3D memory device
By using a pseudo-channel column to connect to the heat dissipation structure in the 3D memory device, the problem of lack of heat dissipation path between the CMOS circuit and the memory cell array in the prior art is solved, effective heat release is achieved, and the device's goodness is improved rate and reliability, and reduce manufacturing costs.
Patent Information
- Application Number
- CN202410980886.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2018-09-28
- Publication Date
- 2024-11-08
AI Technical Summary
Existing 3D memory devices lack heat dissipation paths between CMOS circuits and memory cell arrays, resulting in excessive temperatures, affecting yield and reliability, especially when writing data can lead to errors or device damage.
The pseudo-channel column is used to connect to the heat dissipation structure, providing heat conduction channels and heat dissipation structure, ensuring that heat can be effectively released during the operation of the storage cell array and CMOS circuit, reducing temperature, and improving the yield and reliability of the device.
Through the design of thermal conductivity channels and heat dissipation structure, the working temperature of the memory device is effectively reduced, write data errors and device damage are avoided, yield and reliability of 3D memory devices are improved, and manufacturing costs are reduced.
Smart Images

Figure CN118921987A_ABST
Abstract
Description
[0001] This application is a divisional application of the Chinese invention patent application with an application date of September 28, 2018, application number 201811139608.7, and application name “3D storage device”. Technical Field
[0002] The present invention relates to the field of memory technology, and more particularly, to a 3D memory device. Background Art
[0003] The improvement of storage density of memory devices is closely related to the progress of semiconductor manufacturing process. As the feature size of semiconductor manufacturing process becomes smaller and smaller, the storage density of memory devices becomes higher and higher. In order to further improve the storage density, a three-dimensional memory device (i.e., 3D memory device) has been developed. 3D memory devices include multiple memory cells stacked in a vertical direction, which can multiply the integration on a unit area of a wafer and reduce costs.
[0004] Existing 3D memory devices are mainly used as non-volatile flash memory. The two main non-volatile flash memory technologies use NAND and NOR structures. Compared with NOR memory devices, NAND memory devices have a slightly slower read speed, but a faster write speed, simpler erase operation, and can achieve smaller storage cells, thereby achieving higher storage density. Therefore, 3D memory devices using NAND structures have been widely used.
[0005] In a 3D memory device with a NAND structure, a gate stack structure is used to form a memory cell array. In the 3D memory device, a large number of metal wirings are used to provide electrical connections between the memory cell array and an external circuit. The increase in wiring density will affect the yield and reliability of the 3D memory device. It is expected to further improve the structure of the 3D memory device and the manufacturing method thereof to improve the yield and reliability of the 3D memory device. Summary of the invention
[0006] In view of the above problems, an object of the present invention is to provide a 3D memory device, wherein the present invention uses a dummy channel pillar connected to a heat dissipation structure to provide a heat dissipation path, thereby improving the yield and reliability of the 3D memory device.
[0007] According to one aspect of the present invention, a 3D memory device is provided, comprising: a gate stack structure, the gate stack structure comprising a plurality of gate conductors and a plurality of interlayer insulating layers alternately stacked; a plurality of channel pillars, the channel pillars penetrating the gate stack structure to form transistors; and a plurality of dummy channel pillars, the dummy channel pillars penetrating at least part of the gate conductors in the gate stack structure to provide support, wherein at least one of the plurality of dummy channel pillars is connected to a heat dissipation structure.
[0008] Preferably, the dummy channel pillar comprises a heat dissipation material.
[0009] Preferably, it also includes: a first semiconductor substrate, a first surface of the first semiconductor substrate is adjacent to the gate stack structure; a second insulating layer located on the second surface of the first semiconductor substrate, the second surface and the first surface of the first semiconductor substrate are opposite to each other; and a first insulating layer covering the gate stack structure.
[0010] Preferably, the heat dissipation structure is located on the first insulating layer.
[0011] Preferably, the heat dissipation structure is located on the second insulating layer.
[0012] Preferably, it further comprises: an insulating liner at least partially surrounding the dummy channel column, for isolating the dummy channel column from the gate stack structure and the first semiconductor substrate.
