Semiconductor memory

The semiconductor memory device addresses bonding defects by using longer bonding pads and controlled electrode formation in high-voltage regions, improving reliability and reducing disconnections through enhanced contact areas and stress management.

JP2024160518AInactive Publication Date: 2024-11-14KIOXIA CORP
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
JP2023075610
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-01
Publication Date
2024-11-14
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing semiconductor memory devices face issues with bonding defects, particularly in regions with high voltage applications, leading to potential disconnections and voids due to uneven electrode surfaces and thermal stress during bonding processes.

Method used

The semiconductor memory device design includes longer bonding pads and contact plugs in high-voltage regions to increase the effective contact area and reduce thermal stress, using a dual damascene method to form electrodes with controlled thickness and aspect ratios, thereby enhancing bonding reliability.

Benefits of technology

This configuration reduces bonding defects by increasing the contact area and minimizing disconnection failures, while maintaining manufacturing efficiency and cost-effectiveness.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024160518000001_ABST
    Figure 2024160518000001_ABST
Patent Text Reader

Abstract

To reduce lamination defects.SOLUTION: According to the embodiment, a semiconductor memory has a first chip 10 including a plurality of first wiring layers WL spaced and stacked in the first direction, a memory pillar MP extending in the first direction and passing through the plurality of first wiring layers, a second wiring layer BL electrically connected to the memory pillar, a first electrode 108 electrically connected to one of the plurality of first wiring layers and a second electrode 108 electrically connected to the second wiring layer, and a second chip 20 including a third electrode 210 bonded to the first electrode, and a fourth electrode 210 bonded to the second electrode. The length Hmw2 of the first electrode in the first direction is greater than the length Hmc2 of the second electrode in the first direction.SELECTED DRAWING: Figure 8
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to a semiconductor memory device. [Background technology]

[0002] A NAND flash memory is known as one type of semiconductor memory device. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2021-136271 A [Patent Document 2] JP 2022-95359 A Summary of the Invention [Problem to be solved by the invention]

[0004] An embodiment of the present invention provides a semiconductor memory device capable of reducing bonding defects. [Means for solving the problem]

[0005] A semiconductor memory device according to an embodiment includes a first chip including a plurality of first wiring layers stacked at a distance in a first direction, a memory pillar extending in the first direction and passing through the plurality of first wiring layers, a second wiring layer electrically connected to the memory pillar, a first electrode electrically connected to any one of the plurality of first wiring layers, and a second electrode electrically connected to the second wiring layer, and a second chip including a third electrode bonded to the first electrode, and a fourth electrode bonded to the second electrode. The length of the first electrode in the first direction is greater than the length of the second electrode in the first direction. [Brief description of the drawings]

[0006] [Figure 1] 1 is a block diagram showing an overall configuration of a semiconductor memory device according to a first embodiment. [Diagram 2]1 is a circuit diagram of a memory cell array included in a semiconductor memory device according to a first embodiment. [Diagram 3] 1 is a perspective view showing an outline of a bonding structure of a semiconductor memory device according to a first embodiment. [Figure 4] 1 is a plan view of a memory cell array included in a semiconductor memory device according to a first embodiment. [Diagram 5] An enlarged view of region ER in Figure 4 . [Figure 6] 6 is a cross-sectional view of the semiconductor memory device taken along line IIX-IIX in FIG. 5. [Figure 7] 6 is a cross-sectional view of the semiconductor memory device taken along line IX-IX in FIG. 5. [Figure 8] FIG. 2 is a cross-sectional view of a bonding pad included in the semiconductor memory device according to the first embodiment. [Figure 9] 1 is a cross-sectional view of a memory cell array included in the semiconductor memory device according to the first embodiment. [Figure 10] FIG. 10 is a cross-sectional view taken along line XX in FIG. [Figure 11] 3A to 3C are cross-sectional views showing a manufacturing process of the array chip included in the semiconductor memory device according to the first embodiment. [Figure 12] 3A to 3C are cross-sectional views showing a manufacturing process of the array chip included in the semiconductor memory device according to the first embodiment. [Figure 13] 3A to 3C are cross-sectional views showing a manufacturing process of the array chip included in the semiconductor memory device according to the first embodiment. [Figure 14] 3A to 3C are cross-sectional views showing a manufacturing process of the array chip included in the semiconductor memory device according to the first embodiment. [Figure 15] 3A to 3C are cross-sectional views showing a manufacturing process of the array chip included in the semiconductor memory device according to the first embodiment. [Figure 16] 3A to 3C are cross-sectional views showing a manufacturing process of the array chip included in the semiconductor memory device according to the first embodiment. [Figure 17] 3A to 3C are cross-sectional views showing a manufacturing process of the array chip included in the semiconductor memory device according to the first embodiment. [Figure 18] 3A to 3C are cross-sectional views showing a manufacturing process of the array chip included in the semiconductor memory device according to the first embodiment. [Figure 19] 3A to 3C are cross-sectional views showing a manufacturing process of the array chip included in the semiconductor memory device according to the first embodiment. [Figure 20] 3A to 3C are cross-sectional views showing a manufacturing process of the array chip included in the semiconductor memory device according to the first embodiment. [Figure 21] FIG. 11 is a cross-sectional view of a semiconductor memory device according to a second embodiment. [Figure 22] 10A to 10C are cross-sectional views showing a manufacturing process of an array chip included in a semiconductor memory device according to a second embodiment. [Figure 23] 10A to 10C are cross-sectional views showing a manufacturing process of an array chip included in a semiconductor memory device according to a second embodiment. [Figure 24] 10A to 10C are cross-sectional views showing a manufacturing process of an array chip included in a semiconductor memory device according to a second embodiment. [Diagram 25] 10A to 10C are cross-sectional views showing a manufacturing process of an array chip included in a semiconductor memory device according to a second embodiment. [Figure 26] FIG. 11 is a cross-sectional view of a semiconductor memory device according to a third embodiment. [Figure 27] FIG. 13 is a cross-sectional view of a bonding pad included in the semiconductor memory device according to the third embodiment. [Figure 28] 11A to 11C are cross-sectional views showing a manufacturing process of an array chip included in a semiconductor memory device according to a third embodiment. [Figure 29] 11A to 11C are cross-sectional views showing a manufacturing process of an array chip included in a semiconductor memory device according to a third embodiment. [Diagram 30] 11A to 11C are cross-sectional views showing a manufacturing process of an array chip included in a semiconductor memory device according to a third embodiment. [Diagram 31] 11A to 11C are cross-sectional views showing a manufacturing process of an array chip included in a semiconductor memory device according to a third embodiment. [Diagram 32] 11A to 11C are cross-sectional views showing a manufacturing process of an array chip included in a semiconductor memory device according to a third embodiment. [Diagram 33] 11A to 11C are cross-sectional views showing a manufacturing process of an array chip included in a semiconductor memory device according to a third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0007] Hereinafter, the embodiments will be described with reference to the drawings. In the following description, components having substantially the same functions and configurations are given the same reference numerals. Duplicate description may be omitted if unnecessary. Each embodiment shown below is an example of an apparatus or method for embodying the technical idea of ​​the embodiment. The technical idea of ​​the embodiment does not specify the material, shape, structure, arrangement, etc. of the components as described below. The technical idea of ​​the embodiment can be modified in various ways without departing from the gist of the invention. These embodiments and their modifications are included in the scope of the invention described in the claims and their equivalents.

[0008] 1. First embodiment A semiconductor memory device according to a first embodiment will be described below, taking as an example a three-dimensional stacked NAND flash memory in which memory cell transistors are stacked three-dimensionally on a semiconductor substrate.

[0009] 1.1 Configuration 1.1.1 Overall configuration of semiconductor memory device First, an example of the overall configuration of a semiconductor memory device 1 will be described with reference to Fig. 1. Fig. 1 is a block diagram showing the overall configuration of the semiconductor memory device 1. Note that, although some of the connections between the components are indicated by arrows in Fig. 1, the connections between the components are not limited to these.

[0010] The semiconductor memory device 1 is, for example, a three-dimensional stacked NAND flash memory. The three-dimensional stacked NAND flash memory includes a plurality of non-volatile memory cell transistors arranged three-dimensionally on a semiconductor substrate.

[0011] As shown in FIG. 1, a semiconductor memory device 1 includes an array chip 10 and a circuit chip 20.

[0012] The array chip 10 is a chip provided with a memory cell array including non-volatile memory cell transistors. The circuit chip 20 is a chip provided with a circuit for controlling the array chip 10. The semiconductor memory device 1 of this embodiment has a structure in which the array chip 10 and the circuit chip 20 are bonded together (hereinafter, referred to as a "bonded structure"). Hereinafter, when there is no limitation on either the array chip 10 or the circuit chip 20, it will be simply referred to as a "chip". Note that multiple array chips 10 may be provided. In this case, multiple array chips 10 may be bonded together so as to be stacked on the circuit chip 20.

[0013] The array chip 10 includes one or more memory cell arrays 11. The memory cell array 11 is a region in which non-volatile memory cell transistors (hereinafter also referred to as "memory cells") are arranged three-dimensionally. In the example of FIG. 1, the array chip 10 includes one memory cell array 11.

[0014] The circuit chip 20 includes a sequencer 21, a voltage generating circuit 22, a row decoder 23, and a sense amplifier 24.

[0015] The sequencer 21 is a control circuit for the semiconductor memory device 1. For example, the sequencer 21 is connected to a voltage generating circuit 22, a row decoder 23, and a sense amplifier 24. The sequencer 21 controls the voltage generating circuit 22, the row decoder 23, and the sense amplifier 24. The sequencer 21 also controls the overall operation of the semiconductor memory device 1 under the control of an external controller. More specifically, the sequencer 21 executes a write operation, a read operation, an erase operation, and the like.

[0016] The voltage generation circuit 22 is a circuit that generates voltages used in a write operation, a read operation, an erase operation, etc. For example, the voltage generation circuit 22 is connected to a row decoder 23 and a sense amplifier 24. The voltage generation circuit 22 supplies the generated voltages to the row decoder 23, the sense amplifier 24, etc.

[0017] The row decoder 23 is a circuit that decodes a row address. The row address is an address signal that specifies the row direction wiring of the memory cell array 11. The row decoder 23 supplies the voltage applied from the voltage generating circuit 22 to the memory cell array 11 based on the result of decoding the row address.

[0018] The sense amplifier 24 is a circuit for writing and reading data. During a read operation, the sense amplifier 24 senses data read from the memory cell array 11. During a write operation, the sense amplifier 24 supplies a voltage according to the write data to the memory cell array 11.

