JP5374886B2 - Semiconductor device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve trade-off relation between ON resistance and withstanding voltage in a semiconductor device using a semiconductor substrate of a parallel p-n structure having favorable trade-off relation between turn-off loss and dV<SB>ds</SB>/dt. <P>SOLUTION: In the semiconductor device having a parallel p-n layer with an n-type drift region 2 and a p-type partition region alternately arranged a plurality of times, an insulating film 16 thicker than a gate oxide film 6 is formed on the surface of a region with no p base region 8 between first p-type partition regions 3a having the p base regions 8 among the p-type partition regions. A gate electrode 7 is formed on the n-type drift region 2 and a second p-type partition region 3b with no p base region 8 among the p-type partition regions via the gate oxide film 6 and the insulating film 16. The ratio of an area 2 of a region where the gate electrode 7 covers the insulating film 16 to an area 1 of a region where the gate electrode 7 covers an active region is set to 0.1-0.4. <P>COPYRIGHT: (C)2009,JPO&amp;INPIT

Description

  The present invention relates to a vertical semiconductor device for high power, and more particularly to a semiconductor device having a superjunction layer on a part of a semiconductor substrate.

  2. Description of the Related Art Conventionally, in order to reduce the size and performance of power supply equipment in the power electronics field, power semiconductor devices are required to have higher breakdown voltage and higher current, as well as lower loss, higher breakdown resistance, and higher speed. For this reason, a superjunction substrate has been proposed as the substrate structure of the semiconductor device, and a vertical MOS power device structure has been proposed as the surface structure.

  As a substrate structure of a semiconductor device, a semiconductor substrate having a single conductivity type and a super junction type substrate are widely known. In the super junction type substrate, the first conductivity type and the second conductivity type semiconductor regions are alternately arranged in a direction perpendicular to the semiconductor substrate between the first conductivity type semiconductor substrate and the second conductivity type semiconductor layer. It has a formed super junction layer (see, for example, Patent Document 1, Patent Document 2, and Patent Document 3 below). This superjunction substrate can spread the space charge region over the entire superjunction layer even when the concentration of the semiconductor regions of the first conductivity type and the second conductivity type is high. Therefore, the on-resistance can be reduced particularly in a high breakdown voltage semiconductor device.

Note that in this specification, a semiconductor having n or p means that electrons and holes are majority carriers, respectively. Further, “ ⁺ ” or “ ⁻ ” attached to n or p, such as n ⁺ or n ^−, is relatively higher or lower than the impurity concentration of the semiconductor to which they are not attached. Represents that.

An example of such a vertical MOS device will be described. FIG. 32 is a plan view showing the configuration of the vertical MOS device of the first conventional example. FIG. 33 is a cross-sectional view showing a cross-sectional structure taken along section line AA-AA ′ of FIG. As shown in FIG. 33, an n-type drift region (first conductivity type semiconductor region) 42 and a p-type partition region (on the first main surface side surface of an n ⁺ substrate 41 having a low resistivity which is an n ⁺ drain region) A parallel pn layer (superjunction layer) composed of the second conductivity type semiconductor region) 43 is provided. The parallel pn layer allows a current to flow through the n-type drift region 42 in the on state, and depletes the n-type drift region 42 and the p-type partition region 43 in the off state. Thus, a semiconductor substrate (superjunction substrate) having a parallel pn structure including the parallel pn layers in which the n-type drift regions 42 and the p-type partition regions 43 are alternately arranged and the n ⁺ substrate 41 is formed. Has been.

A planar type MOS structure is formed on the first main surface side of the parallel pn structure semiconductor substrate. A p base region 48 is provided on the p type partition region 43. In the p base region 48, two n ⁺ source regions 49 are provided apart from each other. Although not shown, the n ⁺ source region 49 is often in the form of a ring connected to each other at the end of the stripe. The p base region 48 is provided with a p ⁺ pickup region 50 so as to be in contact with each n ⁺ source region 49. The p ⁺ pickup region 50 occupies a part of the lower side of each n ⁺ source region 49.

A gate electrode 47 is provided on the n-type drift region 42 and the p base region 48 between the n-type drift region 42 and the n ⁺ source region 49 with a gate oxide film 46 interposed therebetween. . Source electrode 51 is in contact with p ⁺ pickup region 50 and n ⁺ source region 49. Therefore, the source electrode 51 is electrically connected to the p-type partition region 43. The drain electrode 52 is in contact with the second main surface side of the semiconductor substrate having the parallel pn structure, that is, the surface of the n ⁺ substrate 41 on the second main surface side.

  The p base region 48 protrudes to the n-type drift region 42 in the vicinity of the interface with the gate oxide film 46. Here, the width (neck length) of the portion other than the p base region 48 (the remaining (neck) portion of the n-type semiconductor) on the surface of the n-type drift region 42 is Ln3.

  As shown in FIG. 32, in the planar structure, the n-type drift region 42 and the p-type partition region 43 are provided in stripes. The gate electrode 47 is provided in a stripe shape on the surface of the semiconductor substrate having a parallel pn structure, and is connected to the adjacent gate electrode at an end (not shown). The source electrode 51 covers the gate electrode 47 in a sheet shape via an interlayer insulating film such as BPSG (not shown). Further, in the region below the gate electrode 47, the neck portion of the n-type drift region 42 is striped in a direction parallel to the longitudinal direction of the gate electrode 47.

Vertical MOS device having a parallel pn layer, the concentration N ₀ of the n-type drift region 42, the concentration P ₀ of the p-type partition regions 43, determines the breakdown voltage by the charge balance of the concentration N ₀ of the n-type drift region 42 Determines the on-resistance. Therefore, the on-resistance-breakdown voltage trade-off relationship is improved as compared with a conventional vertical MOS device using a semiconductor substrate having a single conductivity type. In particular, as shown in FIG. 32, by making the longitudinal direction of the gate electrode 47 parallel to the direction (depth direction) parallel to the interface between the n-type drift region 42 and the p-type partition region 43, wasteful current wraparound occurs. Is suppressed, and the on-resistance is greatly reduced.

  Here, in the semiconductor device shown in FIG. 32 or 33, in order to miniaturize the device, it is necessary to miniaturize the surface structure formed on the first main surface side of the semiconductor substrate having the parallel pn structure. Therefore, the width of the gate electrode 47 must be reduced. On the other hand, the p base region 48 is formed by ion implantation of a p-type impurity such as boron, for example, using the gate electrode 47 as a mask, and performing thermal diffusion. At this time, the implanted impurities are also diffused in the lateral direction and the p base region 48 protrudes to the n type drift region 42. Therefore, when the width Wn of the n type drift region 42 is reduced, the neck length of the n type drift region 42 is increased. Ln3 also narrows and the on-resistance increases. Further, there is a possibility that the neck length Ln3 becomes zero. In this case, the transistor does not turn on.

Further, when the neck length Ln3 becomes narrow due to the miniaturization of a semiconductor substrate having a parallel pn structure, the gate-drain capacitance Cgd decreases and the drain-source voltage (hereinafter referred to as dV _ds / dt) at turn-off increases. To do. Therefore, there is a problem that the trade-off relationship between the turn-off loss Eoff and dV _ds / dt deteriorates. Furthermore, since the breakdown voltage is determined by the charge balance between the concentration N ₀ of the n-type drift region 42 and the concentration P ₀ of the p-type partition region 43, there is a problem that the avalanche resistance is reduced.

In order to solve such a problem, methods have been proposed in which the width of the MOS structure formed on the first main surface side of the semiconductor substrate having the parallel pn structure is widened (for example, Patent Document 4 and Patent Document below). 5). FIG. 34 is a plan view showing the structure of the vertical MOS device of the second conventional example. FIG. 35 is a cross-sectional view showing a cross-sectional structure taken along section line AB-AB ′ in FIG. 34, in a parallel pn structure semiconductor substrate, a p-type partition region in which a p base region 48 and an n ⁺ source region 49 are formed and electrically connected to a source electrode 51 is a first p. A type partition region (first second conductivity type semiconductor region) 43a is used, and the p base region 48 and the n ⁺ source region 49 are not formed and are not electrically connected to the source electrode 51. A partition region (second second conductivity type semiconductor region) 43b is used. In the second conventional example, as shown in FIG. 35, in the repetition of the p-type partition region, one second p-type partition region 43b is provided between the first p-type partition regions 43a.

