150 and the like to the control value of the final doping concentration N F becomes.

[0256] By such a method, the resistance value of the flat portion 150 can be set to a target value even if the volume donor concentration of the semiconductor substrate 10 fluctuates. At a given depth position of the semiconductor substrate 10, the concentration of hydrogen donors to be generated is N VOH 1 and the concentration of hydrogen donors actually produced is N VOH2 The depth position may be in the flat portion 150. The depth position may be the center in the depth direction of the semiconductor substrate 10.

[0257] The concentration of hydrogen donors to be generated N VOH , is represented by expression (2a) to expression (16). N VOH1expressed by expression (16) is an example of a third value. NVOH1=NV+ξCOX+ηCc

[0258] As described above, the vacancy concentration N V , the chemical oxygen concentration C OX , the chemical carbon concentration C C , the oxygen contribution fraction ξ and the carbon contribution fraction η can be determined by measuring the semiconductor substrate 10. In addition, the actually generated hydrogen donor concentration N VOH 2 from a difference between the donor concentration N BO of the semiconductor substrate 10 before treatment by the manufacturing process described above and the donor concentration N F of the semiconductor substrate 10 after treatment. The difference in donor concentration can be measured in the flat section 150. The donor concentration N BOof the semiconductor substrate 10 is obtained by measuring SIMS or SR. According to SIMS, the donor concentration N BO of the semiconductor substrate 10 can also be obtained from the semiconductor substrate 10 after the treatment.

[0259] Fig. Figure 30 is a graph showing another example of the relationship between the vacancy concentration N V and the depth position Z1. Fig. Figure 30 illustrates the relationship between the oxygen contribution fraction ξ and the depth position Z1 for each of the cases where the hydrogen ion dosage is 3 × 10 14 ions / cm 2 , 1 × 10 14 ions / cm 2 , 3 × 10 13 ions / cm 2 , 1 × 10 13 ions / cm 2 , 3 × 10 12 ions / cm 2 or 1 × 10 12 ions / cm 2 The relationship in this example is essentially (almost) the same as that in Fig. 22. As an example, Fig. 30 a case where the depth position Z1 is 120 µm and the dosage D H 5 × 10 12 ions / cm 2 represented by a black square. In this case, the vacancy concentration is N V approximately N V = 6 × 10 12 ions / cm 3 . The value is defined as N V 2. As another example, Fig. 30 a case where the depth position Z1 is 23 µm and the dosage D H 3 × 10 12 ions / cm 2 represented by a black diamond. In this case, the vacancy concentration is N V approximately N V = 1.3 × 10 13 ions / cm 3 . The value is defined as N V 2.

[0260] Fig. Figure 31 is a graph illustrating another example of the relationship between the oxygen contribution fraction ξ and the depth position Z1. Fig. Figure 31 illustrates the relationship between the oxygen contribution fraction ξ and the depth position Z1 for each of the cases where the hydrogen ion dosage is 3 × 10 14 ions / cm 2 , 1 × 10 14 ions / cm 2 , 3 × 10 13 ions / cm 2 , 1 × 10 13 ions / cm 2 , 3 × 10 12 ions / cm 2 or 1 × 10 12 ions / cm 2 In this example, the dependence of the oxygen contribution fraction ξ with respect to the depth position Z1 is smaller in this example, where the dosage of hydrogen ions is high, than in the example in Fig. 21. That is, the slope of the Fig. 31 is small. As an example, Fig. 31 a case where the depth position Z1 is 120 µm and the dosage D H 5 × 10 12 ions / cm 2represented by a black square. In this case, the oxygen contribution fraction ξ is approximately ξ = 1.1 × 10 -4 . The value is defined as ξ2. As another example, Fig. 31 a case where the depth position Z1 is 23 µm and the dosage D H 3 × 10 12 ions / cm 2 represented by a black diamond. In this case, the oxygen contribution fraction ξ is approximately ξ = 1.0 × 10 -4 . The value is defined as ξ3.

[0261] Fig. Figure 32 is a graph illustrating a relationship between the carbon contribution fraction η and the depth position Z1. Similar to the oxygen contribution fraction ξ, the carbon contribution fraction η decreases exponentially with increasing depth position Z1. The dependence of the carbon contribution fraction η on the depth position Z1 is strong. Furthermore, the carbon contribution fraction η increases if the dosage of hydrogen ions is high. As an example, Fig. 32 a case where the depth position Z1 is 120 µm and the dosage D H 5 × 10 12 ions / cm 2 represented by a black square. In this case, the carbon contribution fraction η is approximately η = 2.1 × 10 -4 . The value is defined as η2. As another example, Fig. 32 a case where the depth position Z1 is 23 µm and the dosage D H 3 × 10 12 ions / cm 2represented by a black diamond. In this case, the carbon contribution fraction η is approximately η = 3.0 × 10 -3 . The value is defined as η3.

[0262] As in Fig. 30 and Fig. 32, the dependence of the carbon contribution fraction η on the depth position Z1 is as strong as the dependence of the vacancy concentration N V from the depth position Z1. Thus, it is found that carbon has a strong interaction with vacancies. As can be seen from expression (15), Fig. 25 and the like, it is assumed that the donor consists of at least vacancies (V), oxygen (O), hydrogen (H) and carbon (C), based on the assumption that the slope of the concentration of VOH defects N VOH to the chemical oxygen concentration C OXproperty increases due to a donor other than the VOH defect. As described above, the donor is referred to as a VOH-C defect. VOH-C defects are an example of hydrogen donors.

[0263] As in Fig. As shown in Figure 32, the carbon contribution ratio η is large in a case where the depth position Z1 is shallow and the dosage of charged particles (hydrogen ions) is high. Therefore, in a case where the depth position Z1 is shallower than a predetermined value (e.g., 100 µm), the dosage of charged particles can be calculated based on the chemical carbon concentration C C , as in Fig. 29. The preferred value is 70 µm or 50 µm. If the depth position Z1 is shallower than a preset value and the chemical carbon concentration C C a given value (e.g. 1 × 10 13 atoms / cm 3) or more, the dosage of charged particles can be determined based on the chemical carbon concentration C C The default value can be 5 × 10 13 atoms / cm 3 or 1 × 10 14 Atoms / cm 3 be. (Example 1)

[0264] An example where the depth position Z1 is 120 µm and the dosage of hydrogen ions D H 5 × 10 12 ions / cm 2 is considered. In addition, the chemical oxygen concentration C OX 4.0 × 10 17 atoms / cm 3 , the chemical carbon concentration C C 2.0 × 10 15 atoms / cm 3 , the final doping concentration N F 7 × 10 13 / cm 3 and the volume donor concentration N BO 2 × 10 12 / cm 3 .