[0019] Next, the internal configuration of the memory cell array 11 will be described. The memory cell array 11 has a plurality of blocks BLK. The block BLK is, for example, a collection of a plurality of memory cell transistors from which data is erased collectively. The plurality of memory cell transistors in the block BLK are associated with rows and columns. In the example of FIG. 1, the memory cell array 11 includes blocks BLK0, BLK1, BLK2, and BLK3.

[0020] Each block BLK includes a plurality of string units SU. The string unit SU is a set of a plurality of NAND strings that are collectively selected in, for example, a write operation or a read operation. The NAND string includes a set of a plurality of memory cell transistors connected in series. In the example of FIG. 1, each block BLK includes five string units SU0 to SU4. The number of blocks BLK in the memory cell array 11 and the number of string units SU in each block BLK are arbitrary.

[0021] 1.1.2 Memory cell array circuit configuration Next, an example of a circuit configuration of the memory cell array 11 will be described with reference to Fig. 2. Fig. 2 is a circuit diagram of the memory cell array 11. Note that the example of Fig. 2 shows the circuit configuration of one block BLK.

[0022] As shown in FIG. 2, the string unit SU includes a plurality of NAND strings NS.

[0023] The NAND string NS includes a plurality of memory cell transistors MC and select transistors ST1 and ST2. In the example of Fig. 2, the NAND string NS includes eight memory cell transistors MC0 to MC7. The number of memory cell transistors MC included in the NAND string NS is arbitrary.

[0024] The memory cell transistor MC is a memory element that stores data in a non-volatile manner. The memory cell transistor MC includes a control gate and a charge storage film. The memory cell transistor MC may be of a MONOS (Metal-Oxide-Nitride-Oxide-Silicon) type or a FG (Floating Gate) type. The MONOS type uses an insulating layer for the charge storage film. The FG type uses a conductor for the charge storage film. The following describes the case where the memory cell transistor MC is of the MONOS type.

[0025] The selection transistors ST1 and ST2 are switching elements. The selection transistors ST1 and ST2 are each used to select a string unit SU during various operations. The number of selection transistors ST1 and ST2 included in the NAND string NS is arbitrary. It is sufficient that the NAND string NS includes one or more selection transistors ST1 and ST2.

[0026] The current paths of the select transistor ST2, the memory cell transistors MC0 to MC7, and the select transistor ST1 in the NAND string NS are connected in series. The drain of the select transistor ST1 is connected to a bit line BL. The source of the select transistor ST2 is connected to a source line SL.

[0027] The control gates of the memory cell transistors MC0 to MC7 in the same block BLK are commonly connected to word lines WL0 to WL7, respectively. More specifically, for example, the block BLK includes five string units SU0 to SU4. Each string unit SU includes a plurality of memory cell transistors MC0. The control gates of the plurality of memory cell transistors MC0 in the block BLK are commonly connected to one word line WL0. The same is true for the memory cell transistors MC1 to MC7.

[0028] The gates of the multiple selection transistors ST1 in the string unit SU are commonly connected to one selection gate line SGD. More specifically, the gates of the multiple selection transistors ST1 in the string unit SU0 are commonly connected to a selection gate line SGD0. The gates of the multiple selection transistors ST1 in the string unit SU1 are commonly connected to a selection gate line SGD1. The gates of the multiple selection transistors ST1 in the string unit SU2 are commonly connected to a selection gate line SGD2. The gates of the multiple selection transistors ST1 in the string unit SU3 are commonly connected to a selection gate line SGD3. The gates of the multiple selection transistors ST1 in the string unit SU4 are commonly connected to a selection gate line SGD4.

[0029] The gates of the select transistors ST2 in the block BLK are commonly connected to a select gate line SGS. Similar to the select gate line SGD, a different select gate line SGS may be provided for each string unit SU.

[0030] The word lines WL0 to WL7, the select gate lines SGD0 to SGD4, and the select gate line SGS are each connected to a row decoder 23.

[0031] The bit line BL is commonly connected to one NAND string NS in each string unit SU of each block BLK. The same column address is assigned to the multiple NAND strings NS connected to one bit line BL. Each bit line BL is connected to a sense amplifier 24.

[0032] The source line SL is shared among, for example, a plurality of blocks BLK.

[0033] A set of multiple memory cell transistors MC connected to a common word line WL in one string unit SU is expressed as, for example, a “cell unit CU.” For example, a write operation and a read operation are performed in units of cell units CU.

[0034] 1.1.3 Bonding structure of semiconductor memory device Next, an outline of the bonding structure of the semiconductor memory device 1 will be described with reference to Fig. 3. Fig. 3 is a perspective view showing an outline of the bonding structure of the semiconductor memory device 1.

[0035] As shown in Fig. 3, the array chip 10 includes a plurality of bonding pads MB provided on the surface facing the circuit chip 20. The circuit chip 20 also includes a plurality of bonding pads DB provided on the surface facing the array chip 10 so as to face the bonding pads MB. In the bonding structure, the bonding pads MB of the array chip 10 and the bonding pads DB of the circuit chip 20 are bonded together to form one bonding pad BP. In other words, the bonding pad BP is formed by bonding together an electrode (conductor) constituting the bonding pad MB provided on the array chip 10 and an electrode (conductor) constituting the bonding pad DB provided on the circuit chip 20.

[0036] Hereinafter, the surface where the array chip 10 and the circuit chip 20 are bonded together (hereinafter referred to as the "bonding surface") is referred to as the XY plane. The directions perpendicular to each other on the XY plane are referred to as the X direction and the Y direction. The direction approximately perpendicular to the XY plane, from the array chip 10 to the circuit chip 20, is referred to as the Z1 direction. The direction approximately perpendicular to the XY plane, from the circuit chip 20 to the array chip 10, is referred to as the Z2 direction. When there is no restriction on either the Z1 direction or the Z2 direction, it is referred to as the Z direction.

[0037] 1.1.4 Planar layout of memory cell array Next, an example of a planar layout of the memory cell array 11 will be described with reference to Figures 4 and 5. Figure 4 is a plan view showing an example of a planar layout of the memory cell array 11. Figure 5 is an enlarged view of an area ER in Figure 4. The example in Figure 4 shows areas corresponding to four blocks BLK0 to BLK3 included in the memory cell array 11. Note that in the examples in Figures 4 and 5, interlayer insulating films are omitted.

[0038] 4, the memory cell array 11 includes a WL region WR1, a cell region CR, and a WL region WR2. Hereinafter, when there is no limitation on either the WL region WR1 or the WL region WR2, it will be expressed as a "WL region WR".

[0039] The cell region CR is a region in which the memory cell transistors MC are arranged.

[0040] The WL region WR is a connection region between the word line WL and the select gate lines SGD and SGS and a plurality of contact plugs corresponding thereto. The WL regions WR1 and WR2 are provided at both ends of the cell region CR in the X direction. The WL region WR may be provided within the cell region CR. In the WL region WR, both ends of the word line WL and the select gate lines SGD and SGS extending in the X direction are drawn out in a stepped manner. Hereinafter, the region in which the word line WL and the select gate lines SGD and SGS are drawn out in a stepped manner is referred to as a "terrace". In the WL region WR, the word line WL and the select gate lines SGD and SGS do not have to be drawn out in a stepped manner. Even in this case, it is possible to form a contact plug that is electrically connected to the target wiring layer and is not electrically connected to other wiring layers.

[0041] Moreover, the memory cell array 11 includes, for example, a plurality of slits SLT and a plurality of slits SHE.

[0042] The slit SLT extends in the X direction and crosses the WL region WR1, the cell region CR, and the WL region WR2. A plurality of slits SLT are arranged in the Y direction. The slit SLT has a structure in which an insulator is embedded, for example. The slit SLT may include a conductor connected to the source line SL. The slit SLT divides adjacent wirings (for example, the word lines WL0 to WL7 and the select gate lines SGD and SGS) through the slit SLT. In the memory cell array 11, each of the regions divided by the slit SLT corresponds to one block BLK.

[0043] The slit SHE extends in the X direction and crosses the cell region CR. A plurality of slits SHE are arranged in the Y direction. In this embodiment, four slits SHE are arranged between each pair of adjacent slits SLT in the Y direction. Each slit SHE has, for example, a structure in which an insulator is embedded. Each slit SHE divides adjacent wiring (at least the selection gate line SGD) via the slit SHE. In the memory cell array 11, each of the regions divided by the slits SLT and SHE corresponds to one string unit SU.

[0044] The number of slits SHE arranged between two adjacent slits SLT can be designed to be any number. The number of string units SU included in each block BLK can be changed based on the number of slits SHE arranged between two adjacent slits SLT.

[0045] Next, the planar configuration will be described in detail.

[0046] As shown in FIG. 5, for example, in each block BLK, a plurality of wiring layers functioning as the select gate line SGS, the word lines WL0 to WL7, and the select gate line SGD are stacked at a distance in the Z direction. For example, the select gate line SGS, the word lines WL0 to WL7, and the select gate line SGD extend in the X direction and are drawn out in a stepped manner in the WL region WR. In the example of FIG. 5, in the WL region WR2, terraces corresponding to the select gate line SGS, the word lines WL0 to WL7, and the select gate line SGD are provided from the right side to the left side of the page. Slits SLT are provided on two side surfaces facing the Y direction of the select gate line SGS, the word lines WL0 to WL7, and the select gate line SGD. The slits SLT extend in the X direction and the Z direction. The slits SLT divide the select gate line SGS, the word lines WL0 to WL7, and the select gate line SGD for each block BLK. In the block BLK, the select gate line SGD is divided in the Y direction by the slits SHE. The slits SHE extend in the X direction. In the example of Fig. 5, the selection gate line SGD is divided into five by four slits SHE. Each of the areas divided by the slits SLT and SHE corresponds to one string unit SU. In the example of Fig. 5, the string units SU0 to SU4 correspond to the string units SU0 to SU4 from the top of the page. In other words, the selection gate lines SGD0 to SGD4 are arranged in the Y direction from the top of the page.

[0047] A plurality of memory pillars MP are provided in the cell region CR. The memory pillars MP are pillars corresponding to the NAND strings NS. The structure of the memory pillars MP will be described in detail later. The memory pillars MP extend in the Z direction. The memory pillars MP penetrate (pass through) the select gate lines SGS, the word lines WL0 to WL7, and the select gate lines SGD stacked in the Z direction.

[0048] 5, the memory pillars MP in the cell region CR are arranged in a staggered manner in the X direction. The arrangement of the memory pillars MP can be designed arbitrarily. For example, the arrangement of the memory pillars MP does not have to be a staggered arrangement.

[0049] The WL region includes a plurality of contact plugs CC and a plurality of dummy pillars HR.