[0013] Preferably, it further comprises: a plurality of wiring layers located in the first insulating layer, a first end of the dummy channel column being connected to a corresponding wiring layer of the plurality of wiring layers, and a second end being connected to the heat dissipation structure.
[0014] Preferably, it also includes: a groove extending laterally on the surface of the first insulating layer, the groove extending from the first side wall of the gate stack structure to the second side wall, the first ends of the plurality of pseudo channel columns extending to the groove, and the second ends connected to the heat dissipation structure.
[0015] Preferably, it also includes: a heat conducting strip located in the groove.
[0016] Preferably, it further comprises: a CMOS circuit, which is bonded to the gate stack structure via a connection structure, and comprises: a second semiconductor substrate; a transistor located in the second semiconductor substrate; and a third insulating layer located on the second semiconductor substrate.
[0017] Preferably, the gate stack structure serves as a memory cell array, the surface of the first insulating layer in the memory cell array serves as a first bonding surface, the gate stack structure also includes a first external pad located on the first bonding surface, the surface of the second insulating layer in the CMOS circuit serves as a second bonding surface, and the CMOS circuit also includes a second external pad located on the second bonding surface, wherein the first bonding surface of the gate stack structure and the second bonding surface of the CMOS circuit are in contact with each other, and the first external pad and the second external pad are bonded to each other, thereby realizing electrical connection between the gate stack structure and the CMOS circuit.
[0018] According to the 3D memory device of this embodiment, a gate stack structure is used to form a memory cell array, a heat conduction channel is provided inside the memory cell array, the heat conduction channel includes a dummy channel column, and a heat dissipation structure connected to the heat conduction channel is provided on the free surface of the memory cell array. A large amount of heat is generated during the operation of the memory cell array, which is released through the heat conduction channel and the heat dissipation structure, so that the operating temperature of the 3D memory device can be kept in compliance with the requirements, so that when the 3D memory device writes data using the tunneling principle, no write data error occurs, and damage to the device caused by excessively high temperature is avoided. Therefore, the 3D memory device according to this embodiment can improve the yield and reliability.
[0019] Furthermore, a CMOS circuit is formed using a semiconductor substrate, and then the CMOS circuit and the memory cell array are bonded to each other. A large amount of heat generated at the bonding surface during the operation of the CMOS circuit and the memory cell array can also be released through the heat conduction channel and the heat dissipation structure to avoid damage to the device due to excessive temperature. Therefore, the 3D memory device according to this embodiment can further improve the yield and reliability.
[0020] Furthermore, the heat conduction channel in the 3D memory device is a pseudo channel column and / or a pseudo conductive channel, and at least a portion of the heat conduction channel is formed simultaneously with the channel column and / or the conductive channel, which can reduce additional process steps and the number of masks. Therefore, the 3D memory device according to this embodiment can reduce manufacturing costs. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] The above and other objects, features and advantages of the present invention will become more apparent through the following description of the embodiments of the present invention with reference to the accompanying drawings, in which:
[0022] Figure 1a and 1b An equivalent circuit diagram and a structural schematic diagram of a memory cell string of a 3D memory device are shown respectively.
[0023] Figure 2a and 2b A perspective view and an overall perspective view respectively illustrate an internal structure of a 3D memory device according to a first embodiment of the present invention.
[0024] Figure 3 A cross-sectional view of a 3D memory device according to a first embodiment of the present invention is shown.
[0025] Figure 4 A cross-sectional view of a 3D memory device according to a second embodiment of the present invention is shown.
[0026] Figure 5 FIG. 4 is a cross-sectional view of a 3D memory device according to a third embodiment of the present invention.
[0027] Figures 6a to 6fCross-sectional views showing various stages of a method for manufacturing a 3D memory device according to a first embodiment of the present invention. DETAILED DESCRIPTION
[0028] The present invention will be described in more detail below with reference to the accompanying drawings. In each of the accompanying drawings, the same elements are represented by similar reference numerals. For the sake of clarity, the various parts in the accompanying drawings are not drawn to scale. In addition, some well-known parts may not be shown. For the sake of simplicity, the semiconductor structure obtained after several steps can be described in one figure.
