JP6484733B2 - IE type IGBT - Google Patents

Description

  The present application relates to a semiconductor device (or a semiconductor integrated circuit device) and a method for manufacturing a semiconductor device (or a semiconductor integrated circuit device), and more particularly to a technology effective when applied to an IGBT device technology and an IGBT manufacturing method.

  Japanese Patent Application Laid-Open No. 11-345969 (Patent Document 1) relates to an equal width active cell IE (Injection Enhancement) type IGBT (Integrated Gate Bipolar Transistor) having equidistant trenches. There is disclosed a device structure (so-called “active cell two-dimensional thinning structure”) in which an N + type emitter region is finely divided by a P + body contact region in the longitudinal direction.

  Japanese Patent Laid-Open No. 2005-294649 (Patent Document 2) relates to a wide active cell IE type IGBT in which the trench spacing in the active cell region is wider than the trench spacing in the inactive cell region. There is disclosed a technique in which a floating P-type region extending to the lower ends of trenches on both sides is provided under the inactive cell region. In this document, a floating P-type region is introduced simultaneously with a P-type body region after trench formation.

JP 11-345969 A JP 2005-294649 A

  In mono-width active cell IE-type IGBT, wide-width active cell IE-type IGBT, etc., the trench width in the active cell region and the inactive cell region is equal or the trench width in the inactive cell region is narrow, so the withstand voltage is secured. It is relatively easy to do. However, in such a structure, there is a problem that the structure becomes more complicated if the IE effect is strengthened.

  Means for solving such problems will be described below, but other problems and novel features will become apparent from the description of the present specification and the accompanying drawings.

  An outline of representative ones of the embodiments disclosed in the present application will be briefly described as follows.

  That is, the outline of one embodiment of the present application is that a body contact region is provided in a cross section having an active cell two-dimensional thinning structure, orthogonal to the substrate surface, passing through the emitter region, and orthogonal to the trenches on both sides thereof. It is not a narrow active cell IE type IGBT.

  The effects obtained by the representative ones of the embodiments disclosed in the present application will be briefly described as follows.

  That is, according to the embodiment of the present application, the IE effect can be enhanced while avoiding excessive complication of the device structure.

Narrow active cell IE type trench gate IGBT device chip cell region for explaining outline of device structure in narrow active cell IE type trench gate IGBT of main embodiment (including modification) of main application of this application and upper surface of periphery thereof It is a schematic layout diagram. It is a device schematic cross section corresponding to the A-A 'cross section of the cell region end cutout region R1 of FIG. It is a device schematic cross section corresponding to the B-B 'cross section of the cell region internal cutout region R2 of FIG. FIG. 2 is an enlarged top view of the linear unit cell region of FIG. 1 and its periphery R5 according to an embodiment of the present application. Overall top view of a narrow active cell IE type trench gate IGBT device chip of one embodiment of the present application (common to other embodiments and modifications) (almost corresponding to FIG. 1, but more specific shape) It is close to). FIG. 6 is an enlarged top view of a portion corresponding to the cell region internal cutout region R3 of FIG. 5 for explaining the device structure of the embodiment of the present application (active section dispersion structure in an active cell two-dimensional thinning structure). It is device sectional drawing corresponding to the C-C 'cross section of FIG. FIG. 7 is a device cross-sectional view corresponding to the D-D ′ cross section of FIG. 6. It is device sectional drawing corresponding to the E-E 'cross section of FIG. FIG. 8 is a device sectional view in the manufacturing process (hole barrier region introducing step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. FIG. 8 is a device sectional view in the manufacturing process (P-type floating region introduction step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. FIG. 8 is a device sectional view in the manufacturing process (trench processing hard mask film forming step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application; FIG. 8 is a device sectional view in the manufacturing process (trench hard mask processing step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. FIG. 8 is a device sectional view in the manufacturing process (trench processing step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. FIG. 8 is a device sectional view in the manufacturing process (trench processing hard mask removal step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application; FIG. 8 is a device sectional view in the manufacturing process (stretch diffusion and gate oxidation step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. FIG. 8 is a device sectional view in the manufacturing process (gate polysilicon etchback step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. FIG. 8 is a device sectional view in the manufacturing process (P-type body region and N + emitter region introduction step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. FIG. 9 is a device sectional view in the manufacturing process (P + type body contact region and P + type embedded body contact region introduction step) corresponding to FIG. 8 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application; . FIG. 8 is a device sectional view in the manufacturing process (interlayer insulating film forming step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. FIG. 8 is a device sectional view in the manufacturing process (contact hole forming step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. FIG. 8 is a device sectional view in the manufacturing process (surface metal film forming step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. FIG. 8 is a device sectional view in the manufacturing process (back grinding and back surface impurity introduction step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. FIG. 8 is a device sectional view in the manufacturing process (back surface metal electrode forming step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. Detailed description of the back side device structure of the narrow active cell IE type trench gate IGBT of the one embodiment of the present application, or a local detailed cross section of the back side of the device for explaining a device structure of a modified example (aluminum doped structure) and its manufacturing method FIG. Modification 1 on the surface side device structure of the narrow active cell IE type trench gate IGBT of the one embodiment of the present application (an enlarged top surface corresponding to FIG. 6 for explaining the N + type surface floating region & P + type surface floating region additional structure) FIG. FIG. 27 is a device cross-sectional view corresponding to the F-F ′ cross section of FIG. 26. FIG. 27 is a device cross-sectional view corresponding to the G-G ′ cross section of FIG. 26. CC 'cross section of FIG. 6 corresponding to FIG. 7 for explaining the modification 2 (simplified active cell structure) related to the surface side device structure of the narrow active cell IE type trench gate IGBT of the one embodiment of the present application. It is device sectional drawing corresponding to. The cell region end portion of FIG. 1 corresponding to FIG. 2 for explaining a modification 3 (hole collector cell additional structure) relating to the surface side device structure of the narrow active cell IE type trench gate IGBT of the one embodiment of the present application is cut out. It is a device model sectional view of an AA 'section of field R1. The linear unit cell region of FIG. 1 and its periphery R5 for explaining a modification 3 (hole collector cell additional structure) related to the surface side device structure of the narrow active cell IE type trench gate IGBT of the one embodiment of the present application. It is an enlarged top view. FIG. 7 is an enlarged top view corresponding to FIG. 6 for explaining a modification 3 (hole collector cell additional structure) related to the surface side device structure of the narrow active cell IE type trench gate IGBT of the one embodiment of the present application. FIG. 33 is a device sectional view corresponding to a section taken along line H-H ′ of FIG. 32. FIG. 33 is a device cross-sectional view corresponding to the J-J ′ cross section of FIG. 32; FIG. 33 is a device cross-sectional view corresponding to the K-K ′ cross section of FIG. 32; FIG. 2 is an enlarged top view of the linear unit cell region of FIG. 1 and its periphery R5 for explaining the outline of the device structure of the one embodiment of the present application.

[Outline of Embodiment]
First, an outline of a typical embodiment disclosed in the present application will be described.

1. Narrow active cell IE type trench gate IGBT including:
(A) a silicon-based semiconductor substrate having a first main surface and a second main surface;
(B) an IGBT cell region provided on the first main surface side of the silicon-based semiconductor substrate;
(C) a plurality of linear active cell regions and a plurality of linear inactive cell regions provided in the IGBT cell region;
(D) a plurality of active sections and a plurality of inactive sections arranged alternately along the longitudinal direction of each linear active cell region;
(E) a trench provided on the boundary between each linear active cell region and each linear inactive cell region on the first main surface of the silicon-based semiconductor substrate;
(F) a gate electrode provided in the trench via an insulating film;
(G) an emitter region having a first conductivity type, which is a surface region on the first main surface side of the silicon-based semiconductor substrate and is provided over substantially the entire area of each active section;
(H) a body contact region which is the surface region on the first main surface side of the silicon-based semiconductor substrate and has a second conductivity type provided in each inactive section;
(I) A metal emitter electrode provided on the first main surface of the silicon-based semiconductor substrate and electrically connected to the emitter region and the body contact region.

  2. In the narrow active cell IE type trench gate IGBT according to Item 1, the body contact region is provided over substantially the entire area of each inactive section.

3. The narrow active cell IE type trench gate IGBT according to item 1 or 2 further includes:
(J) The surface region on the first main surface side of the silicon-based semiconductor substrate, provided substantially over the entire area of each linear inactive cell region so as to reach the lower ends of the trenches at both ends thereof. 2 conductivity type floating region.

4). The narrow active cell IE type trench gate IGBT according to any one of Items 1 to 3, further includes:
(K) The surface region on the first main surface side of the silicon-based semiconductor substrate, which is provided in almost the entire area of each linear active cell region to a depth similar to the lower end of the trench at both ends thereof. A hole barrier region having the first conductivity type;

5. The narrow active cell IE type trench gate IGBT according to any one of Items 1 to 4 further includes:
(M) A buried body contact region having the second conductivity type provided on almost the entire lower layer of the body contact region so as to be in contact therewith.

  6). In the narrow active cell IE type trench gate IGBT according to any one of Items 1 to 5, the trench interval at both ends of each linear active cell region is 0.35 micrometers or less.

  7). In the narrow active cell IE type trench gate IGBT according to any one of Items 1 to 6, the width of each active section in the longitudinal direction is 0.5 micrometers or less.

8). The narrow active cell IE type trench gate IGBT according to any one of Items 1 to 7 further includes:
(N) A first region provided in each linear inactive region in the surface region on the first main surface side of the silicon-based semiconductor substrate and extending from the emitter region of the adjacent linear active region. Conductive surface floating area.

9. The narrow active cell IE type trench gate IGBT according to any one of Items 1 to 8 further includes:
(P) The surface region on the first main surface side of the silicon-based semiconductor substrate, provided in each linear inactive region at a position extending the body contact region of the adjacent linear active region. 2 conductivity type surface floating area.

10. The narrow active cell IE type trench gate IGBT according to any one of Items 1 to 9 further includes:
(Q) A hole collector cell region provided so as to replace every other plurality of the linear active cell regions.

11. The narrow active cell IE type trench gate IGBT according to any one of Items 1 to 10 further includes:
(R) a drift region having the first conductivity type provided from the inside to the first main surface over substantially the entire area of the silicon-based semiconductor substrate;
(S) A field stop which is provided on the second main surface side of the drift region over almost the entire area of the silicon-based semiconductor substrate, has the first conductivity type, and has a higher concentration than the drift region. region;
(T) a collector region having the second conductivity type provided on the second main surface side of the field stop region in substantially the entire area of the silicon-based semiconductor substrate;
(V) an aluminum doped region provided on the second main surface side of the collector region and having a concentration higher than that of the collector region in almost the entire area of the silicon-based semiconductor substrate;
(W) a metal collector electrode provided over substantially the entire area of the second main surface of the silicon-based semiconductor substrate;
Here, the portion of the metal collector electrode in contact with the aluminum doped region is a back metal film containing aluminum as a main component.