[0265] As described above, the vacancy concentration N Vapproximately 6 × 10 12 / cm 3 , the oxygen contribution fraction ξ is approximately 1.1 × 10 -4 and the carbon contribution is approximately 2.1 × 10 -4 from the relationship between the black square and the black diamond in Fig. 30, Fig. 31 and Fig. 32 and the relationship shown in each drawing. The Fig. 30, Fig. 31 and Fig. 32 can be experimentally determined in advance by measuring the plurality of semiconductor substrates 10.

[0266] From expression (16) N VOH 1 can be calculated as in the following expression. NVOH1=6.0×1012+1.1×10−4×4.0×10172.1×10−4×2.0×1015=5.04×1013 / cm3

[0267] In addition, N VOH2 can be calculated as in the following expression. NVOH2=NF−NB0=7×1013−2×1012=6.8×1013 / cm3

[0268] Thus, N VOH1 / N VOH2can be calculated as in the following expression. NVOH1 / NVOH2=5.04×1013 / 6.8×1013=0.74

[0269] Since it falls within a range of 0.1 ≤ N described above VOH1 / N VOH2 ≤ 10, it can be determined that N VOH1 and N VOH2 in Example 1 sufficiently match. (Example 2)

[0270] An example where the depth position Z1 is 23 µm and the dosage of hydrogen ions D H 3 × 10 12 ions / cm 2 is considered. In addition, the chemical oxygen concentration C OX 1.5 × 10 17 atoms / cm 3 , the chemical carbon concentration C C 5.1 × 10 14 atoms / cm 3 , the final doping concentration N F 1.4 × 10 14 / cm 3 and the volume donor concentration N BO 7.4 × 10 13 / cm 3 .

[0271] From the Fig. 30, Fig. 31 and Fig. 32, the vacancy concentration N V of approximately 1.3 × 10 13 / cm 3 , the oxygen contribution fraction ξ of approximately 1.0 × 10 -4 and the carbon contribution of approximately 3.0 × 10 -3 . From expression (16) follows N VOH1 is 3.0 × 10 13 / cm 3 . N VOH2 is 6.6 × 10 13 / cm 3 .

[0272] Thus, N VOH1 / N VOH2 can be calculated as in the following expression. NVOH1 / NVOH2=3.0×1013 / 6.6×1013=0.45

[0273] Since it falls within a range of 0.1 ≤ N described above VOH1 / N VOH2 ≤ 10, it can be determined that N VOH1 and N VOH2 in Example 2 are sufficiently consistent.

[0274] In a case where N VOH1 / N VOH2falls within the range, it can be determined that the dosage of charged particles is based on the measurement results of the chemical oxygen concentration C OX and the chemical carbon concentration C C was discontinued, as in the Fig. 29. In addition, N VOH1 / N VOH2 0.2 or more, 0.3 or more, or 0.5 or more. N VOH1 / N VOH2 can be 5 or less, 3 or less, or 2 or less.

[0275] The dosage of charged particles can be calculated by considering the carbon contribution fraction η, similarly to the case where the charged particles are not hydrogen ions. For example, if the charged particles are helium ions, the parameters c to h in expression (13b) are pre-determined with respect to the depth position Z1 at which helium ions are implanted. He is the dosage of helium ions (ions / cm2 ) with respect to the depth position Z1. NVOH=c×(DHe)d+e×(DHe)f×COX+g×(DHe)h×CC

[0276] The parameters can be recorded for a variety of types of depth positions Z1.

[0277] In addition, the chemical oxygen concentration C OX and the chemical carbon concentration C C of the semiconductor substrate 10 is measured and recorded before the start of the manufacturing process. Then, the dosage of helium ions D He which corresponds to the control value of the N to be generated VOH is calculated using expression (13b).

[0278] Fig. 33A, Fig. 33B and Fig. 33C are diagrams illustrating the dependence of the helium ion dosage on the vacancy concentration, the oxygen contribution fraction, and the carbon contribution fraction with respect to the target electrical properties. Fig. 33B and Fig. 33C illustrate three examples where the depth position Z1 at which helium ions are implanted is Z1 = a, Z1 = b, and Z1 = c. Expressions (13), (13a), and (13b) can also be applied to target properties other than the N to be generated. VOH The target properties are, for example, a breakdown voltage (V B ) of the semiconductor device 100, an ON voltage (V CE ) of the IGBT, a switching time (t off ), a forward voltage drop (V F ) of the diode, a reverse recovery time (t rr ) etc.

[0279] To realize the target properties, the semiconductor device can be designed using a flow chart similar to Fig. 29. That is, the concentrations of oxygen and carbon are measured in advance in step S1602, and the implantation amount (dosage) of helium ions is calculated in step S1604. Helium ions are implanted with the implantation amount in S1608. The ion implantation of hydrogen in S1610 may or may not be performed. In these cases, assuming, for example, that the target electrical property is F, Expression (13b) can be converted to the following expression. F=Nv'+ξ'COX+η'CC=c×(DHe)d+e×(DHe)f×COX+g×(DHe)h×CC

[0280] The first term on the right is Nv', which is obtained by multiplying the vacancy concentration Nv formed by helium ion implantation and heat treatment by a coefficient to be converted into the electrical property F. The coefficient part of Cox of the second term is the oxygen contribution fraction ξ' to the target property F and is an amount obtained by multiplying the above-described oxygen contribution fraction ξ by a coefficient to be converted into the electrical property F. The coefficient part of Cc of the third term is the carbon contribution fraction η' to the target property F and is an amount obtained by multiplying the above-described carbon contribution fraction η by a coefficient to be converted into the electrical property F. Expression (15) is determined in advance by generating at least one diagram of Fig. 33A, Fig. 33B and Fig. 33C for 2 to 3 or more types of helium ion depths. That is, the parameters c, d, e, f, g, and h can be experimentally obtained in advance. Therefore, the variable of Expression (15) is only the implantation amount of helium ions, and the right-hand side of Expression (15) is a constant that does not change with respect to the implantation amount.