An n-type surface buffer region 45 is formed on the surfaces of the n-type drift region 42 and the second p-type partition region 43b. That is, the n-type drift region 42 adjacent to the second p-type partition region 43 b is in contact with the n-type surface buffer region 45. The gate electrode 47 is formed on the p-type base region 48 sandwiched between the n-type surface buffer region 45 and the n ⁺ source region 49 and the n-type surface buffer region 45 with a gate oxide film 46 interposed therebetween. Is provided. That is, the gate electrode 47 covers the entire surface of the n-type surface buffer region 45. Therefore, the width of the gate electrode 47 is wider than that of the first conventional example.

34 and 35, the width of the _first p-type partition region 43a is Wp1. The width of the _second p-type partition region 43b is Wp2. A width (cell pitch) between one ends of the source electrode 51 is Sc4. The neck length in the second conventional example is the portion of the surface of the n-type surface buffer region 45 other than the p base region 48, the n ⁺ source region 49 and the p ⁺ pickup region 50 (the remaining n-type surface buffer region 45 ( The width Ln4 of the neck) portion).

Here, for example, when the width of the first p-type partition region 43a and the width of the second p-type partition region 43b are the same, that is, when Wp ₁ = Wp ₂ = Wp, the neck length Ln4 is the same as that of the first conventional example. The unit column (Wn + Wp) of the parallel pn layer is increased from the neck length Ln3. The unit column of the parallel pn layer is the width when the n-type drift region 42 and the p-type partition region 43 are arranged one by one. Therefore, even if the semiconductor substrate having the parallel pn structure is miniaturized, the neck length becomes longer than that of the first conventional example, and therefore, an increase in on-resistance can be suppressed.

  In the second conventional example, in the repetition of the p-type partition region, one second p-type partition region 43b is provided between the first p-type partition regions 43a (Sc4 = 2 · (Wn + Wp)). In the case of further miniaturization, when two second p-type partition regions 43b are provided between the first p-type partition regions 43a (Sc4 = 3 · (Wn + Wp)), three second ps The present invention is also applicable when the mold partition region 43b is provided (Sc4 = 4 · (Wn + Wp)) or when the second p-type partition region 43b is provided. For this reason, since the neck length Ln4 can be further increased, an increase in on-resistance can be suppressed even if the size is reduced.

Further, by securing the neck length, the gate-drain capacitance Cgd per unit area increases, and dV _ds / dt can be lowered as compared with the first conventional example. For this reason, the trade-off relationship between the turn-off loss and dV _ds / dt can be improved.

JP-A-9-266611 US Pat. No. 5,216,275 JP 2004-119611 A JP 2001-267568 A JP 2007-13003 A

  However, in the technique of Patent Document 4 or Patent Document 5 described above, only a thin gate oxide film formed in a normal vertical MOS device is interposed between the gate electrode and the n-type surface buffer region. Therefore, the electric field concentrates on the surface of the n-type surface buffer region, and the breakdown voltage is reduced. For this reason, although the on-resistance can be suppressed, there is a problem that the on-resistance-withstand voltage trade-off relationship deteriorates because the withstand voltage decreases.

  In addition, when the width between the first p-type partition regions is increased, the channel width per unit area is reduced and the on-resistance per unit area is increased, which deteriorates the trade-off relationship between on-resistance and withstand voltage. is there.

In order to solve the above-described problems caused by the prior art, the present invention provides a semiconductor device using a semiconductor substrate having a parallel pn structure and having a good trade-off relationship with the turn-off loss -dV _ds / dt. The purpose is to improve the trade-off relationship.

To solve the above problems and achieve an object, a semiconductor device according to this invention includes a semiconductor substrate of high impurity concentration, said provided in the semiconductor substrate surface, a first conductivity type semiconductor region and the second conductive Parallel pn layers in which type semiconductor regions are alternately arranged, a second conductivity type base region provided in a surface layer of the second conductivity type semiconductor region, and a first conductivity provided in a surface layer of the base region A source electrode of a type, a gate electrode provided on the surface of the parallel pn layer via a gate oxide film, electrically connected to the source region and the base region, and provided apart from the gate electrode A source electrode, and an insulating film that is selectively provided between a surface of the parallel pn layer and the gate electrode, and is thicker than the gate oxide film. the parallel The first area of the surface of the n layer that covers the neck portion other than the base region and the source region, and the second area of the first area in which the gate electrode covers the insulating film are 0. meets 1 ≦ second area / first area ≦ 0.4, wherein the first conductivity type semiconductor region, in the direction orthogonal to the interface between the second conductivity type semiconductor region, the first conductivity type semiconductor adjacent A plurality of islands of the insulating film are provided on the surface of the neck portion between the regions, and the insulating layers are different from each other so as to cover a plurality of electric field concentration regions generated between the adjacent first conductive type semiconductor regions. A film is provided .

The semiconductor device according to this invention is the invention described above, wherein the the longitudinal direction of the gate electrode, a longitudinal and the direction parallel to the insulating film.

Further, the semiconductor device according to this invention, in the invention described above, the first conductivity type semiconductor region, in the direction orthogonal to the interface between the second conductivity type semiconductor region, the insulating film on a surface of the neck portion A plurality of islands are provided.

The semiconductor device according to this invention is the invention described above, as the gate electrode, the insulating film, a first conductive type semiconductor region, a surface parallel to a direction of the second conductive type semiconductor region In the present invention, it is formed so as to have a stripe shape.

Further, the semiconductor device according to this invention, in the invention described above, in the parallel pn layer, the second conductive type semiconductor region, a first second conductivity type semiconductor region in which the base region is formed, the The second and second conductivity type semiconductor regions in which the base region is not formed are alternately formed.

The semiconductor device according to this invention is the invention described above, during the pre SL between the first second conductive type semiconductor regions are formed at regular intervals, before Symbol first second-conductivity type semiconductor region In addition, a plurality of the second second conductivity type semiconductor regions are formed.

The semiconductor device according to this invention is the invention described above, with the semiconductor substrate, between the parallel pn layer, characterized in that the back surface buffer region of the first conductivity type is provided.

The semiconductor device according to this invention is the invention described above, with the semiconductor substrate, between the second second-conductivity type semiconductor region, the back surface buffer region of the first conductivity type is provided It is characterized by.

In the semiconductor device according to the present invention, the width, thickness and concentration of the first conductive semiconductor region are Wn, t ₂ and N ₀ in the above-described invention, and the first second conductive semiconductor The width, thickness and concentration of the region are Wp ₁ , t _3a and P ₁ , and the width, thickness and concentration of the second second conductivity type semiconductor region are Wp ₂ , t _3b and P ₂ When the thickness and concentration of the surface buffer region of the first conductivity type formed in the surface layer of the neck portion are t ₅ and N ₁ , 0.85 <(Wp ₁ · t _3a · P ₁ + Wp _{2 · t 3b · P 2) /} [2 · Wn · t 2 · N 0 + {Ln1 + (Sc1-Ln1) · Ln2 / Sc2} · t 5 · N 1] and satisfies the <1.15.

The semiconductor device according to this invention is the invention described above, the thickness t ₅ of the previous SL surface buffer _{region, t 5 <5 · t 2} (N 1 / N 0) or N _1> 0.2 · (T ₅ / t ₂ ) · N ₀ is satisfied.

The semiconductor device according to this invention is the invention described above, the semiconductor substrate is characterized by a one conductivity type of the first conductivity type or the second conductivity type.

According to the invention described above, it is possible to suppress the electric field to the first main surface side of the parallel pn layer is concentrated, it is possible to increase the breakdown voltage. Further, the increase rate of the on-resistance can be suppressed to 5% or less as compared with the case where there is no insulating film thicker than the gate oxide film on the surface of the parallel pn layer.

Further, according to the above-described invention, the channel width per unit area is increased, so that the on-resistance can be lowered.

According to the semiconductor device and the semiconductor method of the present invention, in a semiconductor device using a semiconductor substrate having a parallel pn structure and having a good trade-off relationship with the turn-off loss -dV _ds / dt, a trade-off between on-resistance and breakdown voltage. There is an effect that the relationship can be improved.