[0050] One end of the contact plug CC is connected to a terrace of any one of the wiring layers of the select gate line SGS, the word lines WL0 to WL7, and the select gate line SGD. The contact plug CC is not electrically connected to other wiring layers. The other end of the contact plug CC is electrically connected to the row decoder 23. In the example shown in FIG. 5, 14 contact plugs CC are provided corresponding to the select gate line SGS, the word lines WL0 to WL7, and the select gate lines SGD0 to SGD4.

[0051] One method for forming the word lines WL and the select gate lines SGD and SGS is, for example, to form structures corresponding to each wiring layer using a sacrificial layer, and then to form the wiring layer by replacing the sacrificial layer with a conductive material (hereinafter referred to as "replacement"). In replacement, the sacrificial layer is removed to form a gap, and then the gap is filled with a conductive material. During replacement, the dummy pillars HR function as pillars that support the interlayer insulating film having the gap. The dummy pillars HR are not electrically connected to the word lines WL and the select gate lines SGD and SGS.

[0052] The dummy pillars HR extend in the Z direction. The dummy pillars HR penetrate (pass through) the select gate lines SGS, the word lines WL0 to WL7, and the select gate lines SGD that are stacked in the Z direction. The dummy pillars HR may be arranged in any desired location.

[0053] 1.1.5 Cross-sectional structure of semiconductor memory device Next, an example of the cross-sectional structure of the semiconductor memory device 1 will be described with reference to FIG. 6 and FIG. 7. FIG. 6 is a cross-sectional view showing an example of the cross-sectional structure of the semiconductor memory device 1 taken along line IIX-IIX in FIG. 5. FIG. 7 is a cross-sectional view showing an example of the cross-sectional structure of the semiconductor memory device 1 taken along line IX-IX in FIG. 5. In the following, the description will focus on a cross section of a region in which the memory cell array 11 is provided in the array chip 10 (hereinafter, also referred to as a "memory cell array region"). In the example of FIG. 7, one dummy pillar HR is shown in each terrace for the sake of simplicity. In the following description, when the description is made that the two are roughly the same, an error due to manufacturing variations may be included. In the following description, the description will focus on the WL region WR2, but the structure of the WL region WR1 is also similar.

[0054] As shown in FIG. 6, the semiconductor memory device 1 has a laminated structure in which an array chip 10 and a circuit chip 20 are laminated together.

[0055] The array chip 10 includes a semiconductor layer 101, a wiring layer 102, conductors 103 and 104, an M0 wiring layer 105, a conductor 106, an M1 wiring layer 107, an electrode 108, a conductor 109, insulating layers 121 to 126, and a memory pillar MP. The conductor 103 functions as a contact plug CH, and the conductor 104 functions as a contact plug VY. The conductor 106 functions as a contact plug V0. The electrode 108 functions as a bonding pad MB. The conductor 109 functions as a contact plug CC. The number of layers of the wiring layer 102 and the multilayer wiring structure on the wiring layer 102 provided on the array chip 10 can be designed arbitrarily.

[0056] The circuit chip 20 includes a semiconductor substrate 201, transistors TRs and TRr, a gate insulating film 202, a gate electrode 203, a conductor 204, a D0 wiring layer 205, a conductor 206, a D1 wiring layer 207, a conductor 208, a D2 wiring layer 209, an electrode 210, and insulating layers 211 to 215. The conductor 204 functions as a contact plug CS. The conductor 206 functions as a contact plug C0. The conductor 208 functions as a contact plug C1. The number of layers in the multilayer wiring structure provided in the circuit chip 20 can be designed arbitrarily.

[0057] 1.1.5.1 Cross-sectional structure of array chip The cross-sectional structure of the array chip 10 will now be described with reference to FIGS.

[0058] 6, the semiconductor layer 101 is provided on a surface of the insulating layer 121 facing the Z1 direction. The semiconductor layer 101 extends in the X direction and the Y direction. The semiconductor layer 101 in the memory cell array region functions as a source line SL. For example, the insulating layer 121 contains silicon oxide (SiO) as an insulating material. For example, the semiconductor layer 101 contains silicon.

[0059] In the cell region CR, a plurality of insulating layers 122 and a plurality of wiring layers 102 are alternately stacked one by one on the surface of the semiconductor layer 101 facing the Z1 direction. In the example of FIG. 6, ten insulating layers 122 and ten wiring layers 102 are alternately stacked one by one. In other words, a plurality of wiring layers 102 stacked and spaced apart in the Z direction are provided between the semiconductor layer 101 and the circuit chip 20. The wiring layers 102 and the insulating layer 122 extend in the X direction. The insulating layer 122 includes, for example, silicon oxide. The wiring layer 102 includes, for example, tungsten (W) as a conductive material. In the example of FIG. 6, the ten wiring layers 102 function as a select gate line SGS, word lines WL0 to WL7, and a select gate line SGD, in that order from the side closest to the semiconductor layer 101.

[0060] In the WL region WR, the wiring layers 102 and the insulating layers 122 are drawn out in a stepped shape. The length of the wiring layers 102 in the X direction gradually decreases from the semiconductor layer 101 side toward the circuit chip 20 side.

[0061] A plurality of memory pillars MP are provided in the cell region CR. One memory pillar MP corresponds to one NAND string NS. The memory pillar MP has, for example, a cylindrical shape extending in the Z direction. The memory pillar MP penetrates (passes through) a plurality of insulating layers 122 and a plurality of wiring layers 102. The end (bottom surface) of the memory pillar MP in the Z2 direction reaches inside the film of the semiconductor layer 101. The structure of the memory pillar MP will be described in detail later.

[0062] The insulating layer 123 is provided so as to cover the insulating layer 122, the wiring layer 102, and the memory pillar MP. For example, the insulating layer 123 includes silicon oxide. In this embodiment, the Z-direction position (hereinafter also referred to as "height position") of the surface of the insulating layer 123 facing the Z1 direction is different between the cell region CR and the WL region WR. The distance from the bonding surface of the array chip 10 and the circuit chip 20 to the surface of the insulating layer 123 facing the Z1 direction is longer in the WL region WR than in the cell region CR. In other words, the height from the surface of the semiconductor layer 101 facing the Z1 direction to the surface of the insulating layer 123 facing the Z1 direction is lower in the WL region WR than in the cell region CR.

[0063] A plurality of conductors 109 are provided in the WL region WR. The conductors 109 function as contact plugs CC. The conductors 109 have, for example, a cylindrical shape extending in the Z direction. The conductors 109 contain, for example, tungsten as a conductive material. The conductors 109 are connected to one of the wiring layers 102 and are not electrically connected to the other wiring layers 102. The length of the conductors 109 in the Z direction differs depending on the wiring layer 102 to which they are connected. The length (height of the plug) of the plurality of conductors 109 in the Z direction is the longest for the conductor 109 connected to the wiring layer 102 functioning as the select gate line SGS, and the shortest for the conductor 109 connected to the wiring layer 102 functioning as the select gate line SGD.

[0064] A conductor 103 is provided on a surface of the memory pillar MP facing the Z1 direction. The conductor 103 functions as a contact plug CH. The conductor 103 has, for example, a cylindrical shape extending in the Z direction. The conductor 103 contains, for example, tungsten as a conductive material.

[0065] A conductor 104 is provided on a surface facing the Z1 direction of the conductor 103 in the cell region CR and the conductor 109 in the WL region WR. The conductor 104 functions as a contact plug VY. The conductor 104 has, for example, a cylindrical shape extending in the Z direction. The conductor 104 contains, for example, tungsten as a conductive material.

[0066] An M0 wiring layer 105 is provided on a surface of the conductor 104 facing the Z1 direction. For example, the multiple M0 wiring layers 105 in the cell region CR each extend in the Y direction and are arranged side by side in the X direction. Each of the multiple memory pillars MP is electrically connected to one of the multiple M0 wiring layers 105 via the conductor 103 (contact plug CH) and the conductor 104 (contact plug VY). The M0 wiring layer 105 to which the memory pillar MP is connected functions as a bit line BL. The M0 wiring layer 105 contains, for example, copper (Cu) as a conductive material.

[0067] A conductor 106 is provided on the surface of the M0 wiring layer 105 facing the Z1 direction. The conductor 106 functions as a contact plug V0. The conductor 106 has, for example, a cylindrical shape extending in the Z direction. The conductor 106 contains, for example, tungsten as a conductive material.

[0068] An M1 wiring layer 107 is provided on a surface of the conductor 106 facing the Z1 direction. The M1 wiring layer 107 contains, for example, tungsten as a conductive material. The conductor 106 and the M1 wiring layer 107 may be formed by a dual damascene method. In the case of the dual damascene method, the patterns of the conductor 106 (contact plug V0) and the M1 wiring layer 107 are processed at the same time. Then, the conductor 106 and the M1 wiring layer 107 are buried at the same time with a conductive material (for example, tungsten).

[0069] The height positions of the conductor 104, the M0 wiring layer 105, the conductor 106, and the M1 wiring layer 107 in the cell region CR are different from the height positions of the conductor 104, the M0 wiring layer 105, the conductor 106, and the M1 wiring layer 107 in the WL region WR. The distances from the bonding surface to the conductor 104, the M0 wiring layer 105, the conductor 106, and the M1 wiring layer 107 are longer in the WL region WR than in the cell region CR.

[0070] An insulating layer 124 is provided on the surface of the insulating layer 123 facing the Z1 direction. The insulating layer 124 covers the conductors 103 and 104, the M0 wiring layer 105, the conductor 106, and the M1 wiring layer 107. For example, the insulating layer 124 includes silicon oxide. The film thickness of the insulating layer 124 in the cell region CR is approximately the same as the film thickness of the insulating layer 124 in the WL region WR. Therefore, similar to the insulating layer 123, the height position of the surface of the insulating layer 124 facing the Z1 direction differs between the cell region CR and the WL region WR. The distance from the bonding surface to the surface of the insulating layer 124 facing the Z1 direction is longer in the WL region WR than in the cell region CR.

[0071] An insulating layer 125 made of a material different from that of the insulating layer 124 is provided on the surface of the insulating layer 124 facing the Z1 direction. The insulating layer 125 functions as an etching stopper when the electrode 108 is formed by the dual damascene method. The insulating layer 125 includes, for example, silicon nitride (SiN) or silicon carbonitride (SiCN). Note that the insulating layer 125 may be made of any insulating material that can provide a sufficient etching selectivity with respect to the insulating layers 124 and 126 in the etching step of the dual damascene method.