[0281] The dosage of helium ions to be implanted into the semiconductor substrate 10 can be calculated by numerically solving expression (15). The implantation amount D obtained from expression (15) He of helium ions may have a width (error) that reflects a fluctuation in the values ​​of the data during the fitting described below. That is, if the implantation amount D He of the helium ions falls within the range of ±50% with respect to the value obtained from Expression (15), the implantation amount can be regarded as a value obtained from Expression (15).

[0282] The target electrical property F can be transformed with respect to each property of the example described above as in the following expression. VB=c×(DHe)d+e×(Dhe)f×COX+g×(DHe)h×CC VCE=c×(DHe)d+e×(DHe)f×COX+g×(DHe)e×Cc toff=c×(DHe)d+e×(DHe)f×COX+g×(DHe)e×Cc VF=c×(DHe)d+e×(DHe)f×COX+g×(DHe)e×Cc trr=c×(DHe)d+e×(DHe)f×COX+g×(DHe)e×Cc

[0283] In these expressions, the parameters c to h have different values ​​according to the target properties. The units of the parameters c to h can be units corresponding to the unit of the target property, the unit of the dosage of helium ions, the unit of the chemical oxygen concentration, and the unit of the chemical carbon concentration.

[0284] Fig. 34A, Fig. 34B and Fig. 34C are diagrams illustrating the dependence of the depth of helium ions on each of the converted vacancy concentration Nv', the converted oxygen contribution fraction ξ', and the converted carbon contribution fraction η' in the target electrical properties. In each drawing, six examples are shown in which the dosages of helium ions to the depth position Z1 D He 1, D He 2, D He 3, D He 4, D He 5 and D He 6. The converted vacancy concentration Nv', the converted oxygen contribution fraction ξ' and the converted carbon contribution fraction η' with respect to the depth of the helium ions can also be calculated in advance using the previously experimentally measured parameters c, d, e, f, g and h. Using this diagram, one can see whether the semiconductor device can be tested based on the process in Fig. 29 was manufactured.

[0285] The implantation depth Z1 of helium ions is a depth from an implantation surface of a peak value of an atomic density distribution (chemical concentration distribution) of helium atoms in the semiconductor substrate 10, which is measured, for example, by the SIMS. The implantation surface can be the main surface on which the chemical concentration distribution of helium atoms extends from the peak position into the depth direction. The dosage of helium ions can be an integrated value obtained by integrating the measured chemical concentration distribution of helium atoms in the depth direction from the implantation surface. By interpolating the values ​​shown in the diagrams above, Fig. 34A, Fig. 34B and Fig. From the calculated depth and helium ion dosage described in Figure 34C, the converted vacancy concentration Nv', the converted oxygen contribution fraction ξ', and the converted carbon contribution fraction η' can be determined. The chemical concentrations of oxygen and carbon can be determined by SIMS. By substituting these values ​​into expression (15), the electrical property value F1 is obtained.

[0286] On the other hand, an actual electrical characteristic F2 can be determined by electrical measurement of the semiconductor device 100. As a result, it can be said that the semiconductor device is based on the flow of the Fig. 29, if the ratio between the calculated value F1 and the measured value F2 is in the range of 0.1 to 10. The first value or third value described above is an example of the calculated value F1 and the second value is an example of the measured value F2.

[0287] Fig. 35 is a diagram illustrating another example of the manufacturing method of the semiconductor device 100. The manufacturing method of this example differs in that a parameter acquisition step S3502 and a condition setting step S3503 are performed instead of the concentration measurement step S1602 and the implantation amount calculation step S1604 of the example in Fig. 16 to Fig. 34C. The other steps are similar to the examples shown in Fig. 16 to Fig. 34C. Although the oxygen introduction step S1802 in Fig. 35 is not shown, the manufacturing process of this example can be used in Fig. 18 may or may not include the oxygen introduction step S1802 described.

[0288] In the parameter acquisition step S3502, parameters related to the semiconductor substrate 10 prepared in the substrate preparation step 1600 are acquired. Fig. 16 and the like is an example of the parameter acquisition step S3502. The parameters can be as shown in Fig. 16 to 34C, a parameter such as the chemical oxygen concentration, the chemical carbon concentration, etc. may be measured in parameter acquisition step S3502, and a determination value, a design value, etc. of the semiconductor substrate 10 may be acquired. In parameter acquisition step S3502, a parameter affecting the hydrogen donor concentration of the semiconductor substrate 10 or a parameter affecting the breakdown voltage of the semiconductor device 100 may be acquired. In parameter acquisition step S3502, at least one of the chemical oxygen concentration, the chemical carbon concentration, the oxygen contribution ratio ξ, and the carbon contribution ratio η in the semiconductor substrate 10 may be acquired. These parameters may affect the hydrogen donor concentration and the breakdown voltage of the semiconductor device 100.In parameter acquisition step S3202, the thickness of the semiconductor substrate 10 and / or the volume donor concentration can be acquired. These parameters can influence the breakdown voltage of the semiconductor device 100.

[0289] In condition setting step S3503, one or a plurality of conditions of the particle implantation condition in particle implantation step S1608, the hydrogen implantation condition in hydrogen implantation step S1610, and the heat treatment condition in heat treatment step S1612 are adjusted based on at least one parameter acquired in parameter acquisition step S3502. In condition setting step S3503, these conditions can be adjusted so that the hydrogen donor concentration of the semiconductor substrate 10 or the breakdown voltage of the semiconductor device 100 reaches a predetermined target value. The implantation amount calculation step S1604 in Fig. 16 and the like is an example of the condition setting step S3503. The particle implantation condition includes the dosage of the charged particles and / or the implantation depth Z1 (see, for example, Fig. 2). The hydrogen implantation condition includes a dosage of hydrogen ions and / or the implantation depth Z2 (see e.g. Fig. 2). The heat treatment condition includes at least a heat treatment temperature and / or a heat treatment duration.