  Hereinafter, preferred embodiments of a semiconductor device according to the present invention will be described in detail with reference to the accompanying drawings. Note that, in the following description of the embodiments and all the attached drawings, the same reference numerals are given to the same components, and duplicate descriptions are omitted.

(Embodiment 1)
FIG. 1 is a plan view showing a planar structure of the semiconductor device according to the first embodiment of the present invention. 2 is a cross-sectional view showing a cross-sectional structure taken along a cutting line AA ′ in FIG. In the following description of the planar structure of the semiconductor device, the gate oxide film is omitted in order to clarify the structure of the semiconductor device.

As shown in FIG. 2, the semiconductor device according to the first embodiment is manufactured using a semiconductor substrate having a parallel pn structure. A semiconductor substrate having a parallel pn structure includes an n-type drift region (first conductivity type semiconductor region) 2 and a p-type partition region on the surface of the n ⁺ substrate 1 having a low resistivity, which is an n ⁺ drain region, on the first main surface side. A parallel pn layer made of (second conductivity type semiconductor region) 3 is provided.

A p base region 8 is provided above the p-type partition region 3. On the surface of the p base region 8, n ⁺ source regions 9 are provided at two positions apart from each other. Further, a p ⁺ pickup region 10 is provided between each n ⁺ source region 9. The p-type partition region 3 in which the p base region 8 and the n ⁺ source region 9 are formed is defined as a first p-type partition region (first second conductivity type semiconductor region) 3a. The p-type partition region 3 in which the p base region 8 and the n ⁺ source region 9 are not formed is defined as a second p-type partition region (second second conductivity type semiconductor region) 3b. In FIG. 1 and FIG. 2, one second p-type partition region 3 b is provided between the first p-type partition regions 3 a in the repetition of the p-type partition region 3. That is, the first p-type partition region 3a and the second p-type partition region 3b are alternately formed in a stripe shape with the n-type drift region 2 interposed therebetween.

An n-type surface buffer region 5 is provided on the surfaces of the n-type drift region 2 and the second p-type partition region 3b. The above-described n ⁺ substrate 1 to n-type surface buffer region 5, p base region 8, n ⁺ source region 9 and p ⁺ pickup region 10 are semiconductor substrates having a parallel pn structure.

An insulating film 16 thicker than the gate oxide film 6 is selectively provided on the surface of the n-type surface buffer region 5. In the first embodiment, an island of insulating film 16 is provided at one location on the surface of n-type surface buffer region 5 between adjacent n-type drift regions 2. Here, an electric field concentration region (electric field concentration region) is generated on the surface of the semiconductor substrate having the parallel pn structure. For this reason, the end portion of the insulating film 16 coincides with the electric field concentration region (for example, the regions B and B ′ surrounded by the broken line in FIG. 2) or the insulating film 16 covers the entire electric field concentration region. The gate electrode 7 is formed on the p-type base region 8 sandwiched between the n-type surface buffer region 5 and the n ⁺ source region 9 and the n-type surface buffer region 5. It is provided via an insulating film 16.

Source electrode 11 is provided in contact with n ⁺ source region 9 and p ⁺ pickup region 10. Therefore, the source electrode 11 is electrically connected to the first p-type partition region 3a. The drain electrode 12 is provided so as to be in contact with the second main surface side of the n ⁺ substrate 1. Here, as shown in FIG. 1, one thick insulating film 16 is formed in a region below each gate electrode 7, and the longitudinal direction of the gate electrode 7 and the longitudinal direction of the insulating film 16 are Parallel. Further, the gate electrode 7 and the insulating film 16 are striped in a direction parallel to the interface between the n-type drift region 2 and the p-type partition region 3 (hereinafter referred to as a depth direction).

  Here, the area of the gate electrode covering the active region is defined as area 1 (first area), and of the area 1, the area of the region covering the thick insulating film is defined as area 2 (second area). In Embodiment 1, the ratio of area 2 to area 1 (area 2 / area 1) needs to satisfy the following expression (1) in order to suppress the increase in on-resistance to about 5%. The reason will be described later.

    0.1 ≦ (area 2 / area 1) ≦ 0.4 (1)

Next, a method for manufacturing the semiconductor device according to the first embodiment will be described. In the semiconductor device according to the first embodiment, first, a parallel pn layer in which n-type drift regions 2 and p-type partition regions 3 are alternately formed is formed on the first main surface side of the n ⁺ substrate 1. The parallel pn layer is a method of repeatedly performing epitaxial growth of an n-type semiconductor layer for forming the n-type drift region 2 and selective ion implantation of p-type impurities for forming the p-type partition region 3 (multistage epitaxial method) ). The parallel pn layer is a method in which an n-type semiconductor to be the n-type drift region 2 is epitaxially grown in advance on the entire surface, and then a trench reaching the n ⁺ substrate 1 is provided, and the p-type semiconductor is buried in this trench (trench embedding method). May be formed. The trench filling method is preferable to the multi-stage epitaxial method because the manufacturing cost is low. Further, by exposing the surface of the parallel pn layer by polishing or the like, a semiconductor substrate having a parallel pn structure including the n ⁺ substrate 1 and the parallel pn layer is formed.

  Next, the n-type surface buffer region 5 is formed in the surface layer on the first main surface side of the semiconductor substrate having a parallel pn structure. The n-type surface buffer region 5 may be formed by epitaxial growth, or may be formed by ion implantation and thermal diffusion. Next, an island of the insulating film 16 is selectively formed on the surface of the n-type surface buffer region 5, and a gate oxide film 6 is further formed. Next, polysilicon serving as the gate electrode 7 is deposited on the insulating film 16 and the gate oxide film 6. Next, the polysilicon is patterned to remove the polysilicon on the first p-type partition region 3a, and an island of the gate electrode 7 is formed. At this time, each island of the gate electrode 7 and the insulating film 16 are formed in a stripe shape. Further, the ratio of area 2 to area 1 is set to 0.1 or more and 0.4 or less.

Next, a p base region 8 serving as a channel is formed using the gate electrode 7 as a mask, and an n ⁺ source region 9 and a p ⁺ pickup region 10 are formed using each mask. Next, although not shown, an interlayer insulating film is formed on the surfaces of the parallel pn structure semiconductor substrate and the gate electrode 7. Further, the contact hole is formed in the interlayer insulating film, the source electrode 11 is formed, and passivation is performed, whereby the semiconductor device according to the first embodiment is completed.

Next, the breakdown voltage of the semiconductor device according to the first embodiment will be described. In the semiconductor device according to the first embodiment, as shown in FIG. 1 or FIG. 2, since the insulating film 16 is provided, the concentration of the electric field on the surface of the n-type surface buffer region 5 is suppressed, and the breakdown voltage is reduced. do not do. For example, when the thickness of the gate oxide film 6 is 1000 mm and the thickness of the insulating film 16 is 4000 mm, the on-resistance is set to be similar to that of the second conventional example. For example, when the on-resistance is 18 mΩcm ² , the breakdown voltage of the semiconductor device according to the first embodiment is about 700 V or higher, whereas the breakdown voltage of the second conventional example is about 100 V. As described above, in the semiconductor device according to the first embodiment, the breakdown voltage is increased as compared with the conventional one.

FIG. 3 is a characteristic diagram showing the relationship between the normalized on-resistance, dV _ds / dt or breakdown voltage and the ratio of area 2 to area 1 (area 2 / area 1) in the semiconductor device according to the first embodiment. It is. In FIG. 3, the case where the width Wn of the n-type drift region and the width Wp of the p-type partition region are both about 4 μm or less will be described.

As shown in FIG. 3, the reason why the ratio of area 2 to area 1 is set to a value satisfying the above-described expression (1) is that when the ratio of area 2 to area 1 is smaller than 0.1, the n-type surface buffer region. This is because the electric field concentrates on the surface of the film and the withstand voltage decreases. Moreover, when the ratio of the area 2 to the area 1 is larger than 0.4, the on-resistance and dV _ds / dt rapidly increase. For this reason, the trade-off relationship of turn-off loss−dV _ds / dt is deteriorated.