[0072] The thickness of the insulating layer 125 in the cell region CR is approximately the same as the thickness of the insulating layer 125 in the WL region WR. Therefore, like the insulating layers 123 and 124, the height position of the insulating layer 125 differs between the cell region CR and the WL region WR. The distance from the bonding surface to the surface of the insulating layer 125 facing the Z1 direction is longer in the WL region WR than in the cell region CR.

[0073] An insulating layer 126 made of a material different from that of the insulating layer 125 is provided on the surface of the insulating layer 125 facing the Z1 direction. The insulating layer 126 includes, for example, silicon oxide. The surface of the insulating layer 126 facing the Z1 direction is flattened in the cell region CR and the WL region WR. Therefore, the film thickness of the insulating layer 126 is thicker in the WL region WR than in the cell region CR. The surface of the insulating layer 126 facing the Z1 direction contacts the insulating layer 215 of the circuit chip 20. In other words, the surface where the insulating layer 126 and the insulating layer 215 contact is the bonding surface.

[0074] An electrode 108 is provided on the surface of the M1 wiring layer 107 facing the Z1 direction. The electrode 108 includes, for example, copper as a conductive material. The electrode 108 includes a bonding pad MB and a contact plug that connects the bonding pad MB and the M1 wiring layer 107. The structure of the electrode 108 will be described in detail later.

[0075] The bonding pad MB contacts the bonding pad DB provided on the circuit chip 20 on the bonding surface. Hereinafter, when the bonding pad MB provided in the cell region CR is limited, it is written as "bonding pad MBc". When the bonding pad MB provided in the WL region WR is limited, it is written as "bonding pad MBw". Each of the multiple M0 wiring layers 105 (bit lines BL) in the cell region CR is electrically connected to one of the bonding pads MBc. In other words, the memory pillar MP (channel of the memory cell transistor MC) is electrically connected to the bonding pad MBc via the bit line BL. Each of the multiple conductors 109 (contact plugs CC) in the WL region WR is electrically connected to one of the bonding pads MBw. In other words, the word line WL and the selection gate lines SGD and SGS are electrically connected to the bonding pad MBw via the conductor 109.

[0076] The electrode 108 is formed by, for example, a dual damascene method. When the electrode 108 is formed by the dual damascene method, the insulating layer 125 functions as an etching stopper for the bonding pad MB. By using the insulating layer 125, the controllability of the height between the bonding pad MB and the contact plug is improved. The bonding pad MB is formed in the same layer as the insulating layer 126. Therefore, the thickness of the bonding pad MB is determined by the film thickness of the insulating layer 126. In this embodiment, the film thickness of the insulating layer 126 is thicker in the WL region WR than in the cell region CR. Therefore, the thickness of the bonding pad MBw is larger than the thickness of the bonding pad MBc. In addition, the thickness of the bonding pad MBw to which the selection gate line SGD is connected may be the same as the thickness of the bonding pad MBw to which the word line WL or the selection gate line SGS is connected, may be the same as the thickness of the bonding pad MBc, or may be an intermediate thickness between the thickness of the bonding pad MBw to which the word line WL or the selection gate line SGS is connected and the thickness of the bonding pad MBc.

[0077] Next, the dummy pillar HR will be described.

[0078] As shown in FIG. 7, a plurality of dummy pillars HR are provided in the WL region WR. The structure of the dummy pillars HR in this embodiment is similar to that of the memory pillars MP. However, the height from the surface of the semiconductor layer 101 facing the Z1 direction to the surface of the insulating layer 123 facing the Z1 direction is lower in the WL region WR than in the cell region CR. Therefore, the length in the Z direction of the dummy pillars HR (pillar height) is shorter than that of the memory pillars MP. Note that the structure of the dummy pillars HR may be different from that of the memory pillars MP.

[0079] For example, the dummy pillar HR provided on the terrace corresponding to the select gate line SGS penetrates the wiring layer 102 functioning as the select gate line SGS. The dummy pillar HR provided on the terrace corresponding to the word line WL0 penetrates the multiple wiring layers 102 functioning as the select gate line SGS and the word line WL0. The dummy pillar HR provided on the terrace corresponding to the word line WL1 penetrates the multiple wiring layers 102 functioning as the select gate line SGS and the word lines WL0 and WL1. The dummy pillar HR provided on the terrace corresponding to the word line WL2 penetrates the multiple wiring layers 102 functioning as the select gate line SGS and the word lines WL0 to WL2. The same applies to the dummy pillar HR provided on the terrace corresponding to the word lines WL3 to WL7 and the select gate line SGD.

[0080] The surface of the dummy pillar HR facing the Z1 direction is not provided with the conductor 103 or the conductor 104. The dummy pillar HR is not electrically connected to the M0 wiring layer 105 and the like.

[0081] 1.1.5.2 Cross-sectional structure of a circuit chip The cross-sectional structure of the circuit chip 20 will now be described with reference to FIG.

[0082] As shown in Fig. 6, a plurality of transistors TR are provided on a surface of a semiconductor substrate 201 facing the Z2 direction. Each transistor TR includes a gate insulating film 202, a gate electrode 203, and a source and a drain (not shown) formed on the semiconductor substrate 201. The gate insulating film 202 is provided on the surface of the semiconductor substrate 201 facing the Z2 direction. The gate electrode 203 is provided on the surface of the gate insulating film 202 facing the Z2 direction. Hereinafter, the transistor TR used in the row decoder 23 will be referred to as a "transistor TRr". The transistor TR used in the sense amplifier 24 will be referred to as a "transistor TRs".

[0083] A conductor 204 is provided on the gate electrode 203 and the surface of the semiconductor substrate 201 facing the Z2 direction. The conductor 204 functions as a contact plug CS. The conductor 204 includes, for example, tungsten. The conductor 204 has, for example, a cylindrical shape extending in the Z direction.

[0084] A D0 wiring layer 205 is provided on the surface facing the Z2 direction of the conductor 204. The D0 wiring layer 205 includes, for example, tungsten as a conductive material.

[0085] A conductor 206 is provided on the surface of the D0 wiring layer 205 facing the Z2 direction. The conductor 206 functions as a contact plug C0. The conductor 206 includes, for example, tungsten or copper as a conductive material. The conductor 206 has, for example, a cylindrical shape extending in the Z direction.

[0086] A D1 wiring layer 207 is provided on the surface of the conductor 206 facing the Z2 direction. The D1 wiring layer 207 includes, for example, tungsten or copper as a conductive material. The D1 wiring layer 207 and the conductor 206 may be formed together by a dual damascene method.

[0087] A conductor 208 is provided on the surface of the D1 wiring layer 207 facing the Z2 direction. The conductor 208 functions as a contact plug C1. The conductor 208 includes, for example, copper as a conductive material. The conductor 208 has, for example, a cylindrical shape extending in the Z direction.

[0088] A D2 wiring layer 209 is provided on the surface of the conductor 208 facing the Z2 direction. The D2 wiring layer 209 includes, for example, copper as a conductive material. The D2 wiring layer 209 and the conductor 208 may be formed together by a dual damascene method.

[0089] An insulating layer 211 is provided on the surface facing the Z2 direction of the semiconductor substrate 201. The insulating layer 211 covers the transistor TR, the conductor 204, the D0 wiring layer 205, the conductor 206, the D1 wiring layer 207, the conductor 208, and the D2 wiring layer 209. The insulating layer 211 includes, for example, silicon oxide.

[0090] An insulating layer 212 is provided on the surfaces facing the Z2 direction of the insulating layer 211 and the D2 wiring layer 209. The insulating layer 212 functions as a cap insulating layer (anti-oxidation layer) for the D2 wiring layer 209. The insulating layer 212 includes, for example, silicon nitride or silicon carbonitride.

[0091] An insulating layer 213 is provided on the surface of the insulating layer 212 facing the Z2 direction. The insulating layer 213 includes, for example, silicon oxide.

[0092] An insulating layer 214 made of a material different from that of the insulating layer 213 is provided on the surface of the insulating layer 213 facing the Z2 direction. The insulating layer 214 functions as an etching stopper when the electrode 210 is formed by the dual damascene method. The insulating layer 214 includes, for example, silicon nitride or silicon carbonitride. Note that the insulating layer 214 may be made of any insulating material that can provide a sufficient etching selectivity with respect to the insulating layers 213 and 215 in the etching step of the dual damascene method.

[0093] An insulating layer 215 made of a material different from that of the insulating layer 214 is provided on the surface of the insulating layer 214 facing the Z2 direction. The insulating layer 215 includes, for example, silicon oxide. The surface of the insulating layer 215 facing the Z2 direction contacts the insulating layer 126 of the array chip 10.

[0094] An electrode 210 is provided on the surface of the D2 wiring layer 209 facing the Z2 direction. The electrode 210 includes, for example, copper as a conductive material. The electrode 210 includes a bonding pad DB and a contact plug that connects the bonding pad DB and the D2 wiring layer 209. The structure of the electrode 210 will be described in detail later.

[0095] The bonding pad DB contacts the bonding pad MB provided on the array chip 10 on the bonding surface. Hereinafter, when the bonding pad DB connected to the bonding pad MBc is specified, it is written as "bonding pad DBc". When the bonding pad DB connected to the bonding pad MBw is specified, it is written as "bonding pad DBw". The bonding pad DBc of the cell region CR is electrically connected to the transistor TRs used in the sense amplifier 24. The bonding pad DBw of the WL region WR is electrically connected to the transistor TRr used in the row decoder 23.

[0096] The electrode 210 is formed by, for example, a dual damascene method. When the electrode 210 is formed by the dual damascene method, the insulating layer 214 functions as an etching stopper for the bonding pad DB. By using the insulating layer 214, the controllability of the height between the bonding pad DB and the contact plug is improved. The bonding pad DB is formed in the same layer as the insulating layer 215.

[0097] The number of layers in the multilayer wiring structure provided between the transistor TR and the electrode 210 is arbitrary.

[0098] 1.1.6 Cross-sectional structure of the bonding pad Next, the cross-sectional structure of the lamination pad BP will be described with reference to Fig. 8. Fig. 8 is a cross-sectional view showing an example of the cross-sectional structure of the lamination pad BP. In the example of Fig. 8, a lamination pad BP that pairs lamination pads MBc and DBc provided in the cell region CR, and a lamination pad BP that pairs lamination pads MBw and DBw provided in the WL region WR are shown.

[0099] First, the electrodes 108 of the array chip 10 will be described.

[0100] As shown in FIG. 8, the electrode 108 has a bonding pad MB and a contact plug VB that connects the bonding pad MB and the M1 wiring layer 107.