[0029] It should be understood that when describing the structure of a device, when a layer or a region is referred to as being "on" or "over" another layer or another region, it may mean that it is directly on the other layer or another region, or that other layers or regions are included between it and the other layer or another region. Furthermore, if the device is turned over, the layer or a region will be "below" or "beneath" another layer or another region.
[0030] If the purpose is to describe the situation of being directly on another layer or another area, this article will use the expression "directly on..." or "on... and adjacent to...".
[0031] In this application, the term "semiconductor structure" refers to the entire semiconductor structure formed in various steps of manufacturing a memory device, including all layers or regions that have been formed. The term "free surface" refers to the surface of the CMOS circuit and the memory cell array that is opposite to the contact surface between the two, and the term "bonding surface" refers to the surface of the CMOS circuit and the memory cell array that are in contact with each other, and the CMOS circuit and the memory cell array are mechanically and electrically connected using the bonding surface of the two.
[0032] Many specific details of the present invention are described below, such as device structure, materials, dimensions, processing technology and techniques, so as to more clearly understand the present invention. However, as those skilled in the art will appreciate, the present invention may be implemented without following these specific details.
[0033] In a 3D memory device with a NAND structure, a memory cell array is formed using a gate stack structure, which provides gate conductors for selection transistors and memory transistors, and a large number of metal wirings are used to provide electrical connections between the memory cell array and external circuits. Furthermore, a CMOS circuit is formed using a semiconductor substrate, and then the CMOS circuit and the memory cell array are bonded to each other. Both the CMOS circuit and the memory cell array include a wiring layer, in which a large number of metal wirings are used to provide electrical connections between the CMOS circuit and the memory cell array.
[0034] The inventor of the present application has found that a large amount of heat is generated during the operation of the memory cell array. The accumulation of this heat causes the temperature of the 3D memory device to be too high. The 3D memory device uses the tunneling principle to write data. Therefore, the 3D memory device is sensitive to the ambient temperature, and too high a temperature may cause a write data error. In more serious cases, too high a temperature may cause the electrical connection between the CMOS circuit and the memory cell array to be disconnected, resulting in device damage. And during the operation of the CMOS circuit, since the relative surfaces of the CMOS circuit and the memory cell array are bonded to each other, a large number of wirings of the CMOS circuit and the memory cell array are located near the bonding surface, so that the heat is concentrated here and cannot be released, and the temperature of the 3D memory device will further increase. The existing 3D memory device has no heat dissipation path between the CMOS circuit and the memory cell array, thereby affecting the yield and reliability of the 3D memory device.
[0035] The inventors of the present application have noticed the above problems affecting the yield and reliability of 3D memory devices, and thus proposed a further improved 3D memory device.
[0036] The present invention may be embodied in various forms, some examples of which will be described below.
[0037] Figure 1a and 1b 1 and 2 respectively show a circuit diagram and a schematic diagram of a storage cell string of a 3D storage device. The storage cell string shown in this embodiment includes 4 storage cells. It can be understood that the present invention is not limited to this, and the number of storage cells in the storage cell string can be any number, for example, 32 or 64.
[0038] like Figure 1a As shown, the first end of the memory cell string 100 is connected to the bit line BL, and the second end is connected to the source line SL. The memory cell string 100 includes a plurality of transistors connected in series between the first end and the second end, including: a first selection transistor Q1, storage transistors M1 to M4, and a second selection transistor Q2. The gate of the first selection transistor Q1 is connected to the string selection line SSL, and the gate of the second selection transistor Q2 is connected to the ground selection line GSL. The gates of the storage transistors M1 to M4 are respectively connected to the corresponding word lines of the word lines WL1 to WL4.