[0290] For example, if the parameters acquired in parameter acquisition step S3502 indicate a state in which it is difficult to generate a hydrogen donor in the semiconductor substrate 10, in condition setting step S3503, the condition of each step is adjusted to a condition for promoting the generation of a hydrogen donor in the semiconductor substrate 10. For example, if the chemical oxygen concentration of the semiconductor substrate 10 is low, VOH defects are less likely to be generated. On the other hand, the generation of VOH defects is similarly promoted by increasing the dosage of the charged particle beam in the particle implantation step S1608, increasing the dosage of hydrogen ions in the hydrogen ion implantation step S1610, increasing the heat treatment temperature in the heat treatment step S1612, and increasing the heat treatment time in the heat treatment step S1612.The condition setting step S3503 may perform at least one of these operations. Similarly, if the chemical oxygen concentration of the semiconductor substrate 10 is high, VOH defects are likely to be generated. On the other hand, the generation of VOH defects is similarly suppressed by reducing the dosage of the charged particle beam in the particle implantation step S1608, reducing the dosage of hydrogen ions in the hydrogen ion implantation step S1610, lowering the heat treatment temperature in the heat treatment step S1612, and shortening the heat treatment time in the heat treatment step S1612. In the condition setting step S3503, one of these operations may be performed, or a plurality of operations may be performed in combination.Furthermore, if the chemical carbon concentration of the semiconductor substrate 10 is high, VOH defects tend to be more easily generated, and if the chemical carbon concentration is low, VOH defects tend to be more difficult to generate. The oxygen contribution ratio ξ and the carbon contribution ratio η are as described above. Condition setting step S3503 can perform the same operations as the chemical oxygen concentration based on these parameters.

[0291] Furthermore, if the parameter acquired in parameter acquisition step S3502 indicates a state in which the breakdown voltage of the semiconductor device 100 decreases, in condition setting step S3503, the condition of each step is adjusted to a condition for improving the breakdown voltage of the semiconductor device 100. For example, if the thickness of the semiconductor substrate 10 is small, the breakdown voltage of the semiconductor device 100 becomes smaller. On the other hand, if the concentration of hydrogen donors generated in the semiconductor substrate 10 is reduced, for example, the doping concentration of the drift region 18 is reduced. As a result, the breakdown voltage of the semiconductor device 100 is improved. By lowering the dosage of charged particles or the dosage of hydrogen ions, the concentration of hydrogen donors can be reduced.In addition, the concentration of hydrogen donors can be reduced by lowering the heat treatment temperature or shortening the heat treatment time. If the length of the via region 106 in the semiconductor substrate 10 is reduced in the Z-axis direction, the length of the high-concentration flat portion 150 (see FIG. Fig. 3) in the Z-axis direction. As a result, the breakdown voltage of the semiconductor device 100 is improved. The length of the passage region 106 can be shortened by bringing the charged particle implantation position Z1 closer to the bottom surface 23. In condition setting step S3503, one of these operations may be performed, or a plurality of operations may be performed in combination.

[0292] If, in addition, the volume donor concentration N BOof the semiconductor substrate 10 is high, the doping concentration of the drift region 18 increases and the breakdown voltage of the semiconductor device 100 decreases. In the condition setting step S3503, the same operations as for the thickness of the semiconductor substrate 10 can also be performed for the volume donor concentration N BO be performed.

[0293] In condition adjustment step S3503, the condition of each step can be adjusted based on a difference or a relationship between the detected parameter and a preset reference value. The extent to which the condition of each step should be adjusted with respect to the difference and relationship can be determined in advance through an experiment or the like. As described above, the characteristics of the semiconductor device 100 can be adjusted by adjusting the conditions of each step based on the detected parameters. Furthermore, fluctuations in the characteristics of the semiconductor device 100 can be reduced.

[0294] Fig. 36 is a diagram illustrating another example of the manufacturing method of the semiconductor device 100. In the manufacturing method of this example, the measuring step S1602 is implemented as the parameter acquisition step S3505 in Fig. 35. The other steps are identical to those of the example of Fig. 35. In the concentration measuring step S1602, the chemical oxygen concentration of the semiconductor substrate 10 is measured after performing the oxygen introduction step S1802.

[0295] In condition setting step S3503, at least one of the charged particle beam implantation condition in particle implantation step S1608, the hydrogen ion implantation condition in hydrogen implantation step S1610, and the heat treatment condition in heat treatment step S1612 is adjusted according to the chemical oxygen concentration. In condition setting step S5303, the charged particle beam implantation condition can be adjusted similarly to the example of Fig. 16 and the like.

[0296] If the hydrogen ion implantation condition is adjusted in the hydrogen implantation step S1610, the hydrogen ion dosage can be adjusted. By adjusting the hydrogen ion dosage, the concentration of hydrogen diffused into the passage region 106 can be adjusted, and the concentration of hydrogen donors formed in the passage region 106 can be adjusted. In the condition setting step S3503, the hydrogen ion dosage can be set lower than the hydrogen reference value if the chemical oxygen concentration is higher than the oxygen reference value, and the hydrogen ion dosage can be set higher than the hydrogen reference value if the chemical oxygen concentration is lower than the oxygen reference value.As a result, the influence of the fluctuation of the chemical oxygen concentration can be reduced and the doping concentration of the through region 106 can be precisely adjusted.

[0297] If the heat treatment condition is adjusted in heat treatment step S1612, the heat treatment temperature and / or the heat treatment time can be adjusted. By adjusting the heat treatment temperature or the heat treatment time, hydrogen diffusing into the passage region 106 can be adjusted and the generation of hydrogen donors can be adjusted. In condition setting step S3503, if the chemical oxygen concentration is higher than the oxygen reference value, the heat treatment temperature can be set lower than the temperature reference value, and if the chemical oxygen concentration is lower than the oxygen reference value, the heat treatment temperature can be set higher than the temperature reference value.Similarly, if the chemical oxygen concentration is higher than the oxygen reference value, the heat treatment time can be set shorter than the reference time, and if the chemical oxygen concentration is lower than the oxygen reference value, the heat treatment time can be set longer than the reference time. As a result, the influence of the fluctuation in the chemical oxygen concentration can be reduced and the doping concentration of the through region 106 can be precisely adjusted.