  Here, in order to satisfy the above-described expression (1), it is preferable that the width Wn of the n-type drift region and the width Wp of the p-type partition region are about 4 μm or less. The reason is that in the semiconductor device according to the first embodiment, only one island of the insulating film 16 is formed below the island of each gate electrode 7. Therefore, in order for the insulating film 16 to cover the electric field concentration region and satisfy the above-described expression (1), the width between the electric field concentration regions in the n-type surface buffer region 5 (from B to B ′ shown in FIG. 2) (Width) must be narrowed. Therefore, the width of the electric field concentration region can be sufficiently narrowed by reducing the size and reducing the width Wn of the n-type drift region 2 and the width Wp of the p-type partition region to about 4 μm or less.

  Next, a modification of the semiconductor device according to the first embodiment will be described. FIG. 4 is a plan view showing a planar structure of a first modification of the semiconductor device according to the first embodiment. FIG. 5 is a cross-sectional view showing a cross-sectional structure taken along the section line CC ′ of FIG. As shown in FIG. 4 or FIG. 5, in the first modification, the n-type surface buffer region is not formed on the surfaces of the n-type drift region 2 and the second p-type partition region 3b. In the first modification, the on-resistance due to the fact that the n-type surface buffer region is not formed by reducing the impurity concentration of the p base region 8, shortening the time for performing the thermal diffusion treatment, and lowering the on-resistance. Can be offset. Other configurations are the same as those in FIG. 1 or FIG.

  According to the first modification, since there is no n-type surface buffer region, electric field concentration on the surface of a semiconductor substrate having a parallel pn structure can be reduced.

FIG. 6 is a cross-sectional view illustrating a structure of a second modification of the semiconductor device according to the first embodiment. FIG. 7 is a cross-sectional view showing a structure of a semiconductor device in which the second modification is applied to the first modification. The planar structure is the same as in FIG. 1 or FIG. As shown in FIG. 6 or 7, in the second modification, an n-type back buffer region 15 having an impurity concentration different from that of the n-type drift region 2 is provided between the parallel pn layer and the n ⁺ substrate 1. ing.

FIG. 8 is a cross-sectional view illustrating a structure of a third modification of the semiconductor device according to the first embodiment. FIG. 9 is a cross-sectional view showing the structure of a semiconductor device in which the third modification is applied to the first modification. The planar structure is the same as in FIG. 1 or FIG. As shown in FIG. 8 or FIG. 9, in the third modified example, the n-type is provided only between the p-type partition region 3 (the first p-type partition region 3 a and the second p-type partition region 3 b) and the n ⁺ substrate 1. A back buffer region 15 is provided.

According to the second modification and the third modification, the charge balance from the interface between the p-type partition region 3 and the n-type back buffer region 15 to the interface between the n-type back buffer region 15 and the n ⁺ substrate 1 is The charge balance is different from the region on the first main surface side from the interface between the p-type partition region 3 and the n-type back buffer region 15. Therefore, the avalanche resistance can be improved as compared with the first embodiment and the first modification.

Further, according to the second modification and the third modification, the thickness t _total from the surface on the first main surface side of the n ⁺ substrate to the first main surface side of the semiconductor substrate having the parallel pn structure is made the same. In this case, by making the impurity concentration of the n-type back buffer region higher than the impurity concentration of the n-type drift region, the on-resistance can be made lower than in the first embodiment and the first modification.

According to the first embodiment, in a semiconductor device manufactured using a semiconductor substrate having a parallel pn structure and having a wide MOS structure on the surface of the substrate and a good trade-off relationship of turn-off loss −dV _ds / dt, An insulating film thicker than the gate insulating film is formed between the surface and the gate electrode, and the area of the region where the gate electrode covers the insulating film relative to the area of the gate electrode where the gate electrode covers the active region is 0.1. By setting the ratio to less than 0.4, it is possible to improve the trade-off relationship between on-resistance and breakdown voltage.

(Embodiment 2)
Next, a semiconductor device according to the second embodiment will be described. FIG. 10 is a plan view illustrating a planar structure of the semiconductor device according to the second embodiment. As shown in FIG. 10, in the planar structure of a semiconductor device according to the second embodiment, the surface layer of the 2p-type partition regions 3b, island p ⁺ pickup region 10 is formed, the island of p ⁺ pickup region 10 An n ⁺ source region 9 is formed so as to surround. Further, a p base region 8 is formed so as to surround the n ⁺ source region 9, and an n-type surface buffer region 5 is formed in other regions.

The gate electrode 7 is formed on the entire surface on the first main surface side of the semiconductor substrate having a parallel pn structure. An opening is provided so as to expose the central portions of p ⁺ pickup region 10 and n ⁺ source region 9. Source electrode 11 is formed on the center of p ⁺ pickup region 10 and n ⁺ source region 9 apart from gate electrode 7. Therefore, the gate electrode 7 is formed on the outer periphery of the n ⁺ source region 9, the p base region 8, and the n-type surface buffer region 5 so that the planar shape is a mesh shape.

FIG. 11 is a cross-sectional view showing a cross-sectional structure taken along the cutting line GG ′ of FIG. As shown in FIG. 11, source electrode 11 is in contact with the central portions of p ⁺ pickup region 10 and n ⁺ source region 9. Therefore, the source electrode 11 is electrically connected to the first p-type partition region 3a. The gate electrode 7 is formed on the outer periphery of the n ⁺ source region 9, the p base region 8, and the n-type surface buffer region 5 with a gate oxide film 6 interposed therebetween.

12 is a cross-sectional view showing a cross-sectional structure taken along the cutting line HH ′ of FIG. As shown in FIG. 12, an n-type surface buffer region 5 is formed on the surface of the first p-type partition region 3a. A p base region 8 is provided on the surface of the n type surface buffer region 5 so as to be in contact with the first p type partition region 3a. Two n ⁺ source regions 9 are provided on the surface of each p base region 8, and a p ⁺ pickup region 10 is provided so as to be in contact with each n ⁺ source region 9. The gate electrode 7 is formed on the p-type base region 8 between the n-type surface buffer region 5 and the n ⁺ source region 9 and the n-type surface buffer region 5 with a gate oxide film 6 interposed therebetween. Is provided. The source electrode 11 is provided in contact with a part of the n ⁺ source region 9 and the p ⁺ pickup region 10.

  FIG. 13 is a cross-sectional view showing a cross-sectional structure taken along section line II ′ of FIG. As shown in FIG. 13, an n-type surface buffer region 5 is provided on the surface of the second p-type partition region 3b. A gate electrode 7 is provided on the entire surface of the n-type surface buffer region 5 via a gate oxide film 6.

  Next, a method for manufacturing the semiconductor device according to the second embodiment will be described. In the semiconductor device according to the second embodiment, after forming the n-type surface buffer region, a gate oxide film is formed on the entire surface of the n-type surface buffer region. Next, polysilicon for the gate electrode is formed on the surface of the gate oxide film. Further, when patterning the polysilicon for the gate electrode, openings are formed at predetermined intervals in a region parallel to the depth direction in the region above the first p-type partition region. Since the subsequent steps are the same as those in the first embodiment, description thereof is omitted.

  Next, the semiconductor device according to the second embodiment is compared with the first conventional example and the second conventional example. Here, in the first conventional example or the second conventional example, as shown in FIG. 32 or FIG. 34, the gate electrodes 47 are formed in stripes in a direction parallel to the depth direction. In the first conventional example or the second conventional example, the width (cell pitch) between one ends of the adjacent source electrodes is Sc3 or Sc4, respectively. For example, in the second conventional example, since the cell pitch Sc4 is an interval between the adjacent first p-type partition regions 3b, the direction of the cell pitch Sc4 is a direction orthogonal to the depth direction.

Therefore, for example, when the second p-type partition region 43b is formed and the interval between the adjacent first p-type partition regions 43a is increased as in the second conventional example, there is a problem that the on-resistance per unit area increases. In addition, in order to improve the trade-off relationship between the breakdown voltage and the avalanche resistance, the concentration N ₁ of the n-type surface buffer region has to be lowered, which increases the on-resistance.

  First, the on-resistance of the semiconductor device according to the second embodiment will be described. As shown in FIG. 10, the width between one ends of adjacent source electrodes in the direction orthogonal to the depth direction is defined as the cell pitch Sc1, and the remaining portion of the n-type surface buffer region 5 in the same direction is defined as the neck length Ln1. . Sc1 is given by the following equation (2).