[0039] like Figure 1bAs shown, the first selection transistor Q1 and the second selection transistor Q2 of the memory cell string 100 include gate conductors 122 and 123, respectively, and the memory transistors M1 to M4 include gate conductors 121, respectively. The gate conductors 121, 122 and 123 are consistent with the stacking order of the transistors in the memory cell string 100, and the adjacent gate conductors are separated from each other by an interlayer insulating layer, thereby forming a gate stack structure. Further, the memory cell string 100 includes a channel column 110. The channel column 110 runs through the gate stack structure. In the middle part of the channel column 110, a tunneling dielectric layer 112, a charge storage layer 113 and a blocking dielectric layer 114 are sandwiched between the gate conductor 121 and the channel layer 111, thereby forming the memory transistors M1 to M4. At both ends of the channel column 110, a blocking dielectric layer 114 is sandwiched between the gate conductors 122 and 123 and the channel layer 111, thereby forming the first selection transistor Q1 and the second selection transistor Q2.
[0040] In this embodiment, the channel layer 111 is composed of, for example, doped polysilicon, the tunnel dielectric layer 112 and the blocking dielectric layer 114 are respectively composed of oxides, such as silicon oxide, the charge storage layer 113 is composed of an insulating layer containing quantum dots or nanocrystals, such as silicon nitride containing metal or semiconductor particles, and the gate conductors 121, 122 and 123 are composed of metals, such as tungsten. The channel layer 111 is used to provide a channel region for controlling the selection transistor and the storage transistor, and the doping type of the channel layer 111 is the same as that of the selection transistor and the storage transistor. For example, for N-type selection transistors and storage transistors, the channel layer 111 can be N-type doped polysilicon.
[0041] In this embodiment, the core of the channel pillar 110 is the channel layer 111, and the tunnel dielectric layer 112, the charge storage layer 113 and the blocking dielectric layer 114 form a stacked structure around the sidewalls of the core. In an alternative embodiment, the core of the channel pillar 110 is an additional insulating layer, and the channel layer 111, the tunnel dielectric layer 112, the charge storage layer 113 and the blocking dielectric layer 114 form a stacked structure around the core.
[0042] In this embodiment, the first selection transistor Q1 and the second selection transistor Q2, and the memory transistors M1 to M4 use a common channel layer 111 and a blocking dielectric layer 114. In the channel pillar 110, the channel layer 111 provides source and drain regions and channel layers of multiple transistors. In an alternative embodiment, the semiconductor layer and the blocking dielectric layer of the first selection transistor Q1 and the second selection transistor Q2 and the semiconductor layer and the blocking dielectric layer of the memory transistors M1 to M4 may be formed separately in independent steps.
[0043] In the write operation, the memory cell string 100 uses the FN tunneling efficiency to write data into the selected memory transistors among the memory transistors M1 to M4. Taking the memory transistor M2 as an example, when the source line SL is grounded, the ground selection line GSL is biased to a voltage of approximately zero volts, so that the selection transistor Q2 corresponding to the ground selection line GSL is disconnected, and the string selection line SSL is biased to a high voltage VDD, so that the selection transistor Q1 corresponding to the string selection line SSL is turned on. Further, the bit line BIT2 is grounded, the word line WL2 is biased to a programming voltage VPG, for example, about 20V, and the remaining word lines are biased to a low voltage VPS1. Since only the word line voltage of the selected memory transistor M2 is higher than the tunneling voltage, the electrons in the channel region of the memory transistor M2 reach the charge storage layer 113 via the tunneling dielectric layer 112, thereby converting the data into charge stored in the charge storage layer 113 of the memory transistor M2.
[0044] In the read operation, the memory cell string 100 determines the amount of charge in the charge storage layer according to the conduction state of the selected memory transistors among the memory transistors M1 to M4, thereby obtaining data represented by the amount of charge. Taking the memory transistor M2 as an example, the word line WL2 is biased at the read voltage VRD, and the remaining word lines are biased at the high voltage VPS2. The conduction state of the memory transistor M2 is related to its threshold voltage, that is, it is related to the amount of charge in the charge storage layer, so the data value can be determined according to the conduction state of the memory transistor M2. The memory transistors M1, M3 and M4 are always in the conduction state, so the conduction state of the memory cell string 100 depends on the conduction state of the memory transistor M2. The control circuit determines the conduction state of the memory transistor M2 according to the electrical signals detected on the bit line BL and the source line SL, thereby obtaining the data stored in the memory transistor M2.