[0298] In the concentration measuring step S1602, the chemical carbon concentration of the semiconductor substrate 10 can be further measured as in the example of Fig. 29 can be measured. In condition setting step S3503, the condition of each step can be further adjusted based on the chemical carbon concentration. The higher the chemical carbon concentration, the higher the concentration of hydrogen donors can be. In condition setting step S3503, the condition can be set to suppress the formation of the hydrogen donor if the chemical carbon concentration is higher than the carbon reference value, and the condition can be set to promote the formation of the hydrogen donor if the chemical carbon concentration is lower than the carbon reference value.

[0299] In condition setting step S3503, the hydrogen ion implantation condition in the hydrogen implantation step S1610 and the heat treatment condition in the heat treatment step S1612 can be adjusted based on the implantation depth Z1 of the charged particle beam in the particle implantation step S1608. The length of the passage region 106 changes according to the implantation depth Z1 of the charged particle beam, and the total amount of lattice defects formed in the semiconductor substrate 10 changes. The total amount of hydrogen donors formed depends on the total amount of lattice defects. Therefore, the total amount of hydrogen donors formed changes according to the implantation depth Z1 of the charged particle beam.In condition setting step S3503, the hydrogen ion implantation condition and / or the heat treatment condition can be adjusted so that the total amount of hydrogen donors generated reaches a predetermined reference value. In condition setting step S3503, the hydrogen ion implantation condition or the heat treatment condition, which are calculated according to the chemical oxygen concentration, can be corrected according to the implantation depth Z1 of the charged particle beam.

[0300] If the plurality of semiconductor devices 100 are manufactured, the setting of the implantation condition in particle implantation step S1608 and the setting of the implantation condition in hydrogen implantation step S1610 may be performed for each semiconductor substrate 10 (for each semiconductor device 100). Furthermore, the setting of the heat treatment condition in heat treatment step S1612 may be performed collectively for the plurality of semiconductor substrates 10. The heat treatment of the plurality of semiconductor substrates 10 can be performed in parallel by placing the plurality of semiconductor substrates 10 in a common heat treatment furnace. Implanting charged particles or hydrogen ions into each semiconductor substrate 10 can be performed for each semiconductor substrate 10. Through this process, the adjustment of the plurality of semiconductor substrates 10 can be performed efficiently.

[0301] Fig. 37 is a diagram illustrating another example of the manufacturing method of the semiconductor device 100. In the manufacturing method of this example, a concentration detection step S3702 is used as the parameter detection step S3502 in Fig. 35. The other steps are identical to those of the example of Fig. 35 and Fig. 36.

[0302] In concentration acquisition step S3702, information about the impurity concentration in a predetermined region of the semiconductor substrate 10 is acquired. In concentration acquisition step S3702, the information acquired by measuring the semiconductor substrate 10, or a design value, a determination value, or the like of the information of the semiconductor substrate 10 may be acquired. The predetermined region of the semiconductor substrate 10 is, for example, a center position of the semiconductor substrate 10 in the depth direction, but is not limited thereto. The impurity concentration in the upper surface 21 or the lower surface 23 of the semiconductor substrate 10 may be acquired. In concentration acquisition step S3702, the impurity concentration that affects the doping concentration of the drift region 18 at the time of completing the semiconductor device 100 may be acquired.As an example, in the concentration detection step S3702, at least one of the following values ​​is detected: the chemical oxygen concentration, the chemical carbon concentration, and the volume donor concentration N. BO . The operations of the condition setting step S3503 in a case that the chemical oxygen concentration or the chemical carbon concentration is detected may be the same as in the example of Fig. be 36.

[0303] In the condition setting step S3503 of this example, at least one of the implantation condition of the charged particle beam in the particle implantation step S1608, the implantation condition of hydrogen ions in the hydrogen implantation step S1610, and the heat treatment condition in the heat treatment step S1612 is set based on the volume donor concentration N BO adjusted.

[0304] The final doping concentration of the drift region 18 in a state where the fabrication of the semiconductor device 100 is completed depends on the volume donor concentration N BO and the hydrogen donor concentration. If the volume donor concentration N BO fluctuates, the final doping concentration also fluctuates. In the condition setting step S3503, the condition of each step is adjusted so that the deviation between the volume donor concentration N BO and a predetermined reference value. That is, in the condition setting step S3503, if the volume donor concentration N BO is smaller than the reference value, the condition of each step is adjusted so that the amount of hydrogen donors produced increases, and if the volume donor concentration N BOis greater than the reference value, the condition of each step is adjusted so that the produced amount of hydrogen donors decreases. The method for adjusting the produced amount of hydrogen donors in each step is the same as in the examples of Fig. 35 and Fig. 36. This can reduce variations in the final doping concentration.

[0305] In condition setting step S3503, each condition can be adjusted so that the integrated value of the doping concentration in the drift region 18 reaches a predetermined reference value. As an example, condition setting step S3503 adjusts the implantation depth of the charged particle beam in the particle implantation step S1608 according to the detected impurity concentration. The length of the passage region 106 is adjusted by adjusting the implantation depth, so that the integrated value of the doping concentration of the drift region 18 can be adjusted.

[0306] In condition setting step S3503, the condition of each step can be further adjusted based on the oxygen contribution ratio ξ and / or the carbon contribution ratio η. This allows the amount of hydrogen donors formed to be adjusted more precisely.

[0307] Fig. Figure 38 is a graph illustrating a relationship between the volume donor concentration and the charged particle implantation depth Z1. In this example, the volume donor concentration detected in the concentration detection step S3702 is expressed as N BO set, the volume reference value is set to N Br and the relationship between N BO and N Br is set as γ. That is, N BO = γ·N Br . In addition, the implantation depth of charged particles before the condition adjustment is Z1 r, the implantation depth of charged particles after condition adjustment is Z1 and the ratio between Z1 and Z1 r is ε. That is, Z1 = ε·Z1 r . The implantation depth Z denotes a distance from the lower surface 23 of the semiconductor substrate to the implantation position.