12 (A) a silicon-based semiconductor wafer having a first main surface and a second main surface;
(B) an IGBT cell region provided on the first main surface side of the silicon-based semiconductor wafer;
(C) a drift region provided from the inside to the first main surface and having the first conductivity type in substantially the entire region of the silicon-based semiconductor wafer;
(D) a body region having a second conductivity type, which is a surface region on the first main surface side of the silicon-based semiconductor wafer and is provided on substantially the entire surface of the IGBT cell region;
(E) a plurality of linear active cell regions and a plurality of linear inactive cell regions provided in the IGBT cell region;
(F) A plurality of active sections and a plurality of inactive sections arranged alternately along the longitudinal direction of each linear active cell region;
(G) a trench provided on the boundary between each linear active cell region and each linear inactive cell region on the first main surface of the silicon-based semiconductor wafer;
(H) a gate electrode provided in the trench via an insulating film;
(I) The emitter region having the first conductivity type, which is the surface region of the body region and is provided over substantially the entire area of each active section;
(J) A body contact region having the second conductivity type provided in each inactive section, which is the surface region of the body region;
(K) The surface region on the first main surface side of the silicon-based semiconductor wafer, provided in almost the entire region of each linear inactive cell region so as to reach the lower ends of the trenches at both ends thereof, A second conductivity type floating region deeper than the body region;
(M) a metal emitter electrode provided on the first main surface of the silicon-based semiconductor wafer and electrically connected to the emitter region and the body contact region;
A method of manufacturing a narrow active cell IE type trench gate IGBT having the following steps:
(X1) introducing a second conductivity type impurity for forming the second conductivity type floating region into the first main surface of the silicon-based semiconductor wafer;
(X2) a step of forming the trench after the step (x1);
(X3) a step of performing drive-in diffusion on the impurities introduced in the step (x1) after the step (x2);
(X4) a step of forming the gate electrode after the step (x3);
(X5) A step of introducing a second conductivity type impurity for forming the body region after the step (x4).

13. The manufacturing method of the narrow active cell IE type trench gate IGBT according to Item 12 further includes the following:
(X6) A step of introducing a first conductivity type impurity for forming a hole barrier region into the first main surface of the silicon-based semiconductor wafer before the step (x1).

  14 In the method of manufacturing the narrow active cell IE type trench gate IGBT according to the item 12 or 13, the step (x1) includes a second conductivity type impurity for forming a floating field ring provided outside the periphery of the IGBT cell region. Is also used for the introduction.

15. The method for manufacturing a narrow active cell IE type trench gate IGBT according to any one of Items 12 to 14, further includes:
(X7) After the step (x5), a step of introducing a first conductivity type impurity for forming the emitter region.

16. The manufacturing method of the narrow active cell IE type trench gate IGBT according to Item 15 further includes the following:
(X8) A step of introducing a second conductivity type impurity for forming the body contact region after the step (x7).

[Description format, basic terms, usage in this application]
1. In the present application, the description of the embodiment may be divided into a plurality of sections for convenience, if necessary, but these are not independent from each other unless otherwise specified. Each part of a single example, one part is the other part of the details, or part or all of the modifications. Moreover, as a general rule, the same part is not repeated. In addition, each component in the embodiment is not indispensable unless specifically stated otherwise, unless it is theoretically limited to the number, and obviously not in context.

  Further, in the present application, the term “semiconductor device” mainly refers to various transistors (active elements) alone, or a device in which resistors, capacitors, etc. are integrated on a semiconductor chip or the like (for example, a single crystal silicon substrate). And a semiconductor chip packaged. Here, as a representative of various transistors, a MISFET (Metal Insulator Semiconductor Effect Transistor) typified by a MOSFET (Metal Oxide Field Effect Transistor) can be exemplified. At this time, typical examples of various single transistors include power MOSFETs and IGBTs (Insulated Gate Bipolar Transistors). These are categorized into power semiconductor devices, and include power MOSFETs, IGBTs, bipolar power transistors, thyristors, power diodes, and the like.

  A typical form of the power MOSFET is a double diffused vertical power MOSFET having a source electrode on the front surface and a drain electrode on the back surface. This double diffusion type vertical power MOSFET can be classified mainly into two types. The first is a planar gate type described mainly in the embodiment, and the second is a trench gate such as a U-MOSFET. (Trench Gate) type.

  Other power MOSFETs include LD-MOSFETs (Lateral-Diffused MOSFETs).

  2. Similarly, in the description of the embodiment and the like, the material, composition, etc. may be referred to as “X consisting of A”, etc., except when clearly stated otherwise and clearly from the context, except for A It does not exclude what makes an element one of the main components. For example, as for the component, it means “X containing A as a main component”. For example, “silicon member” is not limited to pure silicon, but also includes SiGe alloys, other multi-component alloys containing silicon as a main component, and members containing other additives. Needless to say.

  Similarly, “silicon oxide film”, “silicon oxide insulating film” and the like are not only relatively pure undoped silicon oxide but also other silicon oxide as main components. Including membrane. For example, a silicon oxide insulating film doped with impurities such as TEOS-based silicon oxide (TEOS-based silicon oxide), PSG (phosphorus silicon glass), BPSG (borophosphosilicate glass) is also a silicon oxide film. In addition to a thermal oxide film and a CVD oxide film, a coating system film such as SOG (Spin On Glass) or nano-clustering silica (NSC) is also a silicon oxide film or a silicon oxide insulating film. In addition, a low-k insulating film such as FSG (Fluorosilicate Glass), SiOC (Silicon Oxide silicide), carbon-doped silicon oxide (OSD), or OSG (Organosilicate Glass) is similarly used. It is a membrane. Further, a silica-based Low-k insulating film (porous insulating film, including “porous” or “porous”) including a hole in a member similar to these is also a silicon oxide film or silicon oxide. It is a system insulating film.

  In addition to silicon oxide insulating films, silicon nitride insulating films that are commonly used in the semiconductor field include silicon nitride insulating films. Materials belonging to this system include SiN, SiCN, SiNH, SiCNH, and the like. Here, “silicon nitride” includes both SiN and SiNH unless otherwise specified. Similarly, “SiCN” includes both SiCN and SiCNH, unless otherwise specified.

  3. Similarly, suitable examples of graphics, positions, attributes, and the like are given, but it is needless to say that the present invention is not strictly limited to those cases unless explicitly stated otherwise, and unless otherwise apparent from the context.

  4). In addition, when a specific number or quantity is mentioned, a numerical value exceeding that specific number will be used unless specifically stated otherwise, unless theoretically limited to that number, or unless otherwise clearly indicated by the context. There may be a numerical value less than the specific numerical value.

  5. “Wafer” usually refers to a single crystal silicon wafer on which a semiconductor device (same as a semiconductor integrated circuit device and an electronic device) is formed, but an insulating substrate such as an epitaxial wafer, an SOI substrate, an LCD glass substrate, and the like. Needless to say, a composite wafer such as a semiconductor layer is also included.

  6). As described above for power MOSFETs, IGBTs are generally roughly classified into a planar gate type and a trench gate type. Although this trench gate type IGBT has a relatively low on-resistance, in order to further promote conductivity modulation and further lower the on-resistance, an “IE-type trench gate IGBT” using the IE (Injection Enhancement) effect (or "Active cell thinning-out type trench gate IGBT") has been developed. The IE type trench gate IGBT has an active cell (Active Cell) actually connected to an emitter electrode and an inactive cell (Inactive Cell) having a P type floating region in a cell region, or a comb-like structure. By arranging in a tooth shape, the structure is such that holes are easily accumulated on the device main surface side (emitter side) of the semiconductor substrate. Note that the P-type floating region is not essential, but if there is a P-type floating region deep enough to cover the lower ends of the trenches on both sides (that is, the P-type deep floating region), there is an advantage that the withstand voltage design becomes easy. .

  In the present application, there are a plurality of types of active cells. The first is an intrinsic active cell (specifically, a linear active cell region) that actually has an N + emitter region and in which the trench gate electrode is electrically connected to the metal gate electrode. The second is a pseudo active cell (specifically, a linear hole collector cell region) having no N + emitter region and having a trench gate electrode electrically connected to a metal emitter electrode.

  7). In the present application, among the IE-type trench gate IGBTs, a main active cell whose width is narrower than a main inactive cell is referred to as a “narrow active cell IE-type trench gate IGBT”. More generally, the pitch between the trenches of the active cells (the distance between the centers of the trenches) is smaller than the pitch between the trenches of the inactive cells.

  The direction crossing the trench gate is defined as “cell width direction”, and the extending direction (longitudinal direction) of the trench gate (linear gate portion) orthogonal thereto is defined as “cell length direction”.

  In the present application, mainly “linear unit cell regions” (for example, composed of linear active cell regions and linear inactive cell regions) are mainly handled. The “cell formation region”, that is, the “IGBT cell region” is configured by being repeatedly arranged in the internal region of the semiconductor chip.

  A cell peripheral junction region is usually provided around the cell region, and a floating field ring or a field limiting ring is provided around the cell region. Make up structure. Here, the floating field ring or field limiting ring refers to the following. In other words, the surface of the drift region (device surface) is provided separately from the P-type body region (P-type well region), and has the same conductivity type and similar concentration (reverse voltage is applied to the main junction). An impurity region or a group of impurity regions having a concentration that does not sometimes completely deplete) and surrounding a cell region in a ring shape in a single layer or multiple layers (for example, about 10 layers).

  These floating field rings may be provided with a field plate. This field plate is a conductor film pattern connected to the floating field ring, and extends above the surface (device surface) of the drift region via the insulating film, and the portion surrounding the cell region in a ring shape. say.

  For example, in the example of FIG. 5, the linear unit cell region as the periodic element constituting the cell region is as follows. That is, it is reasonable to treat a set of half-width linear inactive cell regions on both sides centering on the linear active cell region, but the linear inactive cell regions will be specifically described individually. In this case, it is inconvenient because it is separated on both sides. In that case, a specific integral part is called a linear inactive cell region.

  In the following example, in order to realize an “ultra-narrow active region”, a “contact substrate groove” that is deeper than the emitter region that is normally used is not formed, and an interlayer insulating film on a flat substrate surface is formed. A contact groove is formed. Here, in this application, the “ultra-narrow active region” means a distance between the insides of the trenches on both sides of the active cell region, that is, a width of the active region between the trenches of 0.35 μm or less. say. An active section having a width in the longitudinal direction (referred to as “active section width”) of 0.5 μm or less is referred to as an “ultra-narrow active section”.

[Details of the embodiment]
The embodiment will be further described in detail. In the drawings, the same or similar parts are denoted by the same or similar symbols or reference numerals, and description thereof will not be repeated in principle.

  In the accompanying drawings, hatching or the like may be omitted even in a cross section when it becomes complicated or when the distinction from the gap is clear. In relation to this, when it is clear from the description etc., the contour line of the background may be omitted even if the hole is planarly closed. Furthermore, even if it is not a cross section, it may be hatched to clearly indicate that it is not a void.

  In addition, regarding the designation in the case of the alternative, when one is referred to as “first” or the like and the other is referred to as “second” or the like, it is exemplified in association with the representative embodiment. Of course, for example, “first” is not limited to the illustrated option.

  As prior patent applications disclosing IE-type IGBTs having differently spaced trenches, for example, Japanese Patent Application No. 2012-19994 (Japan filing date February 1, 2012), Japanese Patent Application No. 2012-577 (Japan) Application date January 5, 2012) and Japanese Patent Application No. 2011-127305 (Japan application date June 7, 2011).

1. Description of outline of device structure in narrow active cell IE type trench gate IGBT of main embodiment (including modification) of the present application (mainly FIGS. 1 to 4)
In this section, specific examples are shown, the above definitions are supplemented, representative specific examples of the present application are extracted, their outlines are explained, and overall preliminary explanations are given. 2 and 3, in order to ensure the simplicity of the wide area diagram, the structure of a part of the impurity region is greatly simplified (see, for example, FIG. 4 for the detailed structure). .