[0045] Figure 2a and 2b A perspective view and an overall perspective view respectively showing an internal structure of a 3D memory device according to a first embodiment of the present invention, Figure 3 A cross-sectional view of a 3D memory device according to a first embodiment of the present invention is shown, wherein the cross-sectional view is taken along Figure 2a Intercepted along line AA.
[0046] For clarity, in Figure 2a Only the internal structure of the 3D memory device is shown, wherein the semiconductor substrate of the memory cell array, and the insulating layer in the CMOS circuit and the memory cell array are not shown. Figure 2b Only the external structure of the 3D memory is shown.
[0047] The 3D memory device 200 shown in this embodiment includes a stacked CMOS circuit 210 and a memory cell array 220 .
[0048] The CMOS circuit 210 includes a semiconductor substrate 201, a plurality of contact pads 261 on the semiconductor substrate 201, a plurality of wiring layers 263 on the plurality of contact pads 261, a plurality of external pads 264 on the plurality of wiring layers 263, and a conductive channel 262 providing interconnection in a direction perpendicular to the surface of the semiconductor substrate 201. Although not shown, it is understood that a plurality of transistors are formed in the semiconductor substrate 201. The plurality of wiring layers 260 are separated from each other, and the plurality of wiring layers 260 and the contact pads 261 and the external pads 264 are separated from each other by an interlayer insulating layer, and are electrically connected to each other by a conductive channel 262 penetrating the interlayer insulating layer. Figure 2a The interlayer insulating layer is not shown.
[0049] In the CMOS circuit 210, the contact pad 261 is electrically connected to the transistor in the semiconductor substrate 201, and the contact pad 261 is connected to the wiring layer 263 via the conductive channel 262, and then the wiring layer 263 is connected from the external pad 264 via the conductive channel 262. The external pad 264 provides electrical connection between the transistor inside the CMOS circuit 210 and the memory cell array 220.
[0050] The memory cell array 220 includes 2*3 memory cell strings, which are 6 in total, and each memory cell string includes 4 memory cells, thereby forming a memory array of 4*2*3 memory cells, which are 24 in total. It can be understood that the present invention is not limited thereto, and the 3D memory device can include any number of memory cell strings, for example, 1024, and the number of memory cells in each memory cell string can be any number, for example, 32 or 64.
[0051] The memory cell array 220 includes a semiconductor substrate 101, a gate stack structure located on the semiconductor substrate 101, a channel pillar 110 penetrating the gate stack structure, and an interconnection structure located on the gate stack structure. The interconnection structure includes a plurality of conductive channels 161, a plurality of contact pads 162 respectively contacting the plurality of conductive channels 161, a plurality of wiring layers 164 located on the plurality of contact pads 162, a plurality of external pads 165 located on the plurality of wiring layers 164, and a conductive channel 163 providing interconnection in a direction perpendicular to the surface of the semiconductor substrate 101. The gate stack structure includes, for example, gate conductors 121, 122, and 123. The plurality of gate conductors in the gate stack structure are, for example, formed in a step shape to provide a space for the conductive channel 161 to extend to the corresponding gate conductor.
[0052] In the memory cell array 220, the memory cell strings each include their own channel pillars 110, and common gate conductors 121, 122, and 123. The gate conductors 121, 122, and 123 are in the same stacking order as the transistors in the memory cell string 100, and adjacent gate conductors are separated from each other by an interlayer insulating layer, thereby forming a gate stack structure. Figure 2a The interlayer insulating layer is not shown.
[0053] In this embodiment, the internal structure of the channel pillar 110 is as follows: Figure 1b As shown, no further detailed description is given here. The channel pillars 110 penetrate the gate stack structure and are arranged in an array. The semiconductor substrate is located above the gate stack structure, in which a common source region (not shown in the figure) is formed. The first ends of the channel pillars 110 are commonly connected to the common source region, and the second ends of the channel pillars 110 are connected to the corresponding external pads 165 via conductive channels and wiring. The functions of the conductive channels and wiring layers here are the same as those of the bit lines BL.