Sc1 = 2 · Wn + Wp ₁ + Wp ₂ (2)

  Also, as shown in FIG. 10, the width between one end of adjacent source electrodes in the direction parallel to the depth direction is the cell pitch Sc2, and the remaining portion of the n-type surface buffer region 5 in the same direction is the neck length Ln2. And

  The on-resistance RonA per unit area of the semiconductor device according to the second embodiment is given by the following equation (3). However, in the equation (3), a component having the route of the neck length Ln1 in RonA is Ron1, and a component having the route of the neck length Ln2 in RonA is Ron2.

    RonA = (Sc1 · Sc2) / {(1 / Ron1) + (1 / Ron2)} (3)

Ron1 is given by the following equation (4). However, in the equation (4), the channel sheet resistance is R _ch-sh , the channel length is Lch, the resistivity of the n-type drift region 2 is ρ ₂ , the thickness of the n-type drift region 2 is t ₂ , and the n-type surface The resistivity of the buffer region 5 is ρ ₅ , and the thickness of the n-type surface buffer region 5 is t ₅ .

Ron1≡ {R _ch-sh · Lch / (Sc2−Ln2)} + {ρ ₂ · t ₂ / (2 · Wn · Sc2)} + {ρ ₅ · t ₅ / (Ln1 · (Sc2−Ln2))} ... (4)

  Ron2 is given by the following equation (5).

Ron2≡ {Rch _-sh · Lch / (Sc1-Ln1)} + {ρ ₂ · t ₂ / (2 · Ln2 · Sc1)} + {ρ ₅ · t ₅ / (Ln2 · (Sc1−Ln1))} ... (5)

  The on-resistance Ron4 · A per unit area of the semiconductor device of the second conventional example is given by the following equation (6). However, the cell pitch Sc4 of the second conventional example is the same as the cell pitch Sc1, and the neck length Ln4 of the second conventional example is the same as the neck length Ln1.

Ron4 · A≡ {(R _ch−sh · Lch) + {ρ ₂ · t ₂ / (2 · Wn)} · Sc1 + {(ρ ₅ · t ₅ / Ln1)} · Sc1 (6)

The on-resistance Ron3 · A per unit area of the semiconductor device of the first conventional example is given by the following equation (7). However, the distance between the outsides of the n ⁺ source regions in contact with the same source electrode in the first conventional example is defined as W _SOURCE .

Ron3 · A≡ {(R _ch-sh · Lch) + {ρ ₂ · t ₂ / (2 · Wn)} · Sc3 + {(ρ ₅ · t ₅ / Ln3)} · Sc3 (7)

  Here, the cell pitch Sc3 is given by the following equation (8), and the neck length Ln3 is given by the following equation (9).

    Sc3 = Sc4 / 2 = Sc1 / 2 (8)

Ln3 = Sc3- (2.Lch + W _SOURCE ) (9)

  FIG. 14 is a characteristic diagram showing the relationship between the ratio of the neck length and the on-resistance. In FIG. 14, the horizontal axis represents the ratio (Ln2 / Sc2) of the neck length Ln2 to the cell pitch Sc2 calculated using the above-described equations (2) to (9), and the vertical axis represents the above (7 ) Is an on-resistance normalized by the value of Ron3 · A shown in the equation.

  In FIG. 14, the broken line indicates the normalized on-resistance Ron3 · A of the first conventional example. A diamond mark (♦) indicates the on-resistance RonA / Ron3 · A of the semiconductor device according to the second embodiment normalized by the on-resistance Ron3 · A. Square marks (□) indicate the on-resistance Ron4 · A / Ron3 · A of the second conventional example normalized by the on-resistance Ron3 · A.

  As shown in FIG. 14, when the ratio of the neck length Ln2 is 0.1 or more and 0.7 or less, the on-resistance of the semiconductor device according to the second embodiment is lower than that of the first conventional example and the second conventional example. . The on-resistance of the semiconductor device according to the second embodiment has an optimum value when the ratio of the neck length Ln2 is approximately 0.4 to 0.5, and satisfies the following expression (10).

    RonA ≒ 0.35 ・ Ron3 ・ A ≒ 0.7 ・ Ron4 ・ A (10)

  On the other hand, the on-resistance of the semiconductor device according to the second embodiment rapidly increases when the ratio of the neck length Ln2 is 0.8 or more. The reason is that as the neck length Ln2 increases, the channel width (channel density) per unit area decreases, and the on-resistance increases accordingly.

  FIG. 15 is a characteristic diagram showing the relationship between the ratio of the neck length and the gate-drain capacitance. In FIG. 15, the vertical axis represents the Cgd value normalized by the gate-drain capacitance Cgd4 of the first conventional example, and the horizontal axis represents the ratio of the neck length to the cell pitch Sc2 (Ln2 / Sc2). In FIG. 15, the broken line indicates the normalized gate-drain capacitance Cgd3 of the first conventional example, and the rhombus (♦) indicates the gate-drain capacitance of the semiconductor device according to the second embodiment normalized by Cgd3. Cgd / Cgd3 is shown. Square marks (□) indicate the gate-drain capacitance Cgd4 / Cgd3 of the second conventional example normalized by Cgd3.

  As shown in FIG. 15, the gate-drain capacitance Cgd of the semiconductor device according to the second embodiment is larger than those of the first conventional example and the second conventional example. In addition, when the ratio of the neck length Ln2 (Ln2 / Sc2) is about 0.4 to 0.5 where the on-resistance of the semiconductor device according to the second embodiment is the optimum value, the gate-drain capacitance Cgd However, it is about 9 times the gate-drain capacitance Cgd3 of the first conventional example. On the other hand, the gate-drain capacitance Cgd4 of the second conventional example is the gate-drain capacitance of the first conventional example when the ratio of the neck length Ln2 (Ln2 / Sc2) is about 0.4 to 0.5. Only about 6.5 times Cgd3.

Here, dV _ds / dt is the gate - to reduce substantially in inverse proportion to the drain capacitance Cgd, when the turn-off loss Eoff is the same, a semiconductor device according to the second embodiment, than in the first conventional example dV _ds / dt decreases by about one digit.

  Thus, when the ratio of the neck length Ln2 (Ln2 / Sc2) is 0.1 or more and 0.7 or less, both the on-resistance and the gate-drain capacitance are both higher than those of the first and second conventional examples. It becomes good.

  Further, the ratio of the neck length Ln1 to the cell pitch Sc1 (Ln1 / Sc1) is preferably 0.1 or more and 0.7 or less. The reason is that when the ratio of the neck length Ln1 exceeds 0.7, the channel density decreases and the on-resistance increases. Further, when the ratio of the neck length Ln1 (Ln1 / Sc1) is less than 0.1, the on-resistance increases at the neck portion.

  In the second embodiment, the shape of the opening of the gate electrode 7 is a square such as a square or a rectangle. However, the present invention is not limited to this. For example, the shape of the opening of the gate electrode 7 may be a polygon, a circle, an ellipse, or the like. In this case, Ln1 and Ln2 are calculated by the average value of the width of the opening of the gate electrode 7.

  Further, the opening provided in the gate electrode 7 may be displaced in the direction orthogonal to the depth direction. That is, in the planar structure, the openings of the gate electrode 7 may be formed in a staggered pattern.

According to the second embodiment, the channel width per unit area can be increased and the on-resistance can be reduced. Therefore, the trade-off relationship between on-resistance and withstand voltage can be improved. Further, by increasing the area of the neck portion per unit area, the gate-drain capacitance Cgd is increased and dV _ds / dt is decreased, so that the trade-off relationship of turn-off loss −dV _ds / dt can be improved. .

(Embodiment 3)
FIG. 16 is a plan view of the planar structure of the semiconductor device according to the third embodiment. The semiconductor device according to the third embodiment has a structure in which the first embodiment is applied to the second embodiment. As shown in FIG. 16, in the semiconductor device according to the third embodiment, an insulating film 16 is formed between the mesh-like gate electrode 7 and the second p-type partition region 3b. The longitudinal direction of the insulating film 16 is parallel to the depth direction.

  FIG. 17 is a cross-sectional view showing a cross-sectional structure taken along the cutting line LL ′ of FIG. As shown in FIG. 17, an n-type surface buffer region 5 is formed on the entire surface of the second p-type partition region 3b, and an insulating film 16 is formed on the entire surface of the n-type surface buffer region 5. In the third embodiment, the cross-sectional structure along the cutting line JJ ′ is the same as that in FIG. Further, the cross-sectional structure at the cutting line KK ′ is the same as that in FIG.