[0308] As in Fig. 38, in the condition setting step S3503, the implantation depth Z1 of the charged particles is set so that ε decreases while γ increases. Fig. 13 shown volume donor concentration N BOincreases while γ increases. For this reason, the doping concentration of the drift region 18 may increase, and the breakdown voltage of the semiconductor substrate 10 may decrease. On the other hand, by reducing ε to reach the implantation depth Z1 and the bottom surface 23, the length of the high-concentration shallow portion 150 can be shortened, so that the integrated value of the doping concentration in the drift region 18 can be reduced. This makes it possible to suppress a decrease in the breakdown voltage of the semiconductor substrate 10.

[0309] Fig. 39 is a diagram illustrating another example of the manufacturing method of the semiconductor device 100. In the manufacturing method of this example, a step S3902 for measuring the substrate thickness is used as the parameter acquisition step S3502 in Fig. 35. In addition, a grinding step S3901 of grinding the semiconductor substrate 10 to adjust the thickness is performed before the substrate thickness measuring step S3902. The other steps are identical to those of the example of Fig. 35, Fig. 36 or Fig. 37.

[0310] In grinding step S3901, the thickness of the semiconductor substrate 10 can be adjusted according to the breakdown voltage that the semiconductor device 100 is to have. The grinding step S3901 can be performed before the particle implantation step S1608 or before the hydrogen implantation step S1610. In grinding step S3901, the bottom surface 23 of the semiconductor substrate 10 can be ground by CMP or the like.

[0311] In substrate thickness measurement step S3902, the thickness in a predetermined region of the semiconductor substrate 10 is measured. In substrate thickness measurement step S3902, an average value of thicknesses measured at a plurality of points may be used. In substrate thickness measurement step S3902, the thickness in the active portion 160 may be measured, and the thickness in the edge termination structure portion 90 may be measured.

[0312] In condition setting step S3503, at least one of the implantation condition of the charged particle beam in the particle implantation step S1608, the implantation condition of hydrogen ions in the hydrogen implantation step S1610, and the heat treatment condition in the heat treatment step S1612 is adjusted based on the measured thickness of the semiconductor substrate 10.

[0313] If the thickness of the semiconductor substrate 10 varies, the breakdown voltage of the semiconductor device 100 may vary. In condition setting step S3503, the condition of each step is adjusted so that the deviation between the thickness of the semiconductor substrate 10 and that of the predetermined thickness reference value is compensated. That is, in condition setting step S3503, if the thickness of the semiconductor substrate 10 is smaller than the thickness reference value, the condition of each step is set to increase the breakdown voltage, and if the thickness of the semiconductor substrate 10 is larger than the thickness reference value, the condition of each step is set to decrease the breakdown voltage. The breakdown voltage of the semiconductor device 100 can be increased by decreasing the integrated value of the doping concentration in the drift region 18, and the breakdown voltage of the semiconductor device 100 can be decreased by increasing the integrated value.The integrated value of the doping concentration can be adjusted by the amount of hydrogen donors formed in the drift region 18. As described above, the amount of hydrogen donor formed can be adjusted by the condition in each step.

[0314] In this example, the thickness of the semiconductor substrate 10 in the edge termination structure portion 90 can be measured to adjust each implantation condition for the edge termination structure portion 90. The active portion 160 and the edge termination structure portion 90 can have different implantation conditions for charged particles and hydrogen ions. As a result, the doping concentration and breakdown voltage in the edge termination structure portion 90 can be precisely controlled.

[0315] Fig. 40 is a diagram illustrating an example of the equipotential surface 308 in the edge termination structure section 90. The structure of the edge termination structure section 90 is the same as in the example of Fig. 14. A region with a higher concentration than the volume donor concentration is formed on the lower surface side 23 at the second peak 141. Thus, the curvature of the equipotential surface 308 changes near the second peak 141. As a result, the equipotential surface 308 spreads toward the outer peripheral side of the semiconductor device 100 near the upper surface 21 of the semiconductor substrate 10. Therefore, the extent to which the equipotential surface 308 spreads toward the outer peripheral side of the semiconductor device 100 depends on a distance Zb between the upper surface 21 of the semiconductor substrate 10 and the second peak 141.

[0316] In condition setting step S3503, the implantation depth Z1 of the charged particles into the edge termination structure portion 90 can be adjusted based on the thickness of the semiconductor substrate 10 in the edge termination structure portion 90. As a result, the pitch Zb can be accurately controlled. Thus, it is possible to suppress excessive longitudinal expansion of the depletion layer in the edge termination structure portion 90. Thus, the length in the outer circumferential direction of the edge termination structure portion 90 can be shortened, and the area of ​​the upper surface 21 of the semiconductor device 100 can be reduced.

[0317] In condition setting step S3503, each implantation condition for the edge termination structure portion 90 can be adjusted. As a result, the doping concentration under the second peak 141 can be precisely adjusted, so that the propagation type of the equipotential surface 308 can be further controlled.

[0318] It should be noted that the order of execution of each process, such as operations, procedures, steps, and stages in the devices, systems, programs, and methods illustrated in the claims, description, and drawings, may be implemented in any order unless "before," "before," or the like is expressly stated, and unless the result of the previous process is used in the later process. Although the workflow is described in the claims, description, and drawings as "first," "next," and the like for convenience, this does not mean that this order must necessarily be followed. LIST OF REFERENCE SYMBOLS 10 Semiconductor substrate 11 Sink area 12 Emitter area 14 Basic area 15 Contact area 16 Accumulation area 18 Drift area 20 Buffer area 21 upper surface 22 Collector area 23 lower surface 24 Collector electrode 25 Peak 26 Peak 29 linear section 30 Dummy trench section 31 marginal section 32 dielectric dummy film 34 Dummy line section 38 dielectric interlayer film 39 linear section 40 gate ditch section 41 marginal section 42 Gate insulating film 44 gate line section 52 Emitter electrode 54 contact hole 60, 61 Mesa section 70 transistor section 80 diode section 81 Extension area 82 Cathode area 90 edge finish structure section 92 protective ring 94 field plate 100 semiconductor devices 106 Passage area 111 second donor peak 121 first donor peak 130 outer circumferential gate runner 131 active-side gate runner 133 first peak 141 second peak 142 lower apron 143 upper apron 150 flat section 151 second peak of the contributing concentration 160 active section 161 first peak of the contributing concentration 162 End page 164 gate connection area, 171 vacancy peak 174 sewer stoppers 180 area 181 second VOH peak 191 first VOH peak 194 Peak of hydrogen concentration 214 linear approximate distribution 216 band-shaped area 308 Equipotential surface 601, 602, 603, 611, 612, 613, 631, 632, 633, 641, 642, 643 Straight 621, 622, 623 Plot 801, 802, 811, 812, 813, 821, 822, 823, 831, 832, 833, 901, 902 curve