[0054] The gate conductor 122 of the first selection transistor Q1 is divided into different gate lines by a gate line slit 151. The gate lines of the plurality of channel pillars 110 in the same row are connected to the corresponding external pads 165 via conductive channels and wirings, respectively. For the sake of clarity, a portion of the conductive channels and wiring layers between the gate conductor 122 and the contact pads are not shown in the figure. The functions of the conductive channels and wiring layers here are the same as those of the string selection line SSL.
[0055] The gate conductors 121 of the storage transistors M1 and M4 are connected to the corresponding word lines, respectively. If the gate conductors 121 of the storage transistors M1 and M4 are divided into different gate lines by the gate line gaps 151, the gate lines at the same level are connected to the corresponding external pads 165 via conductive channels and wirings, respectively. For the sake of clarity, a portion of the conductive channels and wiring layers between the gate conductors 121 and the contact pads are not shown in the figure. The functions of the conductive channels and wiring layers here are the same as those of the word lines WL1 to WL4.
[0056] The gate conductors of the second selection transistor Q2 are connected as one. If the gate conductor 123 of the second selection transistor Q2 is divided into different gate lines by the gate line gap 151, the gate lines are respectively connected to the corresponding external pads 165 via conductive channels and wiring. The functions of the conductive channels and wiring layers here are the same as those of the ground selection line GSL.
[0057] A gate stack structure and an interlayer insulating layer 104 are respectively disposed on the first and second surfaces of the semiconductor substrate 101, and a heat dissipation structure 105 is formed on the interlayer insulating layer 104. The heat dissipation structure 105 is made of metal or resin with good thermal conductivity, such as copper, silver, aluminum thermal conductive silicone.
[0058] In this embodiment, a heat conduction channel 143 and an insulating liner 144 are formed inside the memory cell array 220. The heat conduction channel 143 is composed of a metal or resin with good thermal conductivity, such as copper, silver, and aluminum thermal conductive silicone. The heat conduction channel 143 passes through the gate stack structure, the semiconductor substrate 101, and the interlayer insulating layer 104, and is insulated from the gate stack structure and the semiconductor substrate 101 by the insulating liner 142. The first end of the heat conduction channel 143 extends to the wiring layer 164 inside the memory cell array 220, and the second end extends to the free surface of the memory cell array 220, that is, the surface opposite to the bonding surface, and is connected to the heat dissipation structure 105, thereby providing a heat dissipation path.
[0059] Preferably, in this embodiment, a dummy channel column 130 may be further included, and the internal structure of the dummy channel column 130 may be the same as that of the channel column 110, and at least pass through at least a portion of the gate conductor in the gate stack structure. However, the dummy channel column 130 is not connected to the external pad 165, so that it only provides mechanical support and is not used to form a selection transistor and a storage transistor. Therefore, the dummy channel column 130 does not form an effective storage unit.
[0060] Preferably, in this embodiment, a conductive channel 141 and an insulating liner 142 may also be included. The conductive channel 141 is used as a part of a through contact via (TAC) to provide a conductive path between the CMOS circuit and the external circuit. The conductive channel 141 passes through the gate stack structure, the semiconductor substrate 101 and the interlayer insulating layer 104, and is insulated from the gate stack structure and the semiconductor substrate 101 by the insulating liner 142. The first end and the second end of the conductive channel 141 extend to two opposite surfaces (i.e., the free surface and the bonding surface) of the memory cell array 220, respectively, to form an external pad.
[0061] Preferably, in this embodiment, an additional conductive channel and an additional insulating liner (not shown in the figure) may also be included. The additional conductive channel, as part of the source line GL, is used to provide an electrical connection between the common source in the semiconductor substrate 101 and the external circuit. In this preferred embodiment, the additional conductive channel passes through the gate stack structure, and the gate stack structure and the semiconductor substrate 101 are insulated from each other by the additional insulating liner. The first end of the additional conductive channel reaches the common source region, and the second end is connected to the wiring layer.