Here, as shown in FIG. 16, the width Wy between the openings of the gate electrode 7 in the direction parallel to the depth direction is given by the following equation (11). However, in (11), the gate electrode 7, the width between the openings in the direction perpendicular to the depth direction Wx, the insulating film 16, the width in the direction orthogonal to the depth direction and W _16.

Wy ≦ (Wx−W ₁₆ ) / 2 (11)

  In the third embodiment, the depletion layer spreads so as to surround the opening of the gate electrode 7. For this reason, when the width Wy satisfies the above formula (11), it is possible to prevent the electric field from concentrating in the n-type surface buffer region 5 under the gate oxide film 6 and to prevent the breakdown voltage from being lowered.

  According to the third embodiment, the effect of combining the first embodiment and the second embodiment can be obtained.

(Embodiment 4)
Next, a semiconductor device according to Embodiment 4 will be described. FIG. 18 is a cross-sectional view illustrating the structure of the semiconductor device according to the fourth embodiment. As shown in FIG. 18, in the fourth embodiment, n second p-type partition regions 3b (two in FIG. 18) are provided between the first p-type partition regions 3a in the repetition of the p-type partition regions. It has been. That is, the cell pitch Sc1 is given by the following equation (12). However, n is 1 or more.

Sc1 = (n + 1) · Wn + Wp ₁ + n · Wp ₂ (12)

  Next, an example in which the fourth embodiment is applied to the second embodiment is compared with a second conventional example. FIG. 19 is a characteristic diagram showing the relationship between the number of second p-type partition regions between adjacent first p-type partition regions and the normalized on-resistance. In FIG. 19, the ratio of the neck length Ln2 to the cell pitch Sc2 (Ln2 / Sc2) is set to the optimum value (eg, 0.4) shown in FIG. 14, and the on-resistance Ron4 of the second conventional example is used.・ Standardized by A. Sc1 and Sc2, Ln1 and Ln2 are constant values. That is, the larger the value of the number n of the second p-type partition regions 3b, the smaller the size.

  In FIG. 19, the square mark (□) indicates the normalized on-resistance Ron4 · A≡1 of the second conventional example. A diamond mark (♦) indicates the on-resistance RonA / Ron4 · A of the semiconductor device according to the fourth embodiment normalized by the on-resistance Ron4 · A. As shown in FIG. 19, in the second conventional example, when the cell pitch is constant, the on-resistance is constant even when the degree of miniaturization increases. However, in the semiconductor device according to the fourth embodiment, the miniaturization is performed. As the degree of increases, the on-resistance decreases. In particular, when the number n of the second p-type partition regions 3b is 4 or more, the on-resistance of the semiconductor device according to the fourth embodiment has a value less than half of the on-resistance of the second conventional example.

  According to the fourth embodiment, the same effects as in the first to third embodiments can be obtained. Further, according to the fourth embodiment, the on-resistance can be lowered even if miniaturization is performed.

(Embodiment 5)
Next, a semiconductor device according to Embodiment 5 will be described. The semiconductor device according to the fifth embodiment has a structure in the case where the number n of the second p-type partition regions 3b is zero in the fourth embodiment. That is, the second p-type partition region 3b is not formed, and the n-type drift region 2 and the first p-type partition region 3a are alternately formed.

  Next, an example in which the fifth embodiment is applied to the second embodiment is compared with a first conventional example. FIG. 20 is a characteristic diagram showing the relationship between the normalized on-resistance and the ratio of the neck length Ln1 to the cell pitch Sc1 (Ln1 / Sc1) in the semiconductor device according to the fifth embodiment. In FIG. 20, the on-resistance is normalized by the on-resistance Ron3 · A of the first conventional example. As shown in FIG. 20, in the semiconductor device according to the fifth embodiment, when the ratio of the neck length Ln1 is increased, the on-resistance decreases in the range up to 0.4, and when it exceeds 0.4, the on-resistance increases. To do. When the ratio of the neck length Ln1 is 0.8 or more, the on-resistance of the first conventional example is greater.

  The reason why the on-resistance is increased is that the width of the n-type drift region 2 and the first p-type partition region 3a is increased as the proportion of the neck length Ln1 is increased. Therefore, the channel density is more decreased by increasing the widths of the n-type drift region 2 and the first p-type partition region 3a than the channel density increased by making the gate electrode 7 mesh. This is because a decrease in on-resistance is prevented by a decrease in channel density.

  FIG. 21 is a characteristic diagram showing the relationship between the normalized gate-drain capacitance Cgd and the ratio of the neck length Ln1 to the cell pitch Sc1 (Ln1 / Sc1) in the fifth embodiment. As shown in FIG. 21, the gate-drain capacitance Cgd of the semiconductor device according to the fifth embodiment is larger than the gate-drain capacitance Cgd3 of the first conventional example. As the ratio of the neck length Ln1 to the cell pitch Sc1 (Ln1 / Sc1) increases, the gate-drain capacitance Cgd of the semiconductor device according to the fifth embodiment and the gate-drain capacitance Cgd3 of the first conventional example The values of are close.

According to the fifth embodiment, when the second p-type partition region is not formed, the on-resistance is increased when the ratio (Ln1 / Sc1) of the neck length Ln1 to the cell pitch Sc1 is 0.1 or more and 0.7 or less. In addition, the trade-off relationship of turn-off loss−dV _ds / dt can be improved.

(Embodiment 6)
FIG. 22 is a plan view showing the structure of the semiconductor device according to the sixth embodiment. FIG. 23 is a cross-sectional view showing a cross-sectional structure taken along a cutting line DD ′ in FIG. As shown in FIG. 23, in the semiconductor device according to the sixth embodiment, insulating films are provided at two positions between the n-type surface buffer region 5 and the gate electrode 7 between the adjacent first p-type partition regions 3a. Sixteen islands are formed. Further, as shown in FIG. 22, the longitudinal directions of the two islands of the insulating film 16 and the longitudinal direction of the gate electrode 7 are parallel between the adjacent first p-type partition regions 3 a.

Next, the semiconductor device according to the sixth embodiment and the semiconductor device according to the first embodiment are compared. FIG. 24 is a characteristic diagram showing the relationship between the normalized on-resistance, dV _ds / dt or breakdown voltage and the ratio of area 2 to area 1 (area 2 / area 1) in the semiconductor device according to the first embodiment. It is. In the semiconductor device according to the first embodiment, FIG. 24 illustrates a case where the width Wn of the n-type drift region and the width Wp of the p-type partition region are both about 6 μm or more. As shown in FIG. 24, the ratio of area 2 to area 1 (area 2 / area 1) is about 0.2, and the on-resistance rapidly increases.

FIG. 25 is a characteristic diagram showing the relationship between the normalized on-resistance, dV _ds / dt or breakdown voltage and the ratio of area 2 to area 1 (area 2 / area 1) in the semiconductor device according to the sixth embodiment. It is. In the semiconductor device according to the sixth embodiment, FIG. 25 illustrates a case where the width Wn of the n-type drift region and the width Wp of the p-type partition region are both about 4 μm or more. As shown in FIG. 25, according to the sixth embodiment, even if the width Wn of the n-type drift region and the width Wp of the p-type partition region are about 4 μm or more, the same characteristics as in the first embodiment are exhibited. This is because in the semiconductor device according to the sixth embodiment, the islands of the insulating film 16 are formed at a plurality of locations. Therefore, even if there are a plurality of electric field concentration regions between adjacent n-type drift regions, separate islands of the insulating film 16 can be formed in each electric field concentration region, so that the area 2 is reduced. Can do.

  As shown in FIG. 25, the ratio of area 2 to area 1 (area 2 / area 1) satisfies the above-described expression (1), as in the first embodiment. In the sixth embodiment, the insulating film 16 may have more than two islands.

  FIG. 26 is a sectional view showing a modification of the semiconductor device according to the sixth embodiment. As shown in FIG. 26, for example, when the parallel pn layer is formed by the trench filling method, the heights of the n-type drift region 2 and the p-type partition region 3 are different, and the surface on the first main surface side is uneven. May be. For this reason, the electric field tends to concentrate on both ends of the convex portion, and the breakdown voltage of this portion is reduced. As shown in FIG. 26, in the sixth embodiment, since a desired number of insulating films can be formed in a desired region, by forming the insulating film 16 so as to cover both ends of the convex portion, the withstand voltage is increased. Can be suppressed.