Claims

[1] A semiconductor device (100) comprising: a semiconductor substrate (10) having an upper surface (21) and a lower surface (23) and containing oxygen; a first peak (133) of a chemical hydrogen concentration located on the lower surface side of the semiconductor substrate (10); and a flat portion (150) arranged on the upper surface side of the semiconductor substrate (10) with respect to the first peak (133), which contains a hydrogen donor and which has a substantially flat distribution of the donor concentration in the depth direction of the semiconductor substrate (10), wherein an oxygen contribution fraction, which indicates a fraction of a chemical oxygen concentration that contributes to generating the hydrogen donor in the chemical oxygen concentration of the oxygen, in the range of 1 × 10 -5 up to 7 × 10 -4 lies, a concentration of oxygen contributing to the generation of the hydrogen donor in the flat section (150) is lower than the chemical hydrogen concentration, and a hydrogen donor concentration in the flat section (150) in the range of 2 × 10 12 / cm3 to 5 × 10 14 / cm 3 lies. [2] A semiconductor device (100) according to claim 1, wherein the semiconductor substrate (10) contains a volume donor, and a donor concentration of the flat section (150) is higher than a volume donor concentration. [3] A semiconductor device (100) according to claim 2, wherein a sum of a value obtained by multiplying a chemical oxygen concentration in the flat section (150) by the oxygen contribution fraction and a vacancy concentration of the flat section (150) is defined as a first value of the hydrogen donor concentration, a difference resulting from subtracting the volume donor concentration from a donor concentration of the flat portion (150) is defined as a second value of the hydrogen donor concentration, and a ratio of the first value of the hydrogen donor concentration to the second value of the hydrogen donor concentration is in the range of 0.1 to 10. [4] A semiconductor device (100) comprising: a semiconductor substrate (10) having an upper surface (21) and a lower surface (23) and containing oxygen; a first peak (133) of a chemical hydrogen concentration located on the lower surface side of the semiconductor substrate (10); and a flat portion (150) arranged on the upper surface side of the semiconductor substrate (10) with respect to the first peak (133), which contains a hydrogen donor and which has a substantially flat distribution of the donor concentration in the depth direction of the semiconductor substrate (10), wherein the semiconductor substrate (10) contains a volume donor, a donor concentration of the flat section (150) is higher than a volume donor concentration, a sum of a value obtained by multiplying a chemical oxygen concentration in the flat portion (150) by an oxygen contribution fraction indicating a fraction of a chemical oxygen concentration contributing to generating the hydrogen donor in the chemical oxygen concentration of the oxygen, and a vacancy concentration of the flat portion (150) is defined as a first value of a hydrogen donor concentration, a difference resulting from subtracting the volume donor concentration from a donor concentration of the flat portion (150) is defined as a second value of the hydrogen donor concentration, and a ratio of the first value of the hydrogen donor concentration to the second value of the hydrogen donor concentration is in the range of 0.1 to 10. [5] A semiconductor device (100) comprising: a semiconductor substrate (10) having an upper surface (21) and a lower surface (23) and containing oxygen and carbon; a first peak (133) of a chemical hydrogen concentration located on the lower surface side of the semiconductor substrate (10); and a flat portion (150) arranged on the upper surface side of the semiconductor substrate (10) with respect to the first peak (133), which contains a hydrogen donor and which has a substantially flat distribution of the donor concentration in the depth direction of the semiconductor substrate (10), wherein the semiconductor substrate (10) contains a volume donor, a donor concentration of the flat section (150) is higher than a volume donor concentration, a sum of a value obtained by multiplying a chemical oxygen concentration in the flat section (150) by an oxygen contribution fraction indicating a fraction of a chemical oxygen concentration that contributes to generating the hydrogen donor in the chemical oxygen concentration of the oxygen, a value obtained by multiplying a chemical carbon concentration in the flat section (150) by a carbon contribution fraction indicating a fraction of a chemical carbon concentration that contributes to generating the hydrogen donor in the chemical carbon concentration of the carbon, and a vacancy concentration of the flat section (150) is defined as a third value of a hydrogen donor concentration, a difference resulting from subtracting the volume donor concentration from the donor concentration of the flat section (150) is defined as a second value of the hydrogen donor concentration, and a ratio of the third value of the hydrogen donor concentration to the second value of the hydrogen donor concentration is in the range of 0.1 to 10. [6] A semiconductor device (100) according to any one of claims 1 to 5, further comprising: a second peak (141) of a chemical concentration of hydrogen or helium arranged on the upper surface side of the semiconductor substrate (10), wherein the flat portion (150) is arranged closer to the lower surface side of the semiconductor substrate (10) than the second peak (141). [7] Semiconductor device (100) according to one of claims 1 to 6, wherein a hydrogen fraction, which indicates a proportion of the chemical hydrogen concentration that contributes to the generation of the hydrogen donor in the chemical hydrogen concentration, is in the range of 0.001 to 0.3, [8] Semiconductor device (100) according to one of claims 1 to 7, wherein a vacancy concentration in the flat section (150) in the range of 1 × 10 11 / cm 3 up to 1 × 10 14 / cm 3 lies. [9] A semiconductor device (100) according to any one of claims 1 to 8, wherein the oxygen contribution ratio is 5 × 10 -4 or less. [10] A semiconductor device (100) according to any one of claims 1 to 9, wherein the oxygen contribution ratio is 1 × 10 -4 or more. [11] A semiconductor device (100) according to claim 6, wherein the chemical hydrogen concentration of the first peak (133) is higher than the chemical hydrogen concentration of the second peak (141). [12] Semiconductor device (100) according to one of claims 1 to 11, wherein a chemical oxygen concentration in the flat section (150) 1 × 10 17 atoms / cm 3 or more. [13] Semiconductor device (100) according to one of claims 1 to 12, wherein a chemical carbon concentration in the flat section (150) in the range of 1 × 10 13 / cm 3 up to 