  FIG. 1 shows a cell region of a narrow active cell IE type trench gate IGBT device chip for explaining an outline of a device structure in a narrow active cell IE type trench gate IGBT according to a main embodiment (including modifications) of the present application, and It is a surrounding upper surface schematic layout figure. FIG. 2 is a device schematic cross-sectional view corresponding to the A-A ′ cross section of the cell region end cutout region R <b> 1 of FIG. 1. 3 is a device schematic cross-sectional view corresponding to the BB ′ cross section of the cell region internal cutout region R2 of FIG. 1. FIG. 4 is a diagram of the linear unit cell region of FIG. It is an enlarged top view. Based on these, the outline of the device structure in the narrow active cell IE type trench gate IGBT of the main embodiment (including the modification) of the present application will be described.

(1) Description of the planar structure of the cell region and its periphery (mainly FIG. 1):
First, a top view of the internal region of the device chip 2 of the IE type trench gate IGBT which is the main object of the present application (the inner part of the guard ring or the like that is the outermost part of the termination structure, that is, the main part of the chip 2) is illustrated. It is shown in 1. As shown in FIG. 1, the main part of the inner region of the chip 2 (semiconductor substrate) is occupied by the IGBT cell region 10. On the outer periphery of the cell region 10, a P-type cell peripheral junction region 35 is provided so as to surround the cell region 10. Outside the cell peripheral junction region 35, a P-type floating field ring 36 (that is, a field limiting ring) is provided with an interval and presenting one or a plurality of rings, and the cell peripheral junction region 35, Together with the guard ring 4 (see FIG. 5) and the like, a termination structure for the cell region 10 is configured.

  In this example, a large number of linear unit cell regions 40 are laid in the cell region 10, and a pair of or more dummy cells (about one or several columns in one case) are provided in these end regions. A region 34 (linear dummy cell region) is arranged.

(2) Description of narrow active cell type unit cells and alternate arrangement method (mainly FIG. 2):
Next, FIG. 2 shows an AA ′ cross section of the cell region end cutout region R1 of FIG. As shown in FIG. 2, a P + type collector region 18 is provided in a semiconductor region (in this example, a silicon single crystal region) on the back surface 1b of the chip 2 (the back main surface or the second main surface of the semiconductor substrate). The metal collector electrode 17 is provided on the surface. Between the N − type drift region 20 (first conductivity type drift region) and the P + type collector region 18 (second conductivity type collector region) constituting the main part of the semiconductor substrate 2, there is an N type field stop region 19. (First conductivity type field stop region) is provided.

  On the other hand, a large number of trenches 21 are provided in the semiconductor region on the surface side 1a (the front-side main surface or the first main surface of the semiconductor substrate) of the N − type drift region 20, and a gate insulating film is included therein. A trench gate electrode 14 is embedded via 22. These trench gate electrodes 14 are connected to the metal gate electrode 5 (see FIG. 5) via the metal gate wiring 7.

  Further, these trenches 21 function to partition each region. For example, the dummy cell region 34 is partitioned from both sides by a pair of trenches 21, and the cell region 10 is formed by one of the trenches 21. The cell peripheral junction region 35 is partitioned. The cell peripheral junction region 35 is connected to the metal emitter electrode 8 through the P + type body contact region 25p. In the present application, unless otherwise specified, the thickness of the gate insulating film 22 in any part of the trench is substantially the same (however, if necessary, the thickness of a part is different from that of the other part and is different. Is not to eliminate). In this manner, by making emitter contact in the cell peripheral junction region 35 and the dummy cell region 34, even when the width of the dummy cell region 34 or the like is changed on the layout, a reduction in breakdown voltage is prevented. Can do. That is, the degree of design freedom is improved.

  A P-type floating field ring 36 is provided in the semiconductor region on the surface side 1a of the N − type drift region 20 outside the cell peripheral junction region 35, and the field plate 4 is provided on the surface 1a. The P + type body contact region 25r is connected to the floating field ring 36.

  Next, the cell region 10 will be further described. The dummy cell region 34 is basically the same in structure and size as the linear active cell region 40a except that it does not have the N + type emitter region 12, and is a P + type body provided on the surface of the P type body region 15. Contact region 25 d is connected to metal emitter electrode 8.

  Most of the inner region of the cell region 10 basically has a repetitive structure to be translated with the linear unit cell region 40 as a unit cell (note that the objectivity in a strict sense is not required. The same applies hereinafter. ) The linear unit cell region 40 as a unit lattice is composed of a linear active cell region 40a and half-width linear inactive cell regions 40i on both sides thereof, specifically, adjacent linear active cell regions 40i. It can be seen that a linear inactive cell region 40i having a full width is disposed between the layers 40a (see FIG. 4).

  A P-type body region 15 is provided in a semiconductor surface region on the front-side main surface 1a (first main surface) side of the semiconductor substrate of the linear active cell region 40a, and an N + -type emitter region 12 ( A first conductivity type emitter region) and a P + type body contact region 25 are provided. N + type emitter region 12 and P + type body contact region 25 are connected to metal emitter electrode 8. In the linear active cell region 40 a, an N-type hole barrier region 24 is provided in the N − -type drift region 20 below the P-type body region 15. In each example of the present application, when the N-type hole barrier region 24 is provided, in principle, it is provided over almost the entire area of the linear active cell region 40a in plan view. Needless to say, this is not indispensable, and may be partially provided as necessary.

  On the other hand, a P-type body region 15 is similarly provided in the semiconductor surface region on the front-side main surface 1a (first main surface) side of the semiconductor substrate in the linear inactive cell region 40i, and the N− The type drift region 20 is provided with a P-type floating region 16 (second conductivity type floating region) that covers the lower ends of the trenches 21 on both sides and is deeper than that. By providing such a P-type floating region 16, the width Wi of the linear inactive cell region can be increased without causing a sharp drop in breakdown voltage. For example, even if the layout is adjusted in order to optimize characteristics such as gate capacitance, on-voltage, and switching characteristics, there is no concern about breakdown voltage reduction, and design freedom can be ensured. For example, even if the concentration of the N-type hole barrier region 24 is increased for optimization, there is almost no influence on the breakdown voltage. This makes it possible to effectively enhance and control the hole accumulation effect. In the IE trench gate IGBT, the contact from the emitter electrode 8 to the P type floating region 16 is not formed. This is because the hole concentration in the N − type drift region 20 (N base region) below the linear active cell region 40a is reduced by blocking the direct hole discharge path from the P type floating region 16 to the emitter electrode 8. It is increasing. As a result, the on-resistance is reduced by improving the concentration of electrons injected from the MOSFET in the IGBT into the N base region.

  In this example, the width Wa of the linear active cell region 40a is narrower than the width Wi of the linear inactive cell region 40i. In the present application, this is referred to as a “narrow active cell unit cell”. Hereinafter, the device having the narrow active cell type unit cell will be specifically described below. However, the present application is not limited thereto, and the device having the “non-narrow active cell type unit cell” is also described. Needless to say, it can be applied.

  In the example of FIG. 2, the linear unit cell region 40 is configured by alternately arranging the linear active cell regions 40a and the linear inactive cell regions 40i. This is called an “arrangement method”. In the following, unless otherwise specified (specifically, basically other than FIG. 3), the description will be made on the premise of the alternating arrangement method, but it is needless to say that the “non-alternating arrangement method” may be used.

  In FIG. 2, the main part that illustratively includes each part of various embodiments of the present application has been described. However, in the following description, these are components such as a cell part (cross section, planar structure), a cell peripheral part, and the like. However, these are not individually separated, and as shown in FIG. 2, various modified examples are replaced with the respective constituent elements to constitute the main part. This applies not only to FIG. 2 but also to the next FIG.

  In FIG. 2 (alternate arrangement method), every other active cell is replaced with a hole collector cell as shown in FIG. 30 or the like. In the non-alternate arrangement method as shown in FIG. It goes without saying that substitution can also be performed.

(3) Description of non-alternate arrangement method (mainly FIG. 3):
Next, a specific example of the non-alternate arrangement type linear unit cell region 40 is shown in FIG. As shown in FIG. 3, in the example of FIG. 2, there is one linear inactive cell region 40i inserted between adjacent linear active cell regions 40a, but in the example of FIG. There are a plurality of linear inactive subcell regions 40is (device elements corresponding to the linear inactive cell region 40i in FIG. 2) inserted between the active cell regions 40a. Also in the example of the non-alternate arrangement method, the width Wa of the linear active cell region 40a is mainly smaller than the width Wis of the linear inactive subcell region 40is. This is called “narrow active cell type unit cell”. That is, the narrow active cell type unit cell is defined not by the width Wi of the linear inactive cell region 40i but by the width Wis of the linear inactive subcell region 40is. The number of linear inactive subcell regions 40is inserted between adjacent linear active cell regions 40a (hereinafter referred to as “insertion number”) does not need to be constant, and varies from one to several depending on the location. You may change between.

Similarly, in the alternate arrangement method, the number of insertions may be plural in a part.
Note that the advantage of the alternating arrangement method is that the number of trenches is small, so that the planar structure can be made relatively simple. In addition, there is an advantage that the gate capacity is not inadvertently increased. On the other hand, the merit of the non-alternating arrangement method is that a relatively wide linear inactive cell region width Wi can be set without reducing the gate capacity and reducing the breakdown voltage. Depending on the application and gate drive conditions, it may be difficult to optimize the overall design if the gate capacity is too small. Therefore, it is effective to secure a means that can be adjusted as a device design if necessary.

(4) Description of active cell two-dimensional thinning structure (mainly FIG. 4)
An example of the detailed planar structure of the main part of the linear unit cell region and the peripheral cutout region R5 in FIG. As shown in FIG. 4, for example, active sections 40aa having a constant length are provided at regular intervals in the length direction of the linear active cell region 40a, and an N + -type emitter region 12 is provided between them. There is no inactive section 40ai. That is, a part of the linear active cell region 40a in the length direction is an active section 40aa in a locally dispersed manner. More specifically, the active section 40aa is provided with the N + -type emitter region 12 on almost the entire surface, and the inactive section 40ai is provided on the almost entire surface with the P + -type body contact region 25 and the P + -type buried body contact region 55. Is provided. On the other hand, in the linear inactive cell region 40i, a P-type body region 15 and a P-type floating region 16 (second conductivity type floating region) are provided on almost the entire surface.

  Here, being distributed at a constant interval and having a constant length means periodic, but being substantially periodic corresponds to a local dispersive distribution, Being distributed is broader and does not necessarily mean periodic or quasi-periodic.

2. Description of device structure of narrow active cell IE type trench gate IGBT in one embodiment (P type deep floating & hole barrier combination structure) (mainly FIGS. 5 to 9)
In this section, based on the description of section 1, an example of a specific chip top surface layout and unit cell structure (active cell one-dimensional thinning structure) common to each embodiment (FIG. 1, FIG. 2, and FIG. 4 of section 1). Will be explained. The cell structure described in this section is an alternating array type narrow active cell unit cell.

  Normally, taking the IGBT element 2 having a withstand voltage of 1200 volts as an example, the chip size varies greatly depending on the assumed current value, such as 3 to 15 millimeters square. Here, for convenience of explanation, a chip having a length of 4 mm and a width of 5.2 mm will be described as an example. Here, description will be made assuming that the breakdown voltage of the device is about 1200 volts, for example.