[0101] The bonding pad MB has a frustum shape in which the area of ​​the surface facing the Z1 direction of the bonding pad MB (hereinafter, referred to as the "lower surface of the bonding pad MB") is larger than the area of ​​the surface facing the Z2 direction (hereinafter, referred to as the "upper surface of the bonding pad MB"). The shapes of the upper and lower surfaces of the bonding pad MB are not limited. The bonding pad MB may be a quadrangular pyramid or a circular cone. The side surface of the bonding pad MB has a tapered shape. The lower surface of the bonding pad MB is the bonding surface.

[0102] The contact plug VB connects the bonding pad MB and the M1 wiring layer 107. The contact plug VB has a frustum shape protruding from the upper surface of the bonding pad MB in the Z2 direction. That is, the contact plug VB contacts the upper surface of the bonding pad MB. The area of ​​the surface of the contact plug VB facing the Z1 direction (hereinafter referred to as the "lower surface of the contact plug VB") is larger than the area of ​​the surface facing the Z2 direction (hereinafter referred to as the "upper surface of the contact plug VB"). And, the area of ​​the lower surface of the contact plug VB is smaller than the area of ​​the upper surface of the bonding pad MB. The shapes of the upper and lower surfaces of the contact plug VB are not limited. The contact plug VB may be a quadrangular pyramid or a circular cone. The side of the contact plug VB has a tapered shape. The upper surface of the contact plug VB contacts the M1 wiring layer 107.

[0103] For example, the bonding pad MB and the contact plug VB are formed at the same time by the dual damascene method. The bonding pad MB is provided in the same layer as the insulating layer 126. The upper surface of the bonding pad MB reaches the insulating layer 125. The insulating layer 125 functions as an etching stopper of the bonding pad MB and a mask of the contact plug VB. The contact plug VB is provided in the insulating layer 124. In the example of FIG. 8, the surface of the M1 wiring layer 107 facing the Z1 direction is in contact with the insulating layer 124, but a cap insulating layer (anti-oxidation layer) may be provided on the surface of the M1 wiring layer 107 facing the Z1 direction. For example, when the M1 wiring layer 107 contains copper as a conductive material, silicon nitride or silicon carbonitride may be provided as the cap insulating layer. In this case, the upper surface of the bonding pad MB may pass through the insulating layer 125 and reach the insulating layer 124.

[0104] The electrode 108 includes a barrier metal (conductor) 108_1 and a conductor 108_2. The barrier metal 108_1 functions as an oxidation prevention layer, a diffusion prevention layer, and an adhesion layer for the conductor 108_2. The barrier metal 108_1 is formed so as to cover the side and upper surface of the bonding pad MB and the side and upper surface of the contact plug VB. The barrier metal 108_1 contacts the M1 wiring layer 107 and the insulating layers 124 to 126. The conductor 108_2 is provided in the barrier metal 108_1 in contact with the barrier metal 108_1, and buries the bonding pad MB and the contact plug VB. The barrier metal 108_1 includes, for example, titanium (Ti), titanium nitride (TiN), tantalum (Ta), or tantalum nitride (TaN). The barrier metal 108_1 may be a laminated film of these materials. The conductor 108_2 includes, for example, copper as a conductive material.

[0105] The thickness of the bonding pad MBw provided in the WL region WR is larger than the thickness of the bonding pad MBc provided in the cell region CR. The area of ​​the lower surface of the bonding pad MBw is approximately the same as the area of ​​the lower surface of the bonding pad MBc. Therefore, the volume of the bonding pad MBw is larger than the volume of the bonding pad MBc.

[0106] More specifically, in the WL region WR, the length from the bonding surface to the upper surface of the bonding pad MBw is Hmw1. Then, the length from the bonding surface to the upper surface of the contact plug VB, that is, the length in the Z direction of the electrode 108 is Hmw2. Also, in the cell region CR, the length from the bonding surface to the upper surface of the bonding pad MBc is Hmc1. Then, the length from the bonding surface to the upper surface of the contact plug VB, that is, the length in the Z direction of the electrode 108 is Hmc2. The length Hmw1 and the length Hmc1 have a relationship of Hmw1>Hmc1. The length in the Z direction of the contact plug VB in the WL region WR (Hmw2-Hmw1) is approximately the same as the length in the Z direction of the contact plug VB in the cell region CR (Hmc2-Hmc1). Therefore, the length Hmw2 and the length Hmc2 have a relationship of Hmw2>Hmc2.

[0107] The height position of the insulating layer 125 is different between the cell region CR and the WL region WR. In other words, the height position of the insulating layer 125 in contact with the bonding pad MBc is different from the height position of the insulating layer 125 in contact with the bonding pad MBw. Therefore, the film thickness of the insulating layer 126 is different between the cell region CR and the WL region WR. The film thickness of the insulating layer 126 in the cell region CR, that is, the distance from the bonding surface to the insulating layer 125 in the cell region CR, is, for example, approximately equal to the length Hmc1 of the bonding pad MBc. In addition, the film thickness of the insulating layer 126 in the WL region WR, that is, the distance from the bonding surface to the insulating layer 125 in the WL region WR, is, for example, approximately equal to the length Hmw1 of the bonding pad MBw.

[0108] In the bonding process for bonding the array chip 10 and the circuit chip 20, when the electrode 108 (bonding pad MB) and the electrode 210 (bonding pad DB) are bonded together, an annealing process is performed. The length Hmw1 of the bonding pad MBw is greater than the length Hmc1 of the bonding pad MBc. That is, the volume of the bonding pad MBw is greater than the volume of the bonding pad MBc. Therefore, the thermal expansion amount of the conductor 108_2 due to the annealing process is greater in the bonding pad MBw than in the bonding pad MBc. Therefore, in the WL region WR, the contact area between the conductor 108_2 and the conductor 210_2 increases. As a result, bonding defects between the bonding pad MBw and the bonding pad DBw can be reduced. It is preferable that the difference (Hmw1-Hmc1) between the thickness of the bonding pad MBw and the thickness of the bonding pad MBc is 50 nm±40 nm. If the thickness difference (Hmw1-Hmc1) is greater than 90 nm, for example, when the conductive material embedded in the M1 wiring layer 107 is removed by CMP, the conductive material residue is likely to occur in the step. Also, if the thickness difference (Hmw1-Hmc1) is less than 10 nm, it becomes difficult to obtain the effect of improving the bonding failure in the WL region WR described above.

[0109] In the cell region CR, the ratio of the length (Hmc2-Hmc1) of the contact plug VB to the length Hmc2 of the electrode 108 is preferably 40% or less. Furthermore, in the WL region WR, the ratio of the length (Hmw2-Hmw1) of the contact plug VB to the length Hmw2 of the electrode 108 is preferably 35% or less. If the ratio of the contact plug VB is larger than these values, the tensile stress of the conductor 108_2 of the contact plug VB pulled by the bonding pad MB during the annealing process increases. As a result, disconnection between the electrode 108 and the M1 wiring layer 107 is likely to occur at the bottom of the contact plug VB. In addition, the aspect ratio of the contact plug VB (the ratio of the height of the contact plug VB to the diameter of the lower surface of the contact plug VB) is preferably 1.5 or less. When the aspect ratio of the contact plug VB becomes larger than 1.5, that is, when the ratio of the height of the contact plug VB becomes large, the tensile stress increases, and disconnection between the electrode 108 and the M1 wiring layer 107 becomes more likely to occur.

[0110] Next, the electrodes 210 of the circuit chip 20 will be described.

[0111] The electrode 210 has a bonding pad DB and a contact plug CB that connects the bonding pad DB and the D2 wiring layer 209. In this embodiment, a case will be described in which the shape of the electrode 210 in the cell region CR is approximately the same as the shape of the electrode 210 in the WL region WR. Note that, like the electrode 108 of the array chip 10, the thickness of the electrode 210 in the WL region WR may be greater than the thickness of the electrode 210 in the cell region CR.

[0112] The bonding pad DB has, for example, a frustum shape in which the area of ​​the surface facing the Z2 direction of the bonding pad DB (hereinafter referred to as the "upper surface of the bonding pad DB") is larger than the area of ​​the surface facing the Z1 direction (hereinafter referred to as the "lower surface of the bonding pad DB"). The shapes of the upper and lower surfaces of the bonding pad DB are not limited. The shape of the upper surface of the bonding pad DB is approximately the same as the shape of the lower surface of the bonding pad MB of the electrode 108. The side surface of the bonding pad DB has a tapered shape. The upper surface of the bonding pad DB is a bonding surface.

[0113] The contact plug CB connects the bonding pad DB and the D2 wiring layer 209. The contact plug CB has a frustum shape protruding from the lower surface of the bonding pad DB in the Z1 direction. That is, the contact plug CB contacts the lower surface of the bonding pad DB. The area of ​​the surface of the contact plug CB facing the Z2 direction (hereinafter referred to as the "upper surface of the contact plug CB") is larger than the area of ​​the surface facing the Z1 direction (hereinafter referred to as the "lower surface of the contact plug CB"). The area of ​​the upper surface of the contact plug CB is smaller than the area of ​​the lower surface of the bonding pad DB. The shapes of the upper and lower surfaces of the contact plug CB are not limited. The contact plug CB may be a quadrangular pyramid or a circular cone. The side of the contact plug CB has a tapered shape. The lower surface of the contact plug CB contacts the D2 wiring layer 209.

[0114] For example, the bonding pad DB and the contact plug CB are formed at the same time by the dual damascene method. The bonding pad DB is provided in the same layer as the insulating layer 215. The lower surface of the bonding pad DB passes through the insulating layer 214 and reaches the insulating layer 213. The insulating layer 214 functions as an etching stopper of the bonding pad DB and a mask of the contact plug CB. The contact plug CB passes through the insulating layer 213 and the insulating layer 212. In the example of FIG. 8, the insulating layer 212 is provided as a cap insulating layer on the surface of the D2 wiring layer 209 facing the Z2 direction, but the insulating layer 212 may be omitted. For example, when the D2 wiring layer 209 contains tungsten as a conductive material, the insulating layer 212 can be omitted. In this case, the lower surface of the bonding pad DB reaches the inside of the insulating layer 214, similar to the upper surface of the bonding pad MB.

[0115] The electrode 210 includes a barrier metal (conductor) 210_1 and a conductor 210_2. The barrier metal 210_1 functions as an oxidation prevention layer, a diffusion prevention layer, and an adhesion layer for the conductor 210_2. The barrier metal 210_1 is formed so as to cover the side and lower surface of the bonding pad DB and the side and lower surface of the contact plug CB. The barrier metal 210_1 contacts the D2 wiring layer 209 and the insulating layers 212 to 215. The conductor 210_2 is provided in the barrier metal 210_1 in contact with the barrier metal 210_1, and embeds the bonding pad DB and the contact plug CB. The barrier metal 210_1 includes, for example, titanium, titanium nitride, tantalum, or tantalum nitride. The barrier metal 210_1 may be a laminated film of these materials. The conductor 210_2 includes, for example, copper as a conductive material.