[0062] After forming the CMOS circuit 210 and the memory cell array 220, the two are bonded to each other to form the 3D memory device 200. The surfaces of the CMOS circuit 210 and the memory cell array 220 facing each other are their respective bonding surfaces. A large number of wirings of the CMOS circuit 210 and the memory cell array 220 are located near their respective bonding surfaces.
[0063] See also Figure 2b According to the 3D memory device 200 of this embodiment, the conductive channels and wiring layers of the CMOS circuit 210 are located in at least one insulating layer 202, and the conductive channels and wiring layers of the memory cell array 220 are located in at least one insulating layer 102. The bonding surfaces of the CMOS circuit 210 and the memory cell array 220 are surfaces of the insulating layers 202 and 102 that are opposite to each other. Further, the external pads 264 of the CMOS circuit 210 and the external pads 165 of the memory cell array 220 are exposed on the corresponding bonding surfaces, respectively, and are arranged opposite to each other. Therefore, when the CMOS circuit 210 and the memory cell array 220 are bonded to each other to form the 3D memory device 200, the external pads 264 of the CMOS circuit 210 and the external pads 165 of the memory cell array 220 are in contact with each other, thereby realizing the electrical connection between the CMOS circuit 210 and the memory cell array 220.
[0064] Further, according to the 3D memory device 200 of this embodiment, a heat conduction channel 143 is formed inside the memory cell array 220, and a heat dissipation structure 105 is formed on the free surface of the memory cell array 220. The first end of the heat conduction channel 143 reaches the wiring layer of the memory cell array 220, and the second end is connected to the heat dissipation structure 105, thereby forming a heat dissipation path. The heat dissipation structure 105 extends laterally on the free surface of the memory cell array 220, for example, having a substantially rectangular shape. In the 3D memory device 200 formed by bonding the CMOS circuit 210 and the memory cell array 220 to each other, the heat conduction channel 143 reaches the wiring layer, and the heat dissipation structure 105 is exposed to the external environment, thereby conducting heat from the wiring layer inside the 3D memory device to the outside, thereby utilizing the heat dissipation structure on one side surface of the 3D memory device to achieve heat release.
[0065] In an alternative embodiment, a heat conduction channel is formed inside the CMOS circuit 210, and a heat dissipation structure is formed on the free surface of the CMOS circuit 210. The free surface is, for example, the free surface of the interlayer insulating layer formed on the semiconductor substrate 201. The first end of the heat conduction channel reaches the wiring layer of the CMOS circuit 210, and the second end is connected to the heat dissipation structure, thereby forming a heat dissipation path. This alternative embodiment uses the heat dissipation structure on the other side surface of the 3D memory device to achieve heat release.
[0066] In a preferred embodiment, conductive channels are formed inside the CMOS circuit 210 and the memory cell array 220, and heat dissipation structures are formed on the free surfaces of the CMOS circuit 210 and the memory cell array 220, so as to achieve heat release by using the heat dissipation structures on the opposite surfaces of the 3D memory device.
[0067] In a preferred embodiment, a transverse through groove is formed on the bonding surface of at least one of the CMOS circuit 210 and the memory cell array 220 to provide an additional heat dissipation path, thereby utilizing the through groove to achieve heat release.
[0068] Figure 4 A cross-sectional view of a 3D memory device according to a second embodiment of the present invention is shown, wherein the cross-sectional view is taken along Figure 2a Intercepted along line AA.
[0069] The 3D memory device 300 shown in this embodiment includes a stacked CMOS circuit 210 and a memory cell array 220. Only the differences between the second embodiment and the first embodiment are described in detail below.
[0070] In this embodiment, a plurality of heat conduction channels 143 and a plurality of insulating liners 144 are formed in the memory cell array 220. The plurality of heat conduction channels 143 respectively pass through the gate stack structure, the semiconductor substrate 101 and the interlayer insulating layer 104, and are insulated from each other by the corresponding insulating liners 142 between the gate stack structure and the semiconductor substrate 101. The first end of the first group of heat conduction channels of the plurality of heat conduction channels 143 extends to the wiring layer 163, and the first end of the second group of heat conduction channels reaches the bonding surface of the memory cell array 220. The second ends of the plurality of heat conduction channels 143 all extend to the free surface of the memory cell array 220, that is, the surface opposite to the bonding surface, and are connected to the heat dissipation structure 105, thereby providing a heat dissipation path.