  FIG. 5 is an overall top view of the narrow active cell IE-type trench gate IGBT device chip according to the embodiment of the present invention (common to other embodiments and modifications) (almost corresponding to FIG. 1, but more specifically, It is close to a typical shape. 6 is an enlarged top view of a portion corresponding to the cell region internal cutout region R3 of FIG. 5 for explaining the device structure of the embodiment of the present invention (active section dispersion structure in the active cell two-dimensional thinning structure). . FIG. 7 is a device cross-sectional view corresponding to the C-C ′ cross section of FIG. 6. FIG. 8 is a device cross-sectional view corresponding to the D-D ′ cross section of FIG. 6. FIG. 9 is a device cross-sectional view corresponding to the E-E ′ cross section of FIG. 6. Based on these, the device structure of the narrow active cell IE type trench gate IGBT in one embodiment (P type deep floating & hole barrier combination structure) of the present application will be described.

  As shown in FIG. 5, an annular guard ring 3 made of, for example, an aluminum-based wiring layer is provided on the outer peripheral portion of the upper surface 1a of the IGBT device chip 2, and an annular floating ring is provided on the inner side thereof. Several (single or plural) annular field plates 4 (for example, composed of the same aluminum-based wiring layer as the above) connected to a field ring or the like are provided. A cell region 10 is provided inside the field plate 4 (floating field ring 36) and in the main part of the inner region of the upper surface 1a of the chip 2, and the cell region 10 extends to the vicinity of the outside, for example, The metal emitter electrode 8 is made of the same aluminum-based wiring layer as before. A central portion of the metal emitter electrode 8 serves as a metal emitter pad 9 for connecting a bonding wire or the like. Between the metal emitter electrode 8 and the field plate 4, for example, the same aluminum-based wiring layer as above. A configured metal gate wiring 7 is arranged. The metal gate wiring 7 is connected to, for example, a metal gate electrode 5 composed of the same aluminum-based wiring layer or the like, and the central portion of the metal gate electrode 5 is a gate pad for connecting a bonding wire or the like. 6

  Next, an enlarged planar layout (mainly showing the layout of the surface region of the semiconductor substrate) of the cell region internal cutout region R3 of FIG. 5 is shown in FIG. As shown in FIG. 6, the N + type emitter region 12 is not formed over almost the entire length of the linear active cell region 40a, and the N + type emitter region 12 is formed in the length direction thereof. The active section 40aa and the inactive section 40ai in which the N + type emitter region 12 is not formed are divided approximately periodically. That is, in the active section 40aa of the linear active cell region 40a, the N + type emitter region 12 is provided almost on the entire surface, and in the inactive section 40ai of the linear active cell region 40a, the P + type is provided on the almost entire surface. A body contact region 25 is provided. On the other hand, in the linear inactive cell region 40i separated by the linear active cell region 40a and the trench gate electrode 14, a P-type body region 15 and a P-type floating region 16 are provided on almost the entire surface.

  Next, FIG. 7 shows a C-C ′ cross section in FIG. 6. As shown in FIG. 7, a P + type collector region 18 and an N type field stop region 19 are formed in the semiconductor region of the back surface 1 b of the semiconductor chip 2 so as to be in contact with the top and bottom, and on the back surface 1 b of the semiconductor chip 2. The metal collector electrode 17 is formed.

  An N-type hole barrier region 24 is arranged in this order from the bottom in the N − type drift region 20 (surface side semiconductor region of the semiconductor substrate) on the surface 1a (first main surface) side of the semiconductor chip 2 in the linear active cell region 40a. , A P-type body region 15 and an N + -type emitter region 12 are provided. An interlayer insulating film 26 is formed on the surface 1a of the semiconductor chip 2, and a contact trench 11 (or contact hole) is formed in the interlayer insulating film 26 portion in the linear active cell region 40a. . The N + -type emitter region 12 is connected to the metal emitter electrode 8 provided on the interlayer insulating film 26 through the contact groove 11 and the like. The presence of the N-type hole barrier region 24 is optional, but in addition to acting as a hole barrier, the presence of the N-type hole barrier region 24 also increases the P-type floating when the width of the linear active cell region 40a becomes very narrow. There is an effect of preventing the region 16 from undesirably spreading toward the linear active cell region 40a. In addition, providing the N-type hole barrier region 24 has an advantage that a sufficient IE effect can be realized even when the depth of the trench is not so deep (for example, about 3 micrometers). In addition, there is also an effect that the characteristic fluctuation range with respect to the variation in trench depth can be greatly reduced.

  Here, the N-type hole barrier region 24 is a barrier region for preventing holes from flowing into the passage from the N− type drift region 20 to the N + type emitter region 12, and the impurity concentration thereof is, for example, N + type. It is lower than the emitter region 12 and higher than the N − type drift region 20. Due to the presence of the N-type hole barrier region 24, holes accumulated in the linear inactive cell region 40i are directed to the emitter passage (from the N− type drift region 20 to the P + type body contact region 25) of the linear active cell region 40a. It is possible to effectively prevent entry into the passage. Further, by locally disposing the N-type hole barrier region 24 only in the active cell region 40a, it is possible to prevent the discharge resistance against holes from being increased unnecessarily at the time of switching off, and to prevent the switching characteristics from deteriorating.

  In contrast, in the linear inactive cell region 40i, the N − type drift region 20 (the surface side semiconductor region of the semiconductor substrate) on the surface 1a (first main surface) side of the semiconductor chip 2 is sequentially arranged from the bottom. A P-type floating region 16 and a P-type body region 15 are provided. The depth of the P-type floating region 16 is deeper than the depth of the trench 21, and is distributed so as to cover the lower end portion of the trench 21. In this way, it is possible to effectively prevent the electric field strength from concentrating on the lower end portion of the trench 21 in the off state.

  Next, FIG. 8 shows a D-D ′ cross section of FIG. 6. As shown in FIG. 8, the difference between this cross section and FIG. 7 is that a P + type body contact region 25 is provided on the surface of the P type body region 15 in the linear active cell region 40a and is in contact therewith. A P + type buried body contact region 55 is provided so as to overlap. The other parts are the same as those in FIG.

  Next, FIG. 9 shows an E-E ′ cross section of FIG. 6. As shown in FIG. 9, a P + type collector region 18 and an N type field stop region 19 are formed in the semiconductor region on the back surface 1 b of the semiconductor chip 2 so as to be in contact with the top and bottom, and on the back surface 1 b of the semiconductor chip 2. The metal collector electrode 17 is formed.

  In the active section 40aa of the linear active cell region 40a, the N − type drift region 20 (surface side semiconductor region of the semiconductor substrate) on the surface 1a (first main surface) side of the semiconductor chip 2 is arranged in order from the bottom. A hole barrier region 24, a P-type body region 15 and an N + type emitter region 12 are provided. On the other hand, in the inactive section 40ai of the linear active cell region 40a, the N − type drift region 20 (the semiconductor region on the surface side of the semiconductor substrate) on the surface 1a (first main surface) side of the semiconductor chip 2 is arranged in order from the bottom. , An N-type hole barrier region 24, a P-type body region 15, a P + type buried body contact region 55, and a P + type body contact region 25 are provided. Similar to the above, a contact groove 11 (or contact hole) is formed on the surface 1 a of the semiconductor chip 2. The N + type emitter region 12 and the P + type body contact region 25 are connected to the metal emitter electrode 8 through the contact groove 11 and the like.

  Here, in order to more specifically illustrate the device structure, examples of main dimensions and main parameters of each part of the device (see FIGS. 2 and 4) are shown. That is, the width Wa of the linear active cell region is about 1.0 μm, and the width Wi of the linear inactive cell region is about 2.5 μm (the width Wa of the linear active cell region is linear in It is desirable that the width is smaller than the width Wi of the active cell region, and the value of “Wi / Wa” is particularly preferably in the range of 2 to 3, for example. The contact width is about 1.0 μm, the trench width is about 0.7 μm (particularly less than 0.8 μm is particularly suitable), the trench depth is about 3 μm, and the N + type emitter region 12 Is about 0.6 micrometers, and the depth of the P-type body region 15 (channel region) is about 1.2 micrometers. The depth of the P-type floating region 16 is about 4.5 micrometers, the thickness of the N-type field stop region 19 is about 1.5 micrometers, and the thickness of the P + type collector region is about 0.5 micrometers. The thickness of the semiconductor substrate 2 is about 120 micrometers (here, an example with a breakdown voltage of about 1200 volts is shown). Note that the thickness of the semiconductor substrate 2 strongly depends on the required breakdown voltage. Therefore, at a withstand voltage of 600 volts, it is about 70 micrometers, for example, and at a withstand voltage of 400 volts, it is about 40 micrometers, for example. Further, the recess depth of the upper surface of the trench gate electrode 14 is, for example, about 0.4 micrometers, and the distance between the trenches on both sides in the linear active cell region 40a (the distance between the insides of the trenches) is, for example, 0.3 micrometers. Degree. The thickness of the P + type body contact region 25 is, for example, about 0.4 micrometers, and the thickness of the P + type embedded body contact region is, for example, about 0.5 micrometers. The width of the active section 40aa in the linear active cell region 40a is approximately 0.4 micrometers, for example, and the inactive section 40ai is strongly dependent on the required saturation current value, but is approximately 10 micrometers, for example. Further, the resistivity of the N − type drift region 20 is, for example, about 70 Ωcm.

  The width of the active section 40aa is preferably 0.5 micrometers or less. In this case, the travel distance of the hole passing through the P-type body region 15 under the N + -type emitter region 12 is estimated to be 0.25 micrometers or less, and is at a level with no problem considering the latch-up resistance.

  In the following example and the example in section 1, the dimensions of the corresponding parts are substantially the same as those shown here, and therefore description thereof will not be repeated.

3. Description of the main manufacturing process corresponding to the manufacturing method of the narrow active cell IE type trench gate IGBT of the one embodiment of the present application (mainly FIGS. 10 to 24)
In this section, an example of a manufacturing method for the device structure described in Section 2 is shown. In the following description, the cell region 10 will be mainly described. For the peripheral portion and the like, reference is made to FIG. 1, FIG. 2, FIG.

  FIG. 10 is a device sectional view in the manufacturing process (hole barrier region introducing step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. FIG. 11 is a device sectional view in the manufacturing process (P-type floating region introduction step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. FIG. 12 is a device sectional view in the manufacturing process (trench processing hard mask film forming step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. FIG. 13 is a device sectional view in the manufacturing process (trench hard mask processing step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. FIG. 14 is a device sectional view in the manufacturing process (trench processing step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. FIG. 15 is a device sectional view in the manufacturing process (trench processing hard mask removal step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the embodiment of the present application. FIG. 16 is a device sectional view in the manufacturing process (stretch diffusion and gate oxidation step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. FIG. 17 is a device sectional view in the manufacturing process (gate polysilicon etch-back step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. 18 is a device sectional view in the manufacturing process (P-type body region and N + type emitter region introducing step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. . FIG. 19 is a device cross section in the manufacturing process (P + type body contact region and P + type embedded body contact region introducing step) corresponding to FIG. 8 for describing the manufacturing method corresponding to the device structure of the embodiment of the present application. FIG. FIG. 20 is a device sectional view in the manufacturing process (interlayer insulating film forming step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. FIG. 21 is a device sectional view in the manufacturing process (contact hole forming step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. FIG. 22 is a device sectional view in the manufacturing process (surface metal film forming step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. FIG. 23 is a device sectional view in the manufacturing process (back grinding and back surface impurity introduction step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. FIG. 24 is a device sectional view in the manufacturing process (back surface metal electrode forming step) corresponding to FIG. 7 for describing the manufacturing method corresponding to the device structure of the one embodiment of the present application. Based on these, the main manufacturing process corresponding to the manufacturing method of the narrow active cell IE type trench gate IGBT of the one embodiment of the present application will be described.