  FIG. 27 is a plan view showing an example in which the sixth embodiment is applied to the second embodiment. As shown in FIG. 27, a plurality of (two in FIG. 27) islands of the insulating film 16 extending in the depth direction are formed between the gate electrode 7 and the neck portion.

  FIG. 28 is a cross-sectional view showing an example in which the sixth embodiment is applied to the fourth embodiment. As shown in FIG. 28, even when there are n (two in FIG. 28) second p-type partition regions 3b between the first p-type partition regions 3a and there are a plurality of electric field concentration regions, An island of the insulating film 16 can be formed for each concentrated region.

According to the sixth embodiment, the same effects as those of the first to fifth embodiments can be obtained. Furthermore, compared with Embodiment 1, in the case of the same breakdown voltage, the on-resistance can be further reduced. Further, since the gate-drain capacitance Cgd can be increased as compared with the first embodiment, the trade-off relationship of the turn-off loss −dV _ds / dt can be further improved.

(Embodiment 7)
FIG. 29 is a plan view showing a planar structure of the semiconductor device according to the seventh embodiment. As shown in FIG. 29, in the seventh embodiment, the insulating film 16 is also formed between the gate electrode 7 and the surface of the neck portion in the direction orthogonal to the depth direction. Further, the insulating film 16 is in contact with a direction parallel to the depth direction and a direction orthogonal to the depth direction.

  FIG. 30 is a cross-sectional view showing a cross-sectional structure taken along the cutting line QQ ′ in FIG. As shown in FIG. 30, in the semiconductor device according to the seventh embodiment, the insulating film 16 is formed on the surface of the neck portion of the n-type surface buffer region 5. In the semiconductor device according to the seventh embodiment, a part of the insulating film 16 may protrude from the gate electrode 7.

  According to the seventh embodiment, the concentration of the electric field on the surface of the n-type surface buffer region can be further suppressed. For this reason, the width Wy between the openings in the depth direction of the gate electrode can be increased. Therefore, even if the area where the gate electrode covers the n-type surface buffer region is increased in order to reduce the on-resistance, it is possible to prevent the breakdown voltage from being lowered.

(Embodiment 8)
FIG. 31 is a plan view showing the structure of the semiconductor device according to the eighth embodiment. As shown in FIG. 31, in the semiconductor device according to the eighth embodiment, the insulating film 16 formed on the surface of the neck portion in the direction orthogonal to the depth direction and the surface of the neck portion in the direction parallel to the depth direction. The formed insulating film 16 is separated. Further, the cross-sectional structure at the cutting line XX ′ in FIG. 31, the cross-sectional structure at the cutting line YY ′, and the cross-sectional structure at the cutting line ZZ ′ are the same as those in FIGS. 2, 30, and 17, respectively. Description is omitted.

  According to the eighth embodiment, since the ratio of area 2 to area 1 (area 2 / area 1) can be reduced as compared with the seventh embodiment, the on-resistance can be lowered.

(Embodiment 9)
Next, a semiconductor device according to Embodiment 9 will be described. The semiconductor device according to the ninth embodiment differs in the width and thickness of each semiconductor region of the parallel pn layer or the concentration of the n-type surface buffer region. In the semiconductor device according to the ninth embodiment, the width Wn of the n-type drift region, the width Wp ₁ of the 1p-type partition regions, the width Wp ₂ of the 2p-type partition regions may be the same as either the respective May be different. In the semiconductor device according to the ninth embodiment, the thickness t ₂ of the n-type drift region, the thickness t _3a of the first p-type partition region, and the thickness t _3b of the second p-type partition region are respectively It may be equivalent to either or different. In the semiconductor device according to the third embodiment, the impurity concentration N ₁ of the n-type surface buffer region and the impurity concentration N ₀ of the n-type drift region may be the same or different.

Next, an example in which the ninth embodiment is applied to the second embodiment will be described. In the semiconductor device according to the ninth embodiment, the optimum value of the withstand voltage is given by the following equation (13). However, in the equation (13), the thickness of the n-type surface buffer region is t ₅ .

Wp ₁ · t _3a · P ₁ + Wp ₂ · t _3b · P ₂ = 2 · Wn · t ₂ · N ₀ + {Ln1 + (Sc1−Ln1) · Ln2 / Sc2} · t ₅ · N ₁ (13) )

  The cell pitch Sc1 is given by the following equation (14).

Sc1≡2 · Wn + Wp ₁ + Wp ₂ (14)

  Here, the above-described equation (13) is an ideal condition for giving an optimum value of the withstand voltage. However, in practice, there are cases where the condition of the equation (13) deviates due to variations in profile shape and concentration of parallel pn layers. In order to suppress a decrease in the yield rate (yield) due to the deviation, it is necessary to satisfy the following equation (15).

0.85 <(Wp ₁ · t _3a · P ₁ + Wp ₂ · t _3b · P ₂ ) / [2 · Wn · t ₂ · N ₀ + {Ln1 + (Sc1−Ln1) · Ln2 / Sc2} · t ₅ · N ₁ ] <1.15 (15)

Further, in order to improve the trade-off relationship between the breakdown voltage and the avalanche resistance, the above-described equation (15) is satisfied, and the effective impurity concentration is slightly p-type or n-type than the above-described equation (13). It is good to make it. In order to prevent the breakdown voltage from decreasing, it is preferable that the impurity concentration N ₁ of the n-type surface buffer region is lower than the impurity concentration N ₀ of the n-type drift region.

Further, the impurity concentration N _{1 of} the n-type surface buffer region, the impurity concentration N _{0 of} the n-type drift region, the thickness t ₅ of the n-type surface buffer region, and the thickness t ₂ of the n-type drift region are as follows: The expression or the expression (17) is satisfied. The reason is that if the thickness t ₅ of the n-type surface buffer region is thicker than the right side of the equation (16), the on-resistance increases. On the other hand, when the impurity concentration N ₁ of the n-type surface buffer region is lower than the right side of the equation (17), Ln1 and Ln2 become shorter, the on-resistance increases, and the trade-off relationship of turn-off loss Eoff−dV _ds / dt This is because it gets worse.

t ₅ <5 · t ₂ (N ₁ / N ₀ ) (16)

N ₁ > 0.2 · (t ₅ / t ₂ ) · N ₀ (17)

  In the ninth embodiment, the same effect can be obtained by adjusting the thickness of each semiconductor region and the width of each semiconductor region instead of adjusting the impurity concentration of each semiconductor region.

  The impurity concentration of the n-type drift region 2, the n-type surface buffer region 5, the first p-type partition region 3a, and the second p-type partition region 3b may be uniform or non-uniform in the depth direction. When the concentration is not uniform in the depth direction, the impurity concentrations of the n-type drift region 2 and the n-type surface buffer region 5 satisfy the above-described equation (17) and are from the second main surface side to the first main surface side. Make it thinner. Alternatively, the impurity concentration of at least one of the first p-type partition region 3a and the second p-type partition region 3b is increased. By doing so, the avalanche resistance is improved and the avalanche resistance-withstand voltage trade-off relationship is improved.

  According to the ninth embodiment, the same effect as in the second embodiment can be obtained. Furthermore, the trade-off relationship of breakdown voltage-avalanche resistance-on-resistance can be improved.

In the above description of the semiconductor device, a MOSFET in which a parallel pn layer is formed on the surface of the first main surface side of the n ⁺ substrate having a low resistivity which is the n ⁺ drain region has been described. The present invention is also applicable to a structure such as an IGBT in which a parallel pn layer is formed on the surface of the low p ⁺ substrate on the first main surface side.

  In the above description of the semiconductor device, the first conductivity type is n-type and the second conductivity type is p-type. However, in the present invention, the first conductivity type is p-type and the second conductivity type is n-type. The same holds true.