[0116] The thickness of the lamination pad DB is approximately the same in the cell region CR and the WL region WR. In other words, the thickness of the lamination pad DBc is approximately the same as that of the lamination pad DBw. That is, the thickness of the electrode 210 in the cell region CR is approximately the same as that of the electrode 210 in the WL region WR.

[0117] The length from the bonding surface to the lower surface of the bonding pad DB is Hd1. The length from the bonding surface to the lower surface of the contact plug CB, that is, the length of the electrode 210 in the Z direction is Hd2. For example, when the length Hd2 is approximately equal to the length Hmc2, that is, when the shape of the electrode 108 in the cell region CR is approximately the same as the shape of the electrode 210, the ratio of the length of the contact plug CB in the Z direction (Hd2-Hd1) to the length Hd2 of the electrode 210 is preferably 40% or less. Also, similar to the electrode 108, the aspect ratio of the contact plug CB (the ratio of the height of the contact plug CB to the diameter of the upper surface of the contact plug CB) is preferably 1.5 or less.

[0118] For example, if copper is used for the conductors 108_2 and 210_2, the copper of the conductor 108_2 and the copper of the conductor 210_2 may become integrated, making it difficult to confirm the boundary between the copper parts. However, the bonding can be confirmed by the distortion of the shape of the bonded electrode 108 and the electrode 210 due to the misalignment of the bonding, and the misalignment of the barrier metal 108_1 and the barrier metal 210_1 (the occurrence of discontinuous parts on the side surfaces).

[0119] Furthermore, when the electrodes 108 and 210 are formed by the dual damascene method, the side surfaces of the electrodes 108 and 210 have a tapered shape, so that the cross section along the Z direction at the portion where the electrodes 108 and 210 are bonded together does not have a straight sidewall but has a non-rectangular shape.

[0120] Furthermore, when the electrodes 108 and 210 are bonded together, the bottom, side, and top surfaces of the copper that forms them are covered with a barrier metal. In contrast, in a typical wiring layer using copper, a cap insulating layer (SiN, SiCN, or the like) that has a function of preventing oxidation of copper is provided on the top surface of the copper, and no barrier metal is provided. Therefore, even if there is no misalignment in the bonding, it is possible to distinguish it from a typical wiring layer.

[0121] 1.1.7 Cross-sectional structure of memory cell array Next, a cross-sectional structure of the memory cell array 11 will be described with reference to Fig. 9. Fig. 9 is a cross-sectional view showing an example of the cross-sectional structure of the memory cell array 11. In Fig. 9, two memory pillars MP included in the memory cell array 11 are shown.

[0122] 9, the semiconductor layer 101 includes, for example, three semiconductor layers 101a, 101b, and 101c. The semiconductor layer 101b is provided on a surface of the semiconductor layer 101a facing the Z1 direction. The semiconductor layer 101c is provided on a surface of the semiconductor layer 101b facing the Z1 direction. The semiconductor layer 101b is formed, for example, by replacing a sacrificial layer provided between the semiconductor layers 101a and 101c. The semiconductor layers 101a to 101c include, for example, silicon. The semiconductor layers 101a to 101c also include, for example, phosphorus (P) as a semiconductor impurity.

[0123] Ten insulating layers 122 and ten wiring layers 102 are alternately stacked on the surface of the semiconductor layer 101 facing the Z1 direction. In the example of FIG. 9, the ten wiring layers 102 function as the select gate line SGS, the word lines WL0 to WL7, and the select gate line SGD, in that order from the side closer to the semiconductor layer 101. Note that a plurality of wiring layers 102 functioning as the select gate lines SGS and SGD may be provided. For example, a titanium nitride (TiN) / tungsten laminated structure may be used as the conductive material of the wiring layer 102. In this case, the titanium nitride is formed so as to cover the tungsten. The titanium nitride functions as a barrier layer for suppressing oxidation of the tungsten or an adhesion layer for improving the adhesion of the tungsten when forming a film of the tungsten by, for example, CVD (chemical vapor deposition). The wiring layer 102 may also include a high-dielectric constant material such as aluminum oxide (AlO). In this case, the high-dielectric constant material is formed so as to cover the conductive material. For example, in each of the wiring layers 102, a high dielectric constant material is provided so as to be in contact with the insulating layers 122 provided above and below the wiring layer 102 and the side surfaces of the memory pillars MP. Then, titanium nitride is provided so as to be in contact with the high dielectric constant material. Then, tungsten is provided so as to be in contact with the titanium nitride and to fill the inside of the wiring layer 102. For example, when aluminum oxide is provided as the high dielectric constant material, the memory cell transistor MC is also referred to as a MANOS (Metal-Aluminum-Nitride-Oxide-Silicon) type.

[0124] An insulating layer 123 is provided on the surface of the wiring layer 102 that functions as the selection gate line SGD, facing the Z1 direction.

[0125] A plurality of memory pillars MP are provided in the memory cell array 11. For example, the memory pillars MP have a generally cylindrical shape extending in the Z direction. The memory pillars MP penetrate ten wiring layers 102. The bottom surfaces of the memory pillars MP reach the semiconductor layer 101. Note that the memory pillars MP may have a structure in which a plurality of pillars are connected in the Z direction.

[0126] Next, the internal configuration of the memory pillar MP will be described. The memory pillar MP includes a block insulating film 140, a charge storage film 141, a tunnel insulating film 142, a semiconductor film 143, a core film 144, and a cap film 145.

[0127] A block insulating film 140, a charge storage film 141, and a tunnel insulating film 142 are laminated on a part of the side surface of the memory pillar MP and the bottom surface facing the Z2 direction, in that order from the outside. More specifically, the block insulating film 140, the charge storage film 141, and the tunnel insulating film 142 on the side surface of the memory pillar MP are removed in the same layer as the semiconductor layer 101b and in the vicinity thereof. A semiconductor film 143 is provided so as to contact the side surface and bottom surface of the tunnel insulating film 142 and the semiconductor layer 101b. The semiconductor film 143 is a region in which the channels of the memory cell transistor MC and the select transistors ST1 and ST2 are formed. The inside of the semiconductor film 143 is filled with a core film 144. At the upper part of the memory pillar MP in the Z1 direction, a cap film 145 is provided at the upper ends of the semiconductor film 143 and the core film 144. The side surface of the cap film 145 contacts the tunnel insulating film 142. The cap film 145 includes, for example, silicon. A conductor 103 is provided on a surface facing the Z1 direction of the cap film 145. A conductor 104 is provided on a surface facing the Z1 direction of the conductor 103. The conductor 104 is connected to an M0 wiring layer 105 that functions as a bit line BL.

[0128] Memory cell transistors MC0-MC7 are formed by combining a memory pillar MP with a wiring layer 102 functioning as word lines WL0-WL7. Similarly, a select transistor ST1 is formed by combining a memory pillar MP with a wiring layer 102 functioning as a select gate line SGD. A select transistor ST2 is formed by combining a memory pillar MP with a wiring layer 102 functioning as a select gate line SGS. Each memory pillar MP can thus function as one NAND string NS.

[0129] An example of a cross-sectional structure of a memory pillar MP along the XY plane is shown with reference to Fig. 10. Fig. 10 is a cross-sectional view along the XX line in Fig. 9. More specifically, Fig. 10 shows the cross-sectional structure of the memory pillar MP in a layer including the wiring layer 102.

[0130] In a cross section including the wiring layer 102, the core film 144 is provided, for example, in the center of the memory pillar MP. The semiconductor film 143 surrounds the side surfaces of the core film 144. The tunnel insulating film 142 surrounds the side surfaces of the semiconductor film 143. The charge storage film 141 surrounds the side surfaces of the tunnel insulating film 142. The block insulating film 140 surrounds the side surfaces of the charge storage film 141. The wiring layer 102 surrounds the side surfaces of the block insulating film 140.

[0131] The semiconductor film 143 is used as a channel (current path) of the memory cell transistors MC0 to MC7 and the select transistors ST1 and ST2. Each of the tunnel insulating film 142 and the block insulating film 140 includes, for example, silicon oxide. The charge storage film 141 has a function of storing charges. The charge storage film 141 includes, for example, silicon nitride (SiN).

[0132] 1.2 Manufacturing method of memory cell array and electrodes Next, an example of a manufacturing method of the memory cell array 11 and the electrodes 108 will be described with reference to Figs. 11 to 20. Figs. 11 to 20 are cross-sectional views showing the manufacturing process of the memory cell array 11 and the electrodes 108 of the array chip 10. The examples of Figs. 11, 12, and 14 to 20 show an example of the cross-sectional structure of the array chip 10 cut along the line IIX-IIX in Fig. 5. The example of Fig. 13 shows an example of the cross-sectional structure of the array chip 10 cut along the line IX-IX in Fig. 5.

[0133] 11, first, an insulating layer 121 is formed on the semiconductor substrate 100 of the array chip 10. Note that the semiconductor substrate 100 of the array chip 10 may be removed after the array chip 10 and the circuit chip 20 are bonded together. That is, the semiconductor substrate 100 may be discarded in the semiconductor memory device 1 having a bonded structure. Next, the semiconductor layer 101 is formed on the insulating layer 121. More specifically, the semiconductor layer 101a, the sacrificial layer 130, and the semiconductor layer 101c are formed in this order on the insulating layer 121. The sacrificial layer 130 is later replaced by the semiconductor layer 101b. Next, in the memory cell array region, a plurality of insulating layers 122 and a plurality of sacrificial layers 131 are alternately stacked one by one. The sacrificial layer 131 is later replaced by the wiring layer 102. The sacrificial layer 131 includes, for example, silicon nitride. After forming a step of ten sacrificial layers 131 in the WL region WR, the insulating layer 123 is formed so as to cover the entire surface of the semiconductor substrate 100 facing the Z1 direction. Next, CMP (Chemical Mechanical Polishing) of the insulating layer 123 is performed. At this time, the insulating layer 123 is not completely flattened, and a step of height Ha is formed between the cell region CR and the WL region WR. The method of forming the step with height Ha in the insulating layer 123 is not limited to the CMP of the insulating layer 123. For example, the step may be formed by using photolithography and dry etching.