First, a 200φ wafer (a wafer having various diameters such as 150φ, 100φ, 300φ, 450φ, etc.) of N-type silicon single crystal (for example, phosphorus concentration of about 2 × 10 ¹⁴ / cm ³ ) is prepared. Here, for example, a wafer by FZ (Floating Zone) method is most suitable, but a wafer by CZ (Czochralski) method may be used. This is because it is easier to obtain a high-resistance wafer having a relatively high quality and a stable concentration by using the FZ method.
On the other hand, when the CZ crystal is annealed at around 450 degrees Celsius, there is a problem that a thermal donor is generated and a substantial N-type impurity ability is increased. Therefore, in this case, among the CZ crystals, it is preferable to use an MCZ (Magnetic Field Applied CZ) method having a relatively low oxygen concentration. Among the MCZ crystals, crystals by the HMCZ (Horizontal MCZ) method, the CMCZ (Cusp MCZ) method and the like are particularly suitable. Oxygen concentration in these hypoxic MCZ crystals are usually, 7x10 ¹⁷ / cm ³ order of 3x10 ¹⁷ / cm ^3. On the other hand, the oxygen concentration of a FZ (Floating Zone) crystal is usually about 1 × 10 ¹⁶ / cm ³ , and the oxygen concentration of a normal CZ crystal that does not use a magnetic field is usually about 1 × 10 ¹⁸ / cm ^3. .

  In the IE-type trench gate IGBT of each embodiment of the present application, even a crystal by the CZ method is generally allowed as a product and device design is possible. The IGBT with enhanced IE effect has a relatively flat overall hole distribution in the on-state due to the hole accumulation effect on the front side. Therefore, even if there is a variation in crystal resistivity, it has an effect on switching loss. This is because is small. Note that the resistivity range of the high-resistance CZ crystal particularly suited to the IGBT is, for example, a range of about 20 Ωcm to about 85 Ωcm assuming a breakdown voltage range of about 600 volts to 1200 volts.

  Here, in the IGBT, when a CZ crystal is used, unlike the FZ crystal having a low oxygen concentration, there is an advantage that the mechanical strength is high and the thermal strain is strong. In addition, compared with the FZ crystal, the CZ crystal has an advantage that it is relatively easy to increase the diameter of the wafer. In addition, since the problem of thermal stress becomes more important as the diameter is increased, it is more advantageous to use a CZ crystal in terms of measures against thermal stress. By applying the structure of the present application, the FZ crystal and the CZ crystal can be properly used depending on the situation.

Next, as shown in FIG. 10, an N-type hole barrier region introduction resist film 31 is formed on almost the entire surface 1a (first main surface) of the semiconductor wafer 1 by coating or the like, and by ordinary lithography. Patterning. Using the patterned N-type hole barrier region introducing resist film 31 as a mask, for example, by ion implantation, the semiconductor substrate 1s (N− type single crystal silicon substrate) on the surface 1a (first main surface) side of the semiconductor wafer 1 is used. An N-type hole barrier region 24 is formed by introducing N-type impurities therein. As ion implantation conditions at this time, for example, ion species: phosphorus, dose amount: about 6 × 10 ¹² / cm ² , and implantation energy: about 80 KeV can be exemplified as preferable ones. Thereafter, the resist film 31 that has become unnecessary is removed by ashing or the like. Thus, introducing the N-type hole barrier region 24 before forming the trench is advantageous in controlling the depth and lateral extension thereof.

Next, as shown in FIG. 11, a P-type floating region introducing resist film 37 is formed on almost the entire surface 1a of the semiconductor wafer 1 by coating or the like and patterned by ordinary lithography. Using the patterned P-type floating region introducing resist film 37 as a mask, P-type impurities are introduced into the semiconductor substrate 1s on the surface 1a (first main surface) side of the semiconductor wafer 1, for example, by ion implantation. Thus, the P-type floating region 16 is formed. As ion implantation conditions at this time, for example, ion species: boron, dose amount: about 3.5 × 10 ¹³ / cm ² , and implantation energy: about 75 KeV can be exemplified as preferable ones. Thereafter, the resist film 37 that has become unnecessary is removed by ashing or the like. Thereafter, activation annealing or the like is performed as necessary (for example, 900 degrees Celsius, about 30 minutes). When the P-type floating region 16 is introduced, the cell peripheral junction region 35 and the floating field ring 36 shown in FIG. As described above, introducing the P-type floating region 16 before forming the trench is advantageous in controlling the depth and lateral extension thereof. Needless to say, the introduction timing of the N-type hole barrier region 24 and the P-type floating region 16 can be reversed.

  Next, as shown in FIG. 12, a trench forming hard mask film 32 such as a silicon oxide insulating film (for example, by CVD (Chemical Vapor Deposition) or the like is formed on almost the entire surface 1 a of the semiconductor wafer 1. A thickness of about 450 nm) is formed.

  Next, as shown in FIG. 13, a resist film 33 for trench hard mask film processing is formed by coating or the like on almost the entire surface 1a of the semiconductor wafer 1, and is patterned by normal lithography. Using the patterned trench hard mask film processing resist film 33 as a mask, the trench forming hard mask film 32 is patterned by, for example, dry etching. Thereafter, the unnecessary resist film 33 is removed by ashing or the like.

Next, as shown in FIG. 14, the trench 21 is formed by anisotropic dry etching, for example, using the patterned trench forming hard mask film 32. As a gas system for this anisotropic dry etching, for example, a Cl ₂ / O ₂ system gas can be exemplified as a suitable one.

  After that, as shown in FIG. 15, the trench forming hard mask film 32 that has become unnecessary is removed by wet etching using, for example, a hydrofluoric acid based silicon oxide film etching solution.

  Next, as shown in FIG. 16, extended diffusion (for example, about 1200 degrees Celsius and about 30 minutes) is performed on the P-type floating region 16 and the N-type hole barrier region 24. As described above, it is advantageous to control the depth and the lateral spread of the P-type floating region 16 and the N-type hole barrier region 24 after the trench formation.

  Subsequently, a gate insulating film 22 (for example, a thickness of about 120 nm) is formed on the entire surface 1a of the semiconductor wafer 1 and almost the entire inner surface of the trench 21 by, for example, thermal oxidation, CVD, or both.

Next, as shown in FIG. 17, phosphorus is doped on the surface 1a of the semiconductor wafer 1 on the gate insulating film 22 and almost the entire inner surface of the trench 21 by, for example, CVD so as to fill the trench 21. A doped polysilicon (Doped Poly-Silicon) film 27 is formed (for example, a thickness of about 600 nm). Next, the trench gate electrode 14 is formed in the trench 21 by, for example, etching back the polysilicon film 27 by dry etching or the like (for example, the gas system is SF ₆ or the like).

Next, as shown in FIG. 18, a P-type body region introducing resist film 38 is formed on the surface 1a of the semiconductor wafer 1 by ordinary lithography. Using this P-type body region introducing resist film 38 as a mask, P-type body region 15 is formed by introducing P-type impurities into almost the entire surface of cell region 10 and other necessary portions by, for example, ion implantation. . As ion implantation conditions at this time, for example, ion species: boron, dose amount: about ² × 10 ¹³ / cm ² , and implantation energy: about 250 KeV can be exemplified as preferable ones. Thereafter, the unnecessary P-type body region introduction resist film 38 is removed by ashing or the like. Thereafter, stretching diffusion (for example, about 1000 degrees Celsius and about 100 minutes) is performed on the P-type body region 15. Introducing the P-type body region 15 after the extension and diffusion of the P-type floating region 16 and the N-type hole barrier region 24 is effective for controlling the profile thereof.

Next, a resist film 39 for introducing an N + type emitter region is formed on the surface 1a of the semiconductor wafer 1 by ordinary lithography. Using this N + type emitter region introduction resist film 39 as a mask, N type impurities are introduced into almost the entire upper surface of the P type body region 15 in the active section 40aa of the linear active cell region 40a, for example, by ion implantation. Thereby, the N + type emitter region 12 is formed. Due to the characteristics of the structure of the present application, the surface of the trench gate electrode 14 is recessed from the surface to a slightly deep position (for example, about 0.40 μm), and accordingly, the N + type emitter region 12 is also formed to a relatively deep position. There is a need to. The ion implantation conditions at this time include, for example, ion species: phosphorus, dose amount: about 1 × 10 ¹⁴ / cm ² , implantation energy: about 175 KeV, ion species: arsenic, dose amount: about 5 × 10 ¹⁵ / cm ² , Implantation energy: A two-stage ion implantation that is about 80 KeV can be illustrated as a preferable example. Thereafter, the resist film 39 for introducing the N + type emitter region that has become unnecessary is removed by ashing or the like.

Next, as shown in FIG. 19, a resist film 56 for introducing a P + type body contact region or the like is formed on the surface 1a of the semiconductor wafer 1 by ordinary lithography. Using this P + type body contact region introduction resist film 56 as a mask, for example, by ion implantation, a P type impurity is formed on almost the entire upper surface of the P type body region 15 in the inactive section 40ai of the linear active cell region 40a. As a result, the P + type body contact region 25 is formed. As ion implantation conditions at this time, for example, ion species: BF ₂ , dose: about 5 × 10 ¹⁵ / cm ² , and implantation energy: about 80 KeV can be exemplified as preferable examples.

Next, using this P + type body contact region introducing resist film 56 as a mask, for example, by ion implantation, almost over the entire upper surface of the P type body region 15 in the inactive section 40ai of the linear active cell region 40a, A P + type buried body contact region 55 is formed by introducing a P type impurity. As ion implantation conditions at this time, for example, ion species: boron, dose amount: about 3 × 10 ¹⁵ / cm ² , and implantation energy: about 80 KeV can be exemplified as preferable ones. Thereafter, the resist film 56 for introducing the P + type body contact region and the like that has become unnecessary is removed by ashing or the like. Thereafter, activation annealing of the N + type emitter region 12, the P + type body contact region 25, and the P + type buried body contact region 55 is performed (for example, about 950 degrees Celsius for about 60 minutes). Needless to say, the order of introduction of the N + type emitter region 12, the P + type body contact region 25, and the P + type buried body contact region 55 can be replaced. The introduction of the N + type emitter region 12, the P + type body contact region 25, and the P + type buried body contact region 55 after the extension diffusion to the P type body region 15 is effective in controlling these profiles. The P + type embedded body contact region 55 is not essential, but the presence of the P + type embedded body contact region 55 is effective in improving latch-up resistance. Furthermore, the P + type body contact region 25 and the P + type buried body contact region 55 can be formed by a single ion implantation. However, the two-stage method is easier to control the concentration distribution and is particularly suitable for improving the latch-up resistance.

  Next, as shown in FIG. 20, for example, a PSG (Phosphosilicate Glass) film is formed as the interlayer insulating film 26 on the almost entire surface 1 a of the semiconductor wafer 1 by, for example, CVD (thickness is). For example, about 600 nm). As the material of the interlayer insulating film 26, in addition to the PSG film, a BPSG (Borophosphosilicate Glass) film, an NSG (Non-doped Silicate Glass) film, an SOG (Spin-On-Glass) film, or a composite film thereof is preferable. Can be exemplified.