  As described above, the semiconductor device according to the present invention is useful for manufacturing a high-power semiconductor element. In particular, the semiconductor device has a semiconductor substrate having a parallel pn structure and achieves both high breakdown voltage and improved on-resistance characteristics. Suitable for a semiconductor device capable of

It is a top view shown about the plane structure of the semiconductor device concerning Embodiment 1 of this invention. FIG. 2 is a cross-sectional view showing a cross-sectional structure taken along a cutting line A-A ′ in FIG. 1. FIG. 6 is a characteristic diagram showing a relationship between a normalized on-resistance, dV _ds / dt or withstand voltage and a ratio of area 2 to area 1 in the semiconductor device according to the first embodiment; FIG. 6 is a plan view showing a planar structure of a first modification of the semiconductor device according to the first embodiment; FIG. 5 is a cross-sectional view showing a cross-sectional structure taken along section line C-C ′ of FIG. 4. FIG. 7 is a cross-sectional view showing a structure of a second modification of the semiconductor device according to the first embodiment; It is sectional drawing shown about the structure of the semiconductor device which applied the 2nd modification to the 1st modification. 7 is a cross-sectional view showing a structure of a third modification of the semiconductor device according to the first embodiment; FIG. It is sectional drawing shown about the structure of the semiconductor device which applied the 3rd modification to the 1st modification. 6 is a plan view showing a planar structure of a semiconductor device according to a second embodiment; FIG. It is sectional drawing shown about the cross-section in the cutting line G-G 'of FIG. It is sectional drawing shown about the cross-sectional structure in the cutting line H-H 'of FIG. FIG. 11 is a cross-sectional view showing a cross-sectional structure taken along section line I-I ′ of FIG. 10. It is a characteristic view shown about the ratio which the neck length occupies, and ON resistance. It is a characteristic view shown about the ratio which the ratio which a neck length occupies, and the capacity | capacitance between gate-drains. FIG. 7 is a plan view showing a planar structure of a semiconductor device according to a third embodiment. FIG. 17 is a cross-sectional view showing a cross-sectional structure taken along a cutting line L-L ′ in FIG. 16. FIG. 6 is a cross-sectional view showing a structure of a semiconductor device according to a fourth embodiment. FIG. 6 is a characteristic diagram showing a relationship between the number of second p-type partition regions between adjacent first p-type partition regions and the normalized on-resistance. FIG. 10 is a characteristic diagram showing a relationship between a normalized on-resistance and a ratio of a neck length Ln1 to a cell pitch Sc1 in the semiconductor device according to the fifth embodiment; FIG. 10 is a characteristic diagram showing the relationship between the normalized gate-drain capacitance Cgd and the ratio of the neck length Ln1 to the cell pitch Sc1 in the fifth embodiment. FIG. 10 is a plan view showing a structure of a semiconductor device according to a sixth embodiment; FIG. 23 is a cross-sectional view showing a cross-sectional structure taken along a cutting line D-D ′ in FIG. 22. FIG. 6 is a characteristic diagram showing a relationship between a normalized on-resistance, dV _ds / dt or withstand voltage and a ratio of area 2 to area 1 in the semiconductor device according to the first embodiment; FIG. 16 is a characteristic diagram showing a relationship between a normalized on-resistance, dV _ds / dt or breakdown voltage and a ratio of area 2 to area 1 in the semiconductor device according to the sixth embodiment; FIG. 10 is a cross-sectional view showing a modification of the semiconductor device according to the sixth embodiment. 10 is a plan view showing an example in which Embodiment 6 is applied to Embodiment 2. FIG. 10 is a cross-sectional view showing an example in which Embodiment 6 is applied to Embodiment 4. FIG. FIG. 10 is a plan view showing a planar structure of a semiconductor device according to a seventh embodiment; FIG. 30 is a cross-sectional view showing a cross-sectional structure taken along section line Q-Q ′ of FIG. 29. FIG. 10 is a plan view showing a structure of a semiconductor device according to an eighth embodiment; It is a top view shown about the structure of the vertical MOS device of a 1st prior art example. FIG. 33 is a cross-sectional view showing a cross-sectional structure taken along a cutting line AA-AA ′ in FIG. 32. It is a top view shown about the structure of the vertical MOS device of a 2nd prior art example. FIG. 35 is a cross-sectional view showing a cross-sectional structure taken along section line AB-AB ′ in FIG. 34.

Explanation of symbols

1 n ⁺ substrate 2 n-type drift region (first conductivity type semiconductor region)
3 p-type partition region (second conductivity type semiconductor region)
3a First p-type partition region (first second conductivity type semiconductor region)
3b Second p-type partition region (second second conductivity type semiconductor region)
5 n-type surface buffer region 6 gate oxide film 7 gate electrode 8 p base region 9 n + source region 10 p + pickup region 11 source electrode 12 drain electrode 16 insulating film

Claims (11)

A semiconductor substrate having a high impurity concentration, a parallel pn layer provided on the surface of the semiconductor substrate, in which a first conductive type semiconductor region and a second conductive type semiconductor region are alternately arranged, and the second conductive type semiconductor region A second conductivity type base region provided in the surface layer; a first conductivity type source region provided in the surface layer of the base region; and a surface of the parallel pn layer provided via a gate oxide film In a semiconductor device comprising: a gate electrode; and a source electrode that is electrically connected to the source region and the base region and is provided apart from the gate electrode.
An insulating film that is selectively provided between the surface of the parallel pn layer and the gate electrode and is thicker than the gate oxide film;
A first area where the gate electrode covers a neck portion of the surface of the parallel pn layer other than the base region and the source region;
Of the first area, the second area where the gate electrode covers the insulating film,
0.1 ≦ second area / first area ≦ 0.4
The filling,
There are a plurality of islands of the insulating film on the surface of the neck portion between the adjacent first conductive type semiconductor regions in a direction orthogonal to the interface between the first conductive type semiconductor region and the second conductive type semiconductor region. Are provided,
2. The semiconductor device according to claim 1, wherein each of the different insulating films is provided so as to cover a plurality of electric field concentration regions generated between the adjacent first conductivity type semiconductor regions. The semiconductor device according to claim 1, wherein a longitudinal direction of the gate electrode and a longitudinal direction of the insulating film are parallel to each other. The insulating film is provided with a plurality of islands on a surface of the neck portion in a direction orthogonal to an interface between the first conductive semiconductor region and the second conductive semiconductor region. 2. The semiconductor device according to 2. The gate electrode and the insulating film are formed in a stripe shape in a direction parallel to an interface between the first conductive type semiconductor region and the second conductive type semiconductor region. The semiconductor device according to claim 1. In the parallel pn layer, the second conductive semiconductor region includes a first second conductive semiconductor region in which the base region is formed, and a second second conductive semiconductor region in which the base region is not formed. And the semiconductor device according to claim 1, wherein the semiconductor devices are alternately formed. The first second conductivity type semiconductor regions are formed at equal intervals, and a plurality of the second second conductivity type semiconductor regions are formed between the first second conductivity type semiconductor regions. 6. The semiconductor device according to claim 5, wherein: The semiconductor device according to claim 1, wherein a back buffer region of a first conductivity type is provided between the semiconductor substrate and the parallel pn layer. 7. The semiconductor device according to claim 5, wherein a back buffer region of a first conductivity type is provided between the semiconductor substrate and the second second conductivity type semiconductor region. The width, thickness and concentration of the first conductivity type semiconductor region are Wn, t ₂ And N ₀ And the width, thickness and concentration of the first second conductivity type semiconductor region are Wp ₁ , T _3a And P ₁ And the width, thickness and concentration of the second second-conductivity-type semiconductor region are Wp ₂ , T _3b And P ₂ And
The thickness and concentration of the surface buffer region of the first conductivity type formed in the surface layer of the neck portion are t _Five And N ₁ If it is,
0.85 <(Wp ₁ ・ T _3a ・ P ₁ + Wp ₂ ・ T _3b ・ P ₂ ) / [2 ・ Wn ・ t ₂ ・ N ₀ + {Ln1 + (Sc1-Ln1) .Ln2 / Sc2} .t _Five ・ N ₁ ] <1.15
The semiconductor device according to claim 5, wherein: Thickness t of the surface buffer region _Five But,
t _Five <5t ₂ (N ₁ / N ₀ )
Or
N ₁ > 0.2 · (t _Five / T ₂ ) ・ N ₀
The semiconductor device according to claim 9, wherein: The semiconductor device according to claim 1, wherein the semiconductor substrate is one of a first conductivity type and a second conductivity type.

Images