[0134] As shown in FIG. 12, a memory pillar MP is formed in the cell region CR. At this time, as shown in FIG. 13, a dummy pillar HR is simultaneously formed in the WL region WR. The following description focuses on the memory pillar MP. More specifically, first, a memory hole corresponding to the memory pillar MP is formed. The memory hole penetrates the insulating layer 123, the sacrificial layer 131, the insulating layer 122, the semiconductor layer 101c, and the sacrificial layer 130. The bottom surface of the memory hole reaches the inside of the semiconductor layer 101a. The block insulating film 140, the charge storage film 141, the tunnel insulating film 142, the semiconductor film 143, and the core film 144 are sequentially formed to fill the memory hole. Next, the semiconductor film 143 and the core film 144 on the upper part of the memory pillar MP are removed, and then the cap film 145 is formed. Thereafter, the block insulating film 140, the charge storage film 141, the tunnel insulating film 142, the semiconductor film 143, the core film 144, and the cap film 145 on the surface of the insulating layer 123 facing the Z1 direction are removed.

[0135] As shown in FIG. 14, in order to protect the surface of the memory pillar MP, for example, an insulating layer 124 corresponding to the height of the conductor 103 is formed. Next, the sacrificial layer 130 is replaced with the semiconductor layer 101b. More specifically, for example, a slit SLT is formed in a region not shown in the memory cell array 11. The bottom surface of the slit SLT reaches into the film of the sacrificial layer 130. For example, the sacrificial layer 130 is removed from the side surface of the slit SLT by wet etching. At this time, parts of the block insulating film 140, the charge storage film 141, and the tunnel insulating film 142 located in the same layer as the sacrificial layer 130 are also removed at the same time. The semiconductor layer 101b is formed in the region where the sacrificial layer 130, the block insulating film 140, the charge storage film 141, and the tunnel insulating film 142 have been removed. As a result, the semiconductor film 143 of the memory pillar MP and the semiconductor layer 101 are connected.

[0136] Next, the sacrificial layer 131 is replaced with the wiring layer 102. More specifically, the sacrificial layer 131 is removed from the side surface of the slit SLT by, for example, wet etching. The wiring layer 102 is formed in the region where the sacrificial layer 131 has been removed.

[0137] 15, conductors 103 and 109 are formed. More specifically, for example, a hole pattern corresponding to the conductors 103 and 109 is formed. After the hole pattern is filled with a conductive material, the excess conductive material on the insulating layer 124 is removed by CMP.

[0138] Using a similar procedure, the conductor 104, the M0 wiring layer 105, the conductor 106, and the M1 wiring layer 107 are sequentially formed. At this time, a step having a height Hb remains between the cell region CR and the WL region WR. The height Hb is equal to or smaller than the height Ha. The step can be reduced from the height Ha to the height Hb by the CMP process used to form the multilayer wiring structure described above.

[0139] 16, insulating layers 124 and 125 are formed. Next, the insulating layer 125 is processed to form a mask pattern for forming the contact plugs VB.

[0140] 17, an insulating layer 126 is formed. At this time, a step having a height Hb remains between the cell region CR and the WL region WR.

[0141] 18, the surface of the insulating layer 126 is planarized by CMP. This substantially eliminates the step between the cell region CR and the WL region WR. As a result, the thickness of the insulating layer 126 in the WL region WR is thicker than the thickness of the insulating layer 126 in the cell region CR.

[0142] As shown in FIG. 19, a pattern of the electrode 108 is formed by the dual damascene method. More specifically, for example, a resist mask corresponding to the bonding pad MB of the electrode 108 is formed on the insulating layer 126. Next, a pattern corresponding to the bonding pad MB of the electrode 108 and the contact plug VB is formed at the same time by dry etching. At this time, the insulating layer 126 is first etched to form a pattern of the bonding pad MB whose bottom surface reaches the insulating layer 125. When the insulating layer 125 is exposed, the insulating layer 125 functions as a mask for the contact plug VB. Furthermore, the insulating layer 124 in the opening portion of the insulating layer 125 is etched. As a result, a pattern of the contact plug VB is formed. The film thickness of the insulating layer 126 in the WL region WR is thicker than the film thickness of the insulating layer 126 in the cell region CR. Therefore, the thickness of the bonding pad MBw in the WL region WR is greater than the thickness of the bonding pad MBc in the cell region CR.

[0143] 20, the electrode 108 is formed. More specifically, a barrier metal 108_1 and a conductor 108_2 are sequentially formed. Next, the excess barrier metal 108_1 and the conductor 108_2 on the insulating layer 126 are removed by CMP, thereby forming the electrode 108.

[0144] 1.3 Effects of this embodiment The configuration according to this embodiment can provide a semiconductor memory device in which bonding defects are reduced.

[0145] In a lamination structure in which an array chip and a circuit chip are laminated together, an electrode (lamination pad) on the array chip side and an electrode (lamination pad) on the circuit chip side are laminated together. For example, each electrode is formed by CMP. When an electrode is formed by CMP, the conductive material (e.g., copper) that embeds the electrode may have a concave shape with respect to the lamination surface, i.e., the surface of the insulating layer, due to the influence of dishing, etc. For this reason, the effective contact area between the conductive material of the electrode on the array chip side and the conductive material of the electrode on the circuit chip side may be smaller than the opposing area between the electrode pattern in the insulating layer on the array chip side and the electrode pattern in the insulating layer on the circuit chip side. In this case, there is a possibility that a void will be formed in the lamination pad on the lamination surface. For example, if there is a void, even if the initial inspection after lamination is judged to be a good product, the subsequent current flow will cause electromigration of the conductive material, increasing the possibility of a break in the line. This tendency is more noticeable in the lamination pads in the WL region corresponding to the word lines to which a relatively high voltage is applied in the write operation, read operation, or erase operation than in the lamination pads in the cell region.

[0146] In addition, when an annealing process is performed in the bonding process, tensile stress is generated in the conductive material in which the electrode is embedded. For example, the electrode has a bonding pad and a contact plug connected to the bonding pad, and the bonding pad and the contact plug are embedded with the same conductive material. In this case, the conductive material of the contact plug is pulled toward the bonding pad by the tensile stress, which may cause a disconnection failure of the contact plug.

[0147] In contrast, in the configuration according to this embodiment, the Z-direction length Hmw1 of the lamination pad MBw of the array chip 10 can be made larger than the Z-direction length Hmc1 of the lamination pad MBc. That is, the Z-direction length Hmw2 of the electrode 108 of the WL region WR can be made larger than the Z-direction length Hmc2 of the electrode 108 of the cell region CR. This makes it possible to increase the volume of the lamination pad MBw of the WL region WR of the array chip 10 more than the volume of the lamination pad MBc of the cell region CR. Therefore, in the annealing process of the lamination process, the thermal expansion amount of the conductor 108_2 of the electrode 108 of the WL region WR can be made larger than the thermal expansion amount of the conductor 108_2 of the electrode 108 of the cell region CR. As a result, in the WL region WR where lamination defects due to electromigration are likely to occur, the effective contact area between the conductor 108_2 of the electrode 108 and the conductor 210_2 of the electrode 210 on the circuit chip side can be increased. Therefore, lamination defects in the WL region WR can be reduced.

[0148] For example, a pattern of relatively wide grooves such as an alignment pattern is formed in the same layer as the electrode 108. In this case, the thickness of the conductive material that fills the electrode 108 is set to a thickness sufficient to fill the wide groove pattern. Therefore, when the depth of the wide groove pattern increases, the thickness of the conductive material increases when it is formed. In contrast, with the configuration according to this embodiment, the thickness of the bonding pad MBw in the WL region WR can be selectively increased without changing the depth of the wide groove pattern. This makes it possible to suppress an increase in the thickness of the conductive material that fills the electrode 108. Therefore, an increase in the manufacturing cost of the semiconductor memory device 1 can be suppressed.

[0149] Furthermore, with the configuration according to this embodiment, the ratio of the Z-direction length (Hmc2-Hmc1) of the contact plug VB to the Z-direction length Hmc2 of the electrode 108 in the cell region CR can be made 40% or less. The ratio of the Z-direction length (Hmw2-Hmw1) of the contact plug VB to the Z-direction length Hmw2 of the electrode 108 in the WL region WR can be made 35% or less. Also, in the circuit chip 20, the ratio of the Z-direction length (Hd2-Hd1) of the contact plug CB to the Z-direction length Hd2 of the electrode 210 can be made 40% or less. This can reduce disconnection failures of the contact plugs VB and CB due to tensile stress.

[0150] Furthermore, with the configuration according to this embodiment, the aspect ratio of the contact plugs VB and CB can be set to 1.5 or less, thereby reducing breakage of the contact plugs VB and CB due to tensile stress.

[0151] 2. Second embodiment Next, a second embodiment will be described. In the second embodiment, a method of forming a step between the cell region CR and the WL region WR, which is different from the first embodiment, will be described. The following mainly describes the points different from the first embodiment.

[0152] 2.1 Cross-sectional structure of semiconductor memory device First, the cross-sectional structure of the array chip 10 will be described with reference to Fig. 21. Fig. 21 is a cross-sectional view showing an example of the cross-sectional structure of the semiconductor memory device 1 taken along line IX-IX in Fig. 5. The semiconductor memory device 1 of this embodiment differs from the first embodiment in the structure of the dummy pillars HR.

[0153] 21, the dummy pillars HR of this embodiment are embedded in an insulator 132. The insulator 132 is, for example, silicon oxide. The other structures are similar to those of the first embodiment shown in FIG.

[0154] 2.2 Manufacturing method of memory cell array and electrodes Next, a method for manufacturing the memory cell array 11 and the electrodes 108 will be described with reference to Figs. 22 to 25. Figs. 22 to 25 are cross-sectional views showing the manufacturing process of the memory cell array 11 and the electrodes 108 of the array chip 10. The examples of Figs. 22 and 24 show an example of the cross-sectional structure of the array chip 10 cut along the line IIX-IIX in Fig. 5. The examples of Figs. 23 and 25 show an example of the cross-sectional structure of the array chip 10 cut along the line IX-IX in Fig. 5.

[0155] As shown in FIG. 22, similar to the description of FIG. 11 of the first embodiment, in the memory cell array region, after forming a staircase made of a plurality of insulating layers 122 and a plurality of sacrificial layers 131, an insulating layer 123 is deposited so as to cover the entire surface facing Z1 of the semiconductor substrate 100. Next, CMP of the insulating layer 123 is performed. In this embodiment, the surface of the insulating layer 123 is flattened. That is, no step is formed between the cell region CR and the WL region WR. Note that a step may be formed similarly to FIG. 11 of the first embodiment.

[0156] 23, a dummy pillar HR is formed in the WL region WR. More specifically, first, a dummy hole corresponding to the dummy pillar HR is formed. Next, the dummy hole is filled with an insulator 132.