Next, as shown in FIG. 21, a contact groove forming resist film is formed on the surface 1a of the semiconductor wafer 1 on the interlayer insulating film 26 by ordinary lithography. Subsequently, the contact groove 11 (or contact hole) is formed by, for example, anisotropic dry etching or the like (gas system is, for example, Ar / CHF ₃ / CF ₄ or the like). Thereafter, the unnecessary resist film is removed by ashing or the like.

  Next, as shown in FIG. 22, for example, an aluminum-based electrode film 8 (which becomes the metal emitter electrode 8) is formed by sputtering or the like. Specifically, for example, the following procedure is executed. First, a TiW film (for example, a thickness of about 200 nm) is formed as a barrier metal film on almost the entire surface 1a of the semiconductor wafer 1 by, for example, sputtering film formation (many portions of titanium in the TiW film are Subsequent heat treatment moves to the silicon interface to form silicide and contributes to improvement of contact characteristics, but these processes are complicated and are not shown in the drawing).

Subsequently, for example, silicide annealing is performed for about 30 minutes in a nitrogen atmosphere at about 650 degrees Celsius. Subsequently, aluminum is used as a main component (for example, several percent of silicon is added, and the rest is aluminum) by sputtering, for example, so as to fill the contact groove 11 almost over the entire surface of the barrier metal film. For example, a thickness of about 5 micrometers is formed. Subsequently, the metal emitter electrode 8 made of an aluminum metal film and a barrier metal film is patterned by normal lithography (for example, Cl ₂ / BCl ₃ as a dry etching gas system). Furthermore, as a final passivation film, for example, an organic film (for example, a thickness of about 2.5 micrometers) or the like containing polyimide as a main component is applied to almost the entire device surface 1a of the wafer 1, and by normal lithography, The emitter pad 9 and the gate pad 6 shown in FIG. 5 are opened.

  Next, a back grinding process (chemical etching or the like for removing damage on the back surface is also performed if necessary) is performed on the back surface 1b of the wafer 1, for example, about the original 800 micrometers (preferable). For example, the thickness of the wafer is about 1000 to 450 micrometers, and is reduced to about 200 to 30 micrometers, for example. For example, if the withstand voltage is about 1200 volts, the final thickness is about 120 micrometers.

Next, as shown in FIG. 23, an N-type field stop region 19 is formed on almost the entire back surface 1b of the semiconductor wafer 1 by introducing N-type impurities, for example, by ion implantation. Here, as ion implantation conditions, for example, ion species: phosphorus, dose: about 7 × 10 ¹² / cm ² , and implantation energy: about 350 KeV can be exemplified as preferable examples. Thereafter, laser annealing or the like is performed on the back surface 1b of the wafer 1 for impurity activation as necessary. Next, a P + type collector region 18 is formed on almost the entire back surface 1b of the semiconductor wafer 1 by introducing N-type impurities, for example, by ion implantation. Here, as ion implantation conditions, for example, ion species: boron, dose amount: about 1 × 10 ¹³ / cm ² , and implantation energy: about 40 KeV can be exemplified as preferable examples. Thereafter, laser annealing or the like is performed on the back surface 1b of the wafer 1 for impurity activation as necessary. Here, for activation annealing for back surface ion implantation, by optimizing the laser annealing conditions, a crystal generated by back surface ion implantation in a portion close to the boundary between the N-type field stop region 19 and the N − -type drift region 20. Defects can be intentionally left behind. This remaining crystal defect functions as a local lifetime control layer and contributes to the improvement of the trade-off characteristic of switching performance-on voltage. Here, as annealing conditions (laser irradiation conditions), for example, annealing method: laser irradiation from the back surface 1b side of the wafer 1, wavelength: 527 nm, pulse width: about 100 ns, energy density: about 1.8 J / cm ² , Irradiation method: two-pulse method, delay time of both pulses: about 500 ns, pulse overlap rate: about 66% can be exemplified as preferable examples.

  Next, as shown in FIG. 24, the metal collector electrode 17 is formed on almost the entire back surface 1b of the semiconductor wafer 1 by, for example, sputtering film formation (see FIG. 25 and its description for specific details). ). Thereafter, it is divided into chip regions of the semiconductor wafer 1 by dicing or the like, and sealed in a package as necessary, thereby completing the device.

4). Detailed description of back side device structure of narrow active cell IE type trench gate IGBT of one embodiment of the present application or description of modified example (aluminum doped structure) (mainly FIG. 25)
The examples described in this section relate to the back side structure of the semiconductor substrate, but all the examples other than this section relate to the front side structure of the semiconductor substrate. Thus, the examples in this section are applicable to all other examples except this section. Needless to say, the present invention can also be applied to IGBTs having other general surface-side structures.

  In this section, for convenience of explanation, the device structure will be described according to the example in Section 2, and the process will be briefly described with reference to Section 3.

  In the following, the IE type trench gate IGBT will be described in detail. However, the back surface structure is not limited to the IE type IGBT and the trench gate IGBT, and can be applied to other forms of IGBTs and the like. Needless to say.

  FIG. 25 is a detailed explanation of the back side device structure of the narrow active cell IE type trench gate IGBT of the one embodiment of the present application, or a device back side for explaining a device structure of a modified example (aluminum doped structure) and its manufacturing method. It is a local detailed sectional view. Based on this, a detailed description or a modified example (aluminum doped structure) on the back side device structure of the narrow active cell IE type trench gate IGBT of the one embodiment of the present application will be described.

FIG. 25 shows an enlarged cross-sectional view of the back side and its vicinity of the semiconductor chip 2 in FIG. As shown in FIG. 25, a relatively thin P-type semiconductor region (for example, a thickness of about 0.04 to 0.1 micrometers) is formed in the semiconductor region at the lower end of the P + type collector region 18 on the back surface side of the semiconductor substrate 2. That is, the aluminum doped region 30 is provided, and the impurity concentration (for example, about 1 × 10 ¹⁹ / cm ³ ) is higher than the impurity concentration of the P + type collector region 18. A metal collector electrode 17 is formed on the back surface 1 b of the semiconductor substrate 2 in contact with the aluminum doped region 30. For example, the metal collector electrode 17 has the following configuration from the side closer to the semiconductor substrate 2. That is, an aluminum back metal film 17a (for example, about 600 nm thick), a titanium back metal film 17b (for example, about 100 nm thick), and a nickel back metal film 17c (for example, 600 nm thick) that are impurity sources of the aluminum doped region 30. And a gold back metal film 17d (for example, a thickness of about 100 nm).

  Next, the production method will be briefly described. In the process of FIG. 24 in section 3, that is, during the sputtering film formation, the aluminum back surface metal film 17a, the titanium back surface metal film 17b, the nickel back surface metal film 17c, and the gold back surface metal film 17d are sequentially formed by sputtering. The heat generated at this time introduces aluminum into the silicon substrate, and the aluminum doped region 30 is formed. Then, when it is divided into chip regions of the semiconductor wafer 1 by dicing or the like, it becomes as shown in FIG. 7 (the detailed structure is not clearly shown in FIG. 7).

  In each of the embodiments of the present application, holes are accumulated on the emitter side in the on state to promote electron injection. On the other hand, the PN diode on the back collector side is a diode having low injection efficiency to reduce switching loss. That is, it is a transparent emitter. Here, in order to form a backside diode with low injection efficiency, the ratio of the carrier concentration “Qp” of the P + type collector region 18 to the carrier concentration “Qn” of the N type field stop region 19 (hereinafter referred to as “carrier concentration ratio”). ), That is, it is effective to reduce “(Qp / Qn)”. However, if the carrier concentration “Qp” of the P + type collector region 18 is lowered too much, the characteristics of the back surface metal contact deteriorate. Therefore, in this example, an aluminum doped region 30 having a higher impurity concentration than the P + type collector region 18 introduced from the aluminum film on the back surface is provided. As the carrier concentration ratio, for example, about 1.5 (for example, about 1.1 to 4 as a range) may be exemplified as a suitable one that can optimize the trade-off performance of switching performance-on-voltage performance. it can. Each embodiment of the present application has a structure for promoting the injection of electrons by accumulating holes on the emitter side in the on state. At this time, electrons injected from the front surface side reach the back collector side, and promote injection of holes from the back surface PN diode. Further, the injected holes reach the surface and promote the injection of electrons from the surface side. When each embodiment of the present application is used, conductivity modulation of the N-drift region 20 is likely to occur due to such a positive feedback phenomenon, so that the PN diode on the back collector side is made to have a specification with lower injection efficiency. Thus, it is possible to realize a device in which an increase in on-voltage is unlikely to occur. When importance is attached to the switching performance, it may be assumed that “(Qp / Qn)” is 1 or less, but even in that case, a rapid increase in the on-voltage can be suppressed by the effect of the surface structure of the present application.

5. Description of Modification 1 (N + type surface floating region & P + type surface floating region additional structure) related to the surface side device structure of the narrow active cell IE type trench gate IGBT of the one embodiment of the present application (mainly FIG. 26 to FIG. 28)
The example described in this section is, for example, a modification of the planar layout of FIG.

  FIG. 26 corresponds to Modification 1 (N + type surface floating region & P + type surface floating region additional structure for explaining the surface side device structure of the narrow active cell IE type trench gate IGBT of the one embodiment of the present application). Fig. 27 is a device sectional view corresponding to the section FF 'in Fig. 26. Fig. 28 is a device sectional view corresponding to the section GG' in Fig. 26. A modification 1 (N + type surface floating region & P + type surface floating region additional structure) relating to the surface side device structure of the narrow active cell IE type trench gate IGBT of the one embodiment of the present application will be described.

  As shown in FIG. 26, unlike FIG. 6, not only the linear active cell region 40a but also the portion of the linear inactive cell region 40i corresponding to the active section 40aa also has an N + type surface corresponding to the N + type emitter region 12. A floating region 12i (first conductivity type surface floating region) is provided. That is, the N + type surface floating region 12i is simultaneously formed by the same process as that of the N + type emitter region 12, for example. Thereby, the linear inactive cell region 40i has a first conductive type floating region forming section in which the N + type surface floating region 12i is formed and a first conductive type in which the N + type surface floating region 12i is not formed in the length direction. It will be divided into sections where the mold floating region is not formed.

  Part of the electrons injected from the MOSFET portion in the IGBT also reaches the N + type surface floating region 12i via the storage layer formed in the N-type layer portion of the trench sidewall and the inversion layer formed in the P-type sidewall portion. Then, it is injected into the P-type floating region 16. When the IGBT is turned off in this state, the electrons are recombined with holes remaining in the P-type floating region 16 and disappear. Thereby, the switching loss at the time of OFF can be reduced.

  Similarly, unlike FIG. 6, not only the linear active cell region 40a but also the portion of the linear inactive cell region 40i corresponding to the active section 40aa is not limited to the P + type body contact region 12 and the P + type embedded body contact. A region corresponding to the region 55 is provided. That is, the P + type surface floating region 25i (second conductivity type surface floating region) and the P + type embedded floating region 55i.

  Therefore, as shown in FIG. 27, the FF ′ cross section of FIG. 26 has an N + type surface floating region 12i provided on the surface 1a of the semiconductor substrate of the active section 40aa in the linear inactive cell region 40i. Except for this, it is almost the same as FIG.

  On the other hand, as shown in FIG. 28, the GG ′ cross section of FIG. 26 shows the P + type body contact region 12 and the P + type embedded body contact also on the surface region of the P type body region 15 of the linear inactive cell region 40i. 8 except that a P + type surface floating region 25i (second conductivity type surface floating region) and a P + type buried floating region 55i corresponding to the region 55 are provided.