1 × 10 16 / cm 3 lies. [14] A method of manufacturing a semiconductor device (100), the method comprising: Measuring a chemical oxygen concentration of a semiconductor substrate (10) having a top surface and a bottom surface; implanting a beam of charged particles from the lower surface of the semiconductor substrate (10) so that they pass through half or more of a thickness of the semiconductor substrate (10) in the depth direction; and performing a heat treatment of the semiconductor substrate (10) after implanting the charged particle beam, wherein an implantation condition of the charged particle beam when implanting the charged particle beam and / or a Heat treatment condition is adjusted according to the chemical oxygen concentration when performing heat treatment. [15] A manufacturing method of the semiconductor device (100) according to claim 14, wherein during the measurement, a chemical carbon concentration of the semiconductor substrate (10) is further measured, and the implantation condition of the charged particle beam when implanting the charged particle beam is adjusted according to the chemical oxygen concentration and the chemical carbon concentration. [16] A manufacturing method of the semiconductor device (100) according to claim 15, wherein at a predetermined depth position of the semiconductor substrate (10), when a concentration of the hydrogen donor to be produced by the manufacturing process is N VOH1 is a concentration of the actually produced hydrogen donor N VOH2 is a vacancy concentration formed by implanting the charged particle beam, N V is a chemical oxygen concentration C OX is a chemical carbon concentration C C is an oxygen contribution fraction indicating a fraction of the chemical oxygen concentration that contributes to generating the hydrogen donor in the chemical oxygen concentration, ξ, and a carbon contribution fraction indicating a fraction of the chemical carbon concentration that contributes to generating the hydrogen donor in the chemical carbon concentration, η, the following relationship is satisfied: NVOH1=NV+ξCOX+ηCC; and 0.1≤NVOH1 / NVOH2≤10. [17] A manufacturing method of the semiconductor device (100) according to any one of claims 14 to 16, wherein When implanting the charged particle beam, hydrogen ions are implanted as the charged particle beam. [18] A manufacturing method of the semiconductor device (100) according to any one of claims 14 to 17, further comprising: Implanting hydrogen ions into the lower surface side of the semiconductor substrate (10) before performing the heat treatment. [19] A manufacturing method of the semiconductor device (100) according to claim 18, wherein an implantation condition of the hydrogen ions when implanting hydrogen is adjusted based on the chemical oxygen concentration of the semiconductor substrate (10). [20] A manufacturing method of the semiconductor device (100) according to claim 18 or 19, wherein When implanting the charged particle beam, an implantation condition of hydrogen ions when implanting hydrogen and the heat treatment condition when performing the heat treatment are adjusted based on an implantation depth of the charged particle beam. [21] A manufacturing method of the semiconductor device (100) according to any one of claims 14 to 20, further comprising: Introducing oxygen into the semiconductor substrate (10). [22] A manufacturing method of the semiconductor device (100) according to any one of claims 14 to 21, wherein the implantation condition of the charged particle beam when implanting the charged particle beam and / or the heat treatment condition when performing the heat treatment is adjusted according to a volume donor concentration of the semiconductor substrate (10). [23] A manufacturing method of the semiconductor device (100) according to any one of claims 14 to 22, further comprising: Grinding the semiconductor substrate (10); and Measuring a thickness of the semiconductor substrate (10) after grinding, wherein the implantation condition of the charged particle beam when implanting the charged particle beam and / or the heat treatment condition when performing the heat treatment is adjusted according to a thickness of the semiconductor substrate (10). [24] A manufacturing method of the semiconductor device (100) according to any one of claims 14 to 23, wherein when implanting the charged particle beam, the implantation condition is adjusted for each of a plurality of semiconductor substrates (10), and when performing the heat treatment, the heat treatment condition is adjusted jointly for the plurality of semiconductor substrates (10). [25] A manufacturing method of the semiconductor device (100) according to claim 18, further comprising: Grinding the semiconductor substrate (10); and Measuring a thickness of the semiconductor substrate (10) after grinding, wherein an implantation condition of the hydrogen ions when implanting hydrogen is adjusted based on the thickness of the semiconductor substrate (10). [26] A manufacturing method of the semiconductor device (100) according to claim 23 or 25, wherein when measuring the thickness of the semiconductor substrate (10), a thickness of the semiconductor substrate (10) is measured in an edge termination structure portion (90) of the semiconductor substrate (10). [27] A manufacturing method of the semiconductor device (100) according to any one of claims 14 to 26, wherein when implanting the charged particle beam, the implantation condition of the charged particle beam is further adjusted based on at least an oxygen contribution ratio ξ indicating a ratio of the chemical oxygen concentration that contributes to generating a hydrogen donor in the chemical oxygen concentration, and a carbon contribution ratio η indicating a ratio of the chemical carbon concentration that contributes to generating a hydrogen donor in the chemical carbon concentration of the semiconductor substrate (10). [28] A method of manufacturing a semiconductor device (100), comprising: Detecting an impurity concentration of a semiconductor substrate (10) having an upper surface and a lower surface; implanting a beam of charged particles from the lower surface of the semiconductor substrate (10) so that they pass through half or more of a thickness of the semiconductor substrate (10) in the depth direction; and performing a heat treatment of the semiconductor substrate (10) after implanting the charged particle beam, wherein an implantation depth of the charged particle beam is adjusted according to the impurity concentration when implanting the charged particle beam. [29] A manufacturing method of the semiconductor device (100) according to claim 28, wherein at least one of a volume donor concentration, a chemical oxygen concentration and a chemical carbon concentration of the semiconductor substrate (10) is detected when detecting the impurity concentration.

Images