[0157] 24 and 25, a memory pillar MP is formed in the cell region CR. At this time, a heat treatment is performed to crystallize the semiconductor film 143 of the memory pillar MP. This causes the insulating layer 123 and the insulator 132 to shrink in regions other than the cell region CR where the memory pillar MP is provided. As a result, a step having a height Ha is formed between the cell region CR and the WL region WR.

[0158] The subsequent steps are the same as those in the first embodiment.

[0159] 2.3 Effects of this embodiment With the configuration according to this embodiment, the same effects as those of the first embodiment can be obtained.

[0160] As a method for adjusting the difference in thickness between the bonding pad MBw and the bonding pad MBc, the first and second embodiments may be combined.

[0161] 3. Third embodiment Next, a third embodiment will be described. In the third embodiment, a method of forming a step between the cell region CR and the WL region WR, which is different from the first and second embodiments, will be described. The following mainly describes the differences from the first and second embodiments.

[0162] 3.1 Cross-sectional structure of semiconductor memory device First, the cross-sectional structure of the array chip 10 will be described with reference to Fig. 26. Fig. 26 is a cross-sectional view showing an example of the cross-sectional structure of the semiconductor memory device 1 taken along line IIX-IIX in Fig. 5. In the semiconductor memory device 1 of this embodiment, the film thickness of the M1 wiring layer 107 is different between the cell region CR and the WL region WR.

[0163] 26, the insulating layer 123 of this embodiment has no step between the cell region CR and the WL region WR. Therefore, the height positions of the conductor 104 (contact plug VY), the M0 wiring layer 105, and the conductor 106 (contact plug V0) are approximately the same in the cell region CR and the WL region WR.

[0164] In this embodiment, the film thickness (wiring layer thickness) of the M1 wiring layer 107 in the cell region CR is larger than that of the M1 wiring layer 107 in the WL region WR. Therefore, the height position of the surface of the M1 wiring layer 107 in the cell region CR facing the Z1 direction is lower than the height position of the surface of the M1 wiring layer 107 in the WL region WR facing the Z1 direction. This causes a step between the cell region CR and the WL region WR. The other structures are similar to those in FIG. 6 of the first embodiment.

[0165] 3.2 Cross-sectional structure of M1 wiring layer and bonding pad Next, an example of the cross-sectional structure of the M1 wiring layer 107 and the bonding pad BP will be described with reference to Fig. 27. Fig. 27 is a cross-sectional view showing an example of the cross-sectional structure of the bonding pad BP.

[0166] As shown in FIG. 27, the structures of the electrodes 108 and 210 are similar to those in FIG. 8 of the first embodiment.

[0167] In this embodiment, the height position of the surface of the M1 wiring layer 107 facing the Z2 direction is approximately the same in the cell region CR and the WL region WR. The length in the Z direction of the M1 wiring layer 107 in the cell region CR is Hc3. The length in the Z direction of the M1 wiring layer 107 in the WL region WR is Hw3. The length Hc3 and the length Hw3 have a relationship of Hc3>Hw3. The difference (Hmw1-Hmc1) between the Z direction length Hmc1 of the bonding pad MBw and the Z direction length Hmc1 of the bonding pad MBc is based on the difference (Hc3-Hw3) in the Z direction length (wiring layer thickness) of the M1 wiring layer 107. Therefore, the relationship between the bonding pad MB and the M1 wiring layer 107 is Hmw1-Hmc1=Hc3-Hw3.

[0168] 3.3 Manufacturing method of memory cell array and electrodes Next, a manufacturing method of the memory cell array 11 and the electrodes 108 will be described with reference to Figs. 28 to 33. Figs. 28 to 33 are cross-sectional views showing the manufacturing process of the memory cell array 11 and the electrodes 108 of the array chip 10. The examples of Figs. 28, 29, and 31 to 33 show an example of the cross-sectional structure of the array chip 10 cut along the line IIX-IIX in Fig. 5. The example of Fig. 30 shows an example of the cross-sectional structure of the array chip 10 cut along the line IX-IX in Fig. 5.

[0169] As shown in FIG. 28, similarly to FIG. 22 of the second embodiment, in the memory cell array region, after forming a staircase made of a plurality of insulating layers 122 and a plurality of sacrificial layers 131, an insulating layer 123 is deposited so as to cover the entire surface facing Z1 of the semiconductor substrate 100. Next, CMP of the insulating layer 123 is performed. In this embodiment, the surface of the insulating layer 123 is flattened. That is, no step is formed between the cell region CR and the WL region WR. Note that a step may be formed similarly to FIG. 11 of the first embodiment.

[0170] 29 and 30, memory pillars MP and dummy pillars HR are formed in the same manner as in the first embodiment described in Fig. 12 and Fig. 13. Unlike the first embodiment, no step is formed between the cell region CR and the WL region WR.

[0171] As shown in FIG. 31, conductors 103 and 109, conductor 104, M0 wiring layer 105, and conductor 106 are formed. Next, M1 wiring layer 107 is formed. At this time, the coverage of M1 wiring layer 107 in WL region WR is made smaller than that of cell region CR. The coverage is the ratio of the area of ​​a pattern arranged in a certain region to the area. More specifically, a pattern corresponding to M1 wiring layer 107 is first formed. When conductor 106 and M1 wiring layer 107 are formed collectively by the dual damascene method, a pattern corresponding to conductor 106 and M1 wiring layer 107 is formed. Next, a conductive material is deposited, and the above-mentioned pattern is embedded with the conductive material. At this stage, no step is formed between cell region CR and WL region WR.

[0172] As shown in Fig. 32, excess conductive material on the insulating layer 124 is removed by CMP. At this time, erosion occurs in the WL region WR, which has a low coverage rate, and the entire surface including the insulating layer 124 is polished. As a result, a step having a height Hb occurs between the cell region CR and the WL region WR. In addition, the M1 wiring layer 107 is formed with different film thicknesses in the cell region CR and the WL region WR.

[0173] 33, an insulating layer 124 and an insulating layer 125 are formed. Next, the insulating layer 125 is processed to form a mask pattern for forming the contact plugs VB. The subsequent steps are the same as those in the first embodiment.

[0174] In this embodiment, the case where the thickness of the M1 wiring layer 107 is different between the cell region CR and the WL region WR has been described, but the present invention is not limited to this. For example, the thickness of the M0 wiring layer 105 may be different between the cell region CR and the WL region WR, or the thicknesses of both the M0 wiring layer 105 and the M1 wiring layer 107 may be different.

[0175] 3.4 Effects of this embodiment With the configuration according to this embodiment, the same effects as those of the first embodiment can be obtained.

[0176] As a method for adjusting the difference in thickness between the bonding pad MBw and the bonding pad MBc, the first to third embodiments may be combined.

[0177] 4. Modifications, etc. The semiconductor memory device according to the embodiment includes a first chip (10) including a plurality of first wiring layers (WL) stacked at a distance in a first direction (Z direction), a memory pillar (MP) extending in the first direction and passing through the plurality of first wiring layers, a second wiring layer (BL) electrically connected to the memory pillar, a first electrode (108 in the WL region) electrically connected to one of the plurality of first wiring layers, and a second electrode (108 in the cell region) electrically connected to the second wiring layer, and a second chip (20) including a third electrode (210 in the WL region) bonded to the first electrode, and a fourth electrode (210 in the cell region) bonded to the second electrode. The length (Hmw2) of the first electrode in the first direction is greater than the length (Hmc2) of the second electrode in the first direction.

[0178] By applying the above embodiment, bonding defects in the semiconductor memory device 1 can be reduced.

[0179] The embodiment is not limited to the above-described embodiment, and various modifications are possible.

[0180] In the above embodiment, "connected" includes a state in which something else, such as a transistor or a resistor, is interposed between them and indirectly connected.

[0181] The embodiments are illustrative and the scope of the invention is not limited thereto. [Explanation of symbols]

[0182] 1...semiconductor memory device, 10...array chip, 11...memory cell array, 20...circuit chip, 21...sequencer, 22...voltage generating circuit, 23...row decoder, 24...sense amplifier, 100, 201...semiconductor substrate, 101, 101a to 101c...semiconductor layer, 102...wiring layer, 103, 104, 106, 108_2, 109, 204, 206, 208, 210_2...conductor, 105...M0 wiring layer, 107...M1 wiring layer, 108, 210...electrode, 108_1, 210_1...barrier metal, 121 to 126, 211 to 215...insulating layer, 130, 131...sacrificial layer, 132...insulator, 140...block insulating film, 141...charge storage film, 142...ton 143: semiconductor film, 144: core film, 145: cap film, 201: semiconductor substrate, 202: gate insulating film, 203: gate electrode, 205: D0 wiring layer, 207: D1 wiring layer, 209: D2 wiring layer, BLK0 to BLK3: blocks, BP, MB, MBc, MBw, DB, DBc, DBw: bonding pads, C0, C1, CB, CC, CH, CS, V0, VB, VY: contact plugs, CR: cell region, MC0 to MC7: memory cell transistors, SGD0 to SGD4: select gate lines, ST1, ST2: select transistors, SU0 to SU4: string units, WL0 to WL7: word lines, WR, WR1, WR2: WL region

Claims

1. a first chip including a plurality of first wiring layers stacked and spaced apart in a first direction, a memory pillar extending in the first direction and passing through the plurality of first wiring layers, a second wiring layer electrically connected to the memory pillar, a first electrode electrically connected to any one of the plurality of first wiring layers, and a second electrode electrically connected to the second wiring layer; a second chip including a third electrode bonded to the first electrode and a fourth electrode bonded to the second electrode; Equipped with The length of the first electrode in the first direction is greater than the length of the second electrode in the first direction. Semiconductor memory device.

2. the first electrode includes a first pad having a lower surface in contact with the third electrode, and a first contact plug in contact with an upper surface of the first pad opposite to the lower surface, the second electrode includes a second pad having a lower surface in contact with the fourth electrode, and a second contact plug in contact with an upper surface of the second pad opposite to the lower surface, A length of the first pad in the first direction is greater than a length of the second pad in the first direction.

2. The semiconductor memory device according to claim 1.

3. a ratio of a length of the second contact plug in the first direction to the length of the second electrode is 40% or less; 3. The semiconductor memory device according to claim 2.

4. a ratio of a length of the first contact plug in the first direction to the length of the first electrode is 35% or less; 3. The semiconductor memory device according to claim 2.

5. an aspect ratio of each of the first contact plug and the second contact plug is 1.5 or less; 3. The semiconductor memory device according to claim 2.

Citation Information

Patent Citations

  • Semiconductor device and method for manufacturing the same

    JP2021136271A

  • Semiconductor device and method for manufacturing the same

    JP2022095359A