  Such a structure has an advantage that the N + type emitter region introducing resist film 39 and the P + type body contact region introducing resist film 56 can have a relatively simple structure crossing the trench 21. . That is, there is an advantage of increasing the process margin in that it is not necessary to extend the end portions of these resist film patterns along the trench. Further, this case is not limited to completely traversing the linear inactive cell region 40i. The resist film pattern of the N + type emitter region introducing resist film 39 and the P + type body contact region introducing resist, etc. May be formed inside the linear inactive cell region 40i.

6). Description of Modification 2 (Simplified Active Cell Structure) Related to the Surface-side Device Structure of the Narrow Active Cell IE Trench Gate IGBT of the One Embodiment of the Present Application (Mainly FIG. 29)
The unit cell structure described in this section is obtained by omitting the P-type floating region 16 and the N-type hole barrier region 24 in the unit cell structure in FIG.

  29 is a cross-sectional view of FIG. 6 corresponding to FIG. 7 for explaining a second modification (simplified active cell structure) related to the surface side device structure of the narrow active cell IE type trench gate IGBT according to the embodiment of the present invention. It is device sectional drawing corresponding to a C 'cross section. Based on this, a modification 2 (simplified active cell structure) related to the surface side device structure of the narrow active cell IE type trench gate IGBT of the one embodiment of the present application will be described.

  As shown in FIG. 27, in this example, the P-type floating region 16 and the N-type hole barrier region 24 are omitted from the unit cell structure shown in FIG. Therefore, the hole concentration in the N − -type drift region 20 portion of the linear active cell region 40a tends to decrease as compared with the structure of FIG. For example, when the linear active cell region 40a is sufficiently narrow, and when the trench 21 is sufficiently deep, it is effective to adopt the structure of FIG. That is, the device structure and the impurity doping process can be simplified. In addition, this structure is advantageous not for applications that place importance on low on-voltage but for applications that require higher switching performance.

7). Description of Modification 3 (Hole Collector Cell Additional Structure) on the Surface-Side Device Structure of the Narrow Active Cell IE Trench Gate IGBT of the One Embodiment of the Present Application (Mainly FIGS. 30 to 35)
The example described in this section is a variation on the basic device structure described in Section 1 (mainly FIG. 2). Therefore, the other figures, such as FIGS. 1, 3 to 29 etc., also apply to this example, with corresponding changes or as they are.

  FIG. 30 is a cell region of FIG. 1 corresponding to FIG. 2 for explaining a third modification (hole collector cell additional structure) relating to the surface side device structure of the narrow active cell IE type trench gate IGBT of the one embodiment of the present application. It is a device schematic cross section of the AA 'cross section of edge part cutting-out area | region R1. FIG. 31 is a linear unit cell region of FIG. 1 for explaining a third modification (hole collector cell additional structure) relating to the surface side device structure of the narrow active cell IE type trench gate IGBT of the one embodiment of the present application and its It is an enlarged top view of periphery R5. FIG. 32 is an enlarged top view corresponding to FIG. 6 for explaining a modification 3 (hole collector cell additional structure) related to the surface side device structure of the narrow active cell IE type trench gate IGBT of the one embodiment of the present application. . 33 is a device cross-sectional view corresponding to the H-H ′ cross section of FIG. 32. FIG. 34 is a device cross-sectional view corresponding to the J-J ′ cross section of FIG. 32. 35 is a device sectional view corresponding to a section taken along line K-K ′ of FIG. 32. Based on these, a third modification (hole collector cell additional structure) related to the surface side device structure of the narrow active cell IE type trench gate IGBT of the one embodiment of the present application will be described.

(1) Description of narrow active cell type unit cells and alternate arrangement method (mainly FIG. 30):
Next, FIG. 30 shows an XX ′ cross section of the cell region end cutout region R1 of FIG. As shown in FIG. 30, a P + type collector region 18 is provided in a semiconductor region (in this example, a silicon single crystal region) on the back surface 1b of the chip 2 (the back main surface or the second main surface of the semiconductor substrate). The metal collector electrode 17 is provided on the surface. An N-type field stop region 19 is provided between the N − -type drift region 20 (first conductivity type drift region) and the P + -type collector region 18 constituting the main part of the semiconductor substrate 2.

  On the other hand, a large number of trenches 21 are provided in the semiconductor region on the surface side 1a (the front-side main surface or the first main surface of the semiconductor substrate) of the N − type drift region 20, and a gate insulating film is included therein. A trench gate electrode 14 is embedded via 22. These trench gate electrodes 14 are connected to the metal gate electrode 5 (specifically, the metal gate wiring 7) or the emitter electrode 8 according to its function.

  These trenches 21 function to partition each region. For example, the dummy cell region 34 is partitioned from both sides by a pair of trenches 21, and a cell forming region is formed by one of the trenches 21. 10 and the cell peripheral junction region 35 are partitioned. The cell peripheral junction region 35 is connected to the metal emitter electrode 8 through the P + type body contact region 25p. In the present application, unless otherwise specified, the thickness of the gate insulating film 22 in any part of the trench is substantially the same (however, if necessary, the thickness of a part is different from that of the other part and is different. Is not to eliminate). As described above, by making emitter contact in the cell peripheral junction region 35 and the dummy cell region 34, it is possible to prevent the breakdown voltage from being lowered even when the width of the dummy cell region 34 or the like is changed in a process. it can.

  A P-type floating field ring 36 is provided in the semiconductor region on the surface side 1a of the N − type drift region 20 outside the cell peripheral junction region 35, and the field plate 4 is provided on the surface 1a. The P + type body contact region 25r is connected to the floating field ring 36.

  Next, the cell formation region 10 will be further described. The dummy cell region 34 is basically the same in structure and size as the linear active cell region 40a except that it does not have the N + type emitter region 12, and is a P + type body provided on the surface of the P type body region 15. Contact region 25 d is connected to metal emitter electrode 8. The dummy cell region 34 can basically have the same structure as the hole collector cell (see FIG. 30).

  Most of the inner region of the cell formation region 10 basically has a repetitive structure to be translated with the linear unit cell region 40 as a unit lattice (note that the objectivity in a strict sense is not required. The same). The linear unit cell region 40 as a unit lattice includes a linear inactive cell region 40i, a linear active cell region 40a on one side thereof, a linear hole collector cell region 40c on the other side, and both sides thereof. The half-width linear inactive cell region 40i. However, specifically, it can be seen that the linear active cell regions 40a and the linear hole collector cell regions 40c are alternately arranged between the full-width linear inactive cell regions 40i (FIG. 31). reference). It can also be seen that the first linear unit cell regions 40f and the second linear unit cell regions 40s are alternately arranged.

  A P-type body region 15 (second conductivity type body region) is provided in a semiconductor surface region on the front-side main surface 1a (first main surface) side of the semiconductor substrate of the linear active cell region 40a. Are provided with an N + type emitter region 12 (first conductivity type emitter region) and a P + type body contact region 25. N + type emitter region 12 and P + type body contact region 25 are connected to metal emitter electrode 8. In the linear active cell region 40 a, an N-type hole barrier region 24 is provided in the N − -type drift region 20 below the P-type body region 15. Note that the trench gate electrodes 14 on both sides of the linear active cell region 40 a are electrically connected to the metal gate electrode 5.

  On the other hand, the structure of the linear hole collector cell region 40c is different in this example only in that the N + type emitter region 12 is not present and the trench gate electrodes 14 on both sides are connected to the emitter electrode 8. The other points are the same as those of the linear active cell region 40a including dimensions and the like.

  On the other hand, a P-type body region 15 is similarly provided in the semiconductor surface region on the front-side main surface 1a (first main surface) side of the semiconductor substrate in the linear inactive cell region 40i, and the N− The type drift region 20 is provided with a P-type floating region 16 (second conductivity type floating region) that covers the lower ends of the trenches 21 on both sides and is deeper than that. By providing such a P-type floating region 16, the width Wi of the linear inactive cell region can be increased without causing a sharp drop in breakdown voltage. For example, even if the layout is adjusted for optimizing the characteristics such as the gate capacitance and the on-voltage, there is no concern about a decrease in breakdown voltage, and a design freedom can be secured. For example, even if the concentration of the N-type hole barrier region 24 is increased for optimization or the like, there is no influence on the breakdown voltage. This makes it possible to effectively enhance the hole accumulation effect. In the IE trench gate IGBT, the contact from the emitter electrode 8 to the P type floating region 16 is not formed. This is because the hole concentration in the N − type drift region 20 (N base region) below the linear active cell region 40a is reduced by blocking the direct hole discharge path from the P type floating region 16 to the emitter electrode 8. It is increasing. As a result, the on-resistance is reduced by improving the concentration of electrons injected from the MOSFET in the IGBT into the N base region.

  In this example, the width Wa of the linear active cell region 40a and the width Wc of the linear hole collector cell region 40c are narrower than the width Wi of the linear inactive cell region 40i. It is called an “active cell type unit cell”. Hereinafter, the device having the narrow active cell type unit cell will be specifically described below. However, the example described here is not limited thereto, and the “non-narrow active cell type unit cell” is referred to as “non-narrow active cell type unit cell”. Needless to say, the present invention can be applied to a device having the same.

  In the example of FIG. 30, the linear unit cell region 40 is configured by alternately arranging the linear active cell region 40a (or the linear Hall collector cell region 40c) and the linear inactive cell region 40i. This configuration is referred to as an “alternate arrangement method” in the present application. In the following description, the description will be made on the premise of the alternating arrangement method unless otherwise specified. Needless to say, the “non-alternating arrangement method” may be used.

  In FIG. 30, the outline (main part and peripheral part) of the embodiment of FIGS. 31 to 35 of the present application has been described. However, in the following description, the cell part (cross section, planar structure), the cell peripheral part, etc. Although the description will be divided into components, it goes without saying that these also give outlines to various modifications.

(2) Description of active cell two-dimensional thinning structure (mainly FIG. 31):
An example of the detailed planar structure of the main portion of the linear unit cell region and the peripheral cutout region R5 in FIG. 1 is shown in FIG. As shown in FIG. 31, in the length direction of the linear active cell region 40a, for example, active sections 40aa having a constant length are provided at regular intervals, and an N + type emitter region 12 is provided therebetween. There is no inactive section 40ai. That is, a part of the linear active cell region 40a in the length direction is an active section 40aa in a locally dispersed manner. More specifically, in the active section 40aa of the linear active cell region 40a, the N + type emitter region 12 is provided in almost the entire region, and in the inactive section 40ai of the linear active cell region 40a, in the almost entire region, P + type body contact region 25 and P + type buried body contact region 55 are provided. On the other hand, in the linear hole collector cell region 40c, the P + type body contact region 25 and the P + type buried body contact region 55 are provided almost all over, and the N + type emitter region 12 is not provided. In the linear inactive cell region 40i, the P-type body region 15 and the P-type floating region 16 (second conductivity type floating region) are provided in almost the entire region as in the other cases.

  Here, being distributed at a constant interval and having a constant length means periodic, but being substantially periodic corresponds to a local dispersive distribution, Being distributed is broader and does not necessarily mean periodic or quasi-periodic.

(3) Detailed description of layout and device structure in which every other active cell is replaced with a hole collector cell (mainly FIGS. 32 to 35):
The structures of the linear active cell region 40a and the linear inactive cell region 40i are the same as those shown in FIGS. 4 and 6 to 9, and only the linear hole collector cell region 40c will be described below.