[0071] According to the 3D memory device of this embodiment, a heat conduction channel is provided inside the memory cell array, and a heat dissipation structure connected to the heat conduction channel is provided on the surface of the memory cell array. The heat conduction channel extends to the wiring layer and the bonding surface, thereby providing a heat dissipation path. A large amount of heat is generated during the operation of the CMOS circuit and the operation of the memory cell array, and is released via the heat dissipation path. The release of this heat can keep the operating temperature of the 3D memory device in compliance with the requirements, so that when the 3D memory device writes data using the tunneling principle, no write data error occurs, and excessive temperature is avoided to cause device damage. Therefore, the 3D memory device according to this embodiment improves yield and reliability.
[0072] Figure 5 3D memory device according to the third embodiment of the present invention are shown in cross-sectional view, the cross-sectional view is taken along Figure 2a Intercepted along line AA.
[0073] The 3D memory device 400 shown in this embodiment includes a stacked CMOS circuit 210 and a memory cell array 220. Only the differences between the third embodiment and the first embodiment are described in detail below.
[0074] In this embodiment, a groove is formed on the bonding surface of the memory cell array 220. The heat conducting bar 413 is located in the groove and extends laterally on the bonding surface of the memory cell array 220. Preferably, the heat conducting bar 413 passes through two opposite sides of the memory cell array 220.
[0075] Further, a plurality of heat-conducting channels 143 and a plurality of insulating liners 144 are formed in the memory cell array 220. The plurality of heat-conducting channels 143 respectively pass through the gate stack structure, the semiconductor substrate 101 and the interlayer insulating layer 104, and are insulated from each other by the corresponding insulating liners 142 between the gate stack structure and the semiconductor substrate 101. The first end of the first group of heat-conducting channels of the plurality of heat-conducting channels 143 extends to the wiring layer 163, and the first end of the second group of heat-conducting channels reaches the heat-conducting bar 413 in the bonding surface of the memory cell array 220. The second ends of the plurality of heat-conducting channels 143 all extend to the free surface of the memory cell array 220, that is, the surface opposite to the bonding surface, and are connected to the heat dissipation structure 105, thereby providing a heat dissipation path.
[0076] According to the 3D memory device of this embodiment, a heat conduction channel is provided inside the memory cell array, and a heat dissipation structure connected to the heat conduction channel is provided on the free surface of the memory cell array, and a heat conduction bar connected to the heat conduction channel is provided on the bonding surface of the memory cell array. The heat conduction channel extends to the wiring layer and the heat conduction bar, thereby providing a heat dissipation path. A large amount of heat is generated during the operation of the CMOS circuit and the operation of the memory cell array. Since the relative surfaces of the CMOS circuit and the memory cell array are bonded to each other, the heat is concentrated near the bonding surface and released via the heat dissipation path. The release of this heat can keep the operating temperature of the 3D memory device in compliance with the requirements, so that when the 3D memory device uses the tunneling principle to write data, no write data error will occur, and excessive temperature will be avoided to cause device damage. Therefore, the 3D memory device according to this embodiment improves yield and reliability.
[0077] Figures 6a to 6f 1 is a cross-sectional view showing various stages of a method for manufacturing a 3D memory device according to a first embodiment of the present invention, wherein Figures 6a to 6d shows the steps of manufacturing a memory cell array, Figure 6e shows the manufacturing steps of CMOS circuits, Figure 6f The cross-sectional view shows the bonding between CMOS and memory cell array. Figure 2a Intercepted along line AA.
[0078] The method starts with a semiconductor structure in which a plurality of well regions have been formed on a semiconductor substrate 101 . In this embodiment, the semiconductor substrate 101 is, for example, a single crystal silicon substrate.
[0079] In this embodiment, in order to facilitate the programming operation of the memory cells in the 3D memory device, a plurality of well regions are formed in the semiconductor substrate 101. For example, the semiconductor substrate