  As shown in FIG. 32, the trench buried electrodes 14c on both sides of the linear hole collector cell region 40c need to be connected to the emitter potential. In this example, two (trench embedded electrodes 14c on both sides) are connected to each other by, for example, a buried electrode connecting portion 28 (including the in-trench electrode 14i in the connecting portion trench 21c) made of a polysilicon film of the same layer, An emitter electrode-buried electrode contact portion 11c (contact hole) is provided on the buried electrode connecting portion 28, and is connected to the metal emitter electrode 8 through this. The characteristics of the linear hole collector cell region 40 c are similar to those of the linear active cell region 40 a, but the N + -type emitter region 12 is not provided, and is almost the same except for the lower part of the buried electrode connecting portion 28. The difference is that a P + type body contact region 25 and a P + type embedded body contact region 55 are provided over the entire area.

  Next, FIG. 33 shows an H-H ′ cross section of FIG. 32. As shown in FIG. 33, it is similar to the cross section of FIG. 7 (linear active cell region 40a), but in this cross section of the linear hole collector cell region 40c, the buried electrode connection portion 28 (polysilicon connection) is located upward. Therefore, the N + type emitter region 12 and the P type body region 15 are not introduced. This is because the polysilicon film 27 on the linear hole collector cell region 40c is left in the step of FIG. 17, and as a result, impurities are introduced below the buried electrode connecting portion 28 in the step of FIG. Because it is not done. The same applies to FIG. Here, below the buried electrode connecting portion 28 is in a floating state. However, if there is no P-type diffusion layer, the electric field strength concentrates at the lower end of the trench in the off state, and the breakdown voltage decreases. Therefore, it is desirable to introduce the P-type floating region 16 below the buried electrode connecting portion 28. Since the P-type floating region 16 is ion-implanted in a step prior to forming the buried electrode, it can be arranged below the buried electrode connecting portion 28. As a result, no matter what size the linear inactive cell region 40i is set to, the withstand voltage can be ensured, and a design with a degree of freedom can be made according to product requirements.

  Next, FIG. 34 shows a J-J ′ cross section of FIG. 32. As shown in FIG. 34, it is basically similar to FIG. 9, but there is a connecting portion trench 21c and a trench buried connecting portion 14i between the linear hole collector cell region 40c and the linear inactive cell region 40i. It is different in point. Another difference is that the linear hole collector cell region 40c does not have the N + type emitter region 12, the P + type body contact region 25, and the P + type buried body contact region 55 for the same reason as described above. Further, the P-type floating region 16 and the P + type body contact region 25 and the P + type buried body contact region 55 connected to the metal emitter electrode 8 are efficiently separated by the connecting portion trench 21c.

  Next, FIG. 35 shows a K-K ′ cross section of FIG. 32. As shown in FIG. 35, it is exactly the same as FIG. 8 except that the trench gate electrode 14 is a trench buried electrode 14c connected to the emitter potential.

  The degree of hole accumulation effect of the surface device structure depends on the geometric shape and the N-type hole barrier region 24. That is, it can be said that there is no significant difference between the linear active cell region 40a and the linear hole collector cell region 40c for holes. That is, the hole accumulation effect is equivalent, and the IE effect is correspondingly equivalent. Thus, by replacing a part of the plurality of linear active cell regions 40a with the linear hole collector cell region 40c, the number of trenches functioning as gate capacitance can be reduced while the hole accumulation effect is the same. That is, it is possible to prevent the gate capacity from increasing even if cell shrinking is performed to the maximum in order to enhance the IE effect.

  Further, when a part of the linear active cell region 40a is replaced with the linear Hall collector cell region 40c, the absolute amount of the N + type emitter region 12 per unit area is reduced and the value of the saturation current is reduced in that state. Become. However, if the length of the inactive section 40ai in the linear active cell region 40a is shortened and the proportion occupied by the active section 40aa is increased and optimized, the saturation current value required as an IGBT for inverter use can be maintained. Can do. In the main embodiment of the present application, a linear layout is basically adopted, and the optimum design of the N + type emitter region 12 is easy. Further, since significant cell shrinkage is realized with respect to the IGBT using the conventional technology, even if the linear hole collector cell region 40c is arranged, the layout in the linear active cell region 40a is optimized. The saturation current required for the entire chip can be ensured.

8). Supplementary explanation about the above-described embodiment (including modifications) and general consideration (mainly FIG. 36)
FIG. 36 is an enlarged top view of the linear unit cell region of FIG. 1 and its periphery R5 for explaining the outline of the device structure of the one embodiment of the present application. Based on this, a supplementary explanation regarding the above-described embodiment (including modifications) and a general consideration will be given.

  In the IE type IGBT, in order to increase the IE effect, it is necessary to shorten the interval between the trenches as much as possible. However, in a structure having a substrate contact substrate groove (or substrate contact groove), it is necessary to secure the width of the substrate contact groove, and shrinking is difficult. On the other hand, even if the width of the trench itself is reduced, the IE effect is not improved. Rather, in order to ensure the thickness of the gate insulating film or the like, it is better not to reduce the width of the trench itself.

  Therefore, in each of the above-described embodiments (including modifications), as shown in FIG. 36, in the two-dimensional thinning structure, an N + type emitter region is formed on almost the entire active section 40aa of the linear active cell region 40a. 12 was spread and the structure without the P + type body contact region 25 was formed.

  Here, from the viewpoint of ensuring latch-up resistance, the width of the active section 40aa is preferably as small as possible (for example, about 0.5 micrometers or less). When it is desired to increase the overall area of the active section 40aa, it is preferable to decrease the pitch (increase the number) while maintaining the width of each active section 40aa as it is. If the width of each individual active section 40aa is increased too much, the latch-up resistance is lowered, and the short-circuit safety operating area is also reduced.

9. Summary The invention made by the present inventor has been specifically described based on the embodiments. However, the present invention is not limited thereto, and it goes without saying that various changes can be made without departing from the scope of the invention.

  For example, in each of the above-described embodiments, an example in which doped polysilicon (Doped Poly-silicon) or the like is used as the gate polysilicon member has been specifically described. However, the present invention is not limited thereto, and is not doped. A polysilicon (Nondoped Poly-silicon) film may be applied, and necessary impurities may be added by ion implantation or the like after the film formation.

  Furthermore, in the above-described embodiment, an example in which a non-epitaxial wafer is used and a high concentration impurity layer is formed from the back surface after back grinding has been described. However, the invention of the present application is not limited thereto, and the epitaxial wafer Needless to say, the present invention can also be applied to those manufactured using the above.

1 Semiconductor wafer 1a Wafer or chip surface (first main surface)
1b Back surface of wafer or chip (second main surface)
1s N-type single crystal silicon substrate 2 Semiconductor chip (semiconductor substrate)
3 Guard ring 4 Field plate 5 Metal gate electrode 6 Gate pad 7 Metal gate wiring 8 Metal emitter electrode 9 Metal emitter pad 10 IGBT cell region (or cell region)
11 Contact groove (or contact hole)
11c Contact between emitter electrode and buried electrode 12 N + type emitter region 12i N + type surface floating region (first conductivity type surface floating region)
14 Trench gate electrode (Main trench gate electrode)
14c Trench buried electrode connected to emitter potential 14i Trench buried connecting portion 15 P type body region 16 P type floating region (second conductivity type floating region)
17 Metal collector electrode 17a Aluminum back surface metal film 17b Titanium back surface metal film 17c Nickel back surface metal film 17d Gold back surface metal film 18 P + type collector region 19 N type field stop region 20 N− type drift region 21 Trench 21c Connecting portion trench 22 Gate insulation Film 24 N-type hole barrier region 25 P + type body contact region 25d P + type body contact region of dummy cell 25i P + type surface floating region (second conductivity type surface floating region)
25p P + type body contact region of cell peripheral junction region 25r P + type body contact region of floating field ring 26 Interlayer insulating film 27 Polysilicon film 28 Embedded electrode connecting portion 30 Aluminum doped region 31 Resist film for introducing N type hole barrier region 32 Trench Hard mask film 33 Resist film for trench hard mask film processing 34 Dummy cell region (linear dummy cell region)
35 Cell peripheral junction region (second conductivity type cell peripheral junction region)
36 Floating field ring (field limiting ring)
37 P-type floating region introduction resist film 38 P-type body region introduction resist film 39 N + type emitter region introduction resist film 40 Linear unit cell region 40a Linear active cell region 40aa Active section 40ai Inactive section 40c Linear hole Collector cell region 40f First linear unit cell region 40i Linear inactive cell region 40is Linear inactive subcell region 40s Second linear unit cell region 55P + type embedded body contact region 55i P + type embedded floating region 56P + type body Contact region introduction resist film R1 Cell region edge cutout region R2, R3 Cell region internal cutout region R5 Linear unit cell region main portion and surrounding cutout region Wa Linear active cell region Width Wc Width of linear hole collector cell region Wf Width of first linear unit cell region Wi Width of linear inactive cell region Wis Width of linear inactive subcell region

Claims (6)

(A) a semiconductor substrate having a first main surface and a second main surface and having a drift region of the first conductivity type;
(B) an IGBT cell region provided on the first main surface side of the semiconductor substrate;
(C) A linear active cell region formed in the IGBT cell region and extending in a first direction in plan view, and adjacent to both sides of the linear active cell region in a second direction intersecting the first direction And a linear inactive cell region extending in the first direction;
(D) a linear hole collector cell region disposed adjacent to the outside of the linear inactive cell region in the second direction and extending in the first direction;
(E) The first main surface of the semiconductor substrate, which is a boundary between the linear active cell region and the linear inactive cell region, and the linear inactive cell region and the linear Hall collector cell. A trench formed at the boundary of the region;
(F) In the trench formed at the boundary between the linear active cell region and the linear inactive cell region, a gate electrode formed on the trench surface via an insulating film;
(G) first conductivity type emitter regions and second conductivity type body contact regions formed in the linear active cell region and alternately arranged in the first direction in plan view;
(H) a metal emitter electrode provided on the first main surface of the semiconductor substrate and electrically connected to the emitter region, the body contact region, and the linear hole collector cell region;
(I) a floating region of a second conductivity type formed in the linear inactive cell region and having a junction interface with the drift region extending to a position deeper than a lower end of the trench;
The IE type IGBT, wherein the width of the linear inactive cell region in the second direction is wider than the width of each of the linear active cell region and the linear hole collector cell region in the second direction. In IE type IGBT of Claim 1,
(J) A surface region on the first main surface side of the semiconductor substrate, and substantially deeper than the lower ends of the trenches at both ends of the linear active cell region and the linear hole collector cell region. It further has a hole barrier region of the first conductivity type . In IE type IGBT of Claim 1,
(K) It further has an embedded body contact region of the second conductivity type provided on and substantially in contact with the lower layer of the body contact region. In IE type IGBT of Claim 1,
(L) The semiconductor substrate further includes a surface floating region of a second conductivity type which is a surface region on the first main surface side of the semiconductor substrate and is provided in the linear inactive cell region. In IE type IGBT of Claim 1,
The width of the linear inactive cell region in the second direction is two to three times the width of the linear active cell region and the linear hole collector cell region in the second direction. In IE type IGBT of Claim 2 ,
The lower end of the hole barrier region of the first conductivity type is deeper than the lower end of the trench,
The lower end of the second conductivity type floating region is deeper than the lower end of the first conductivity type hole barrier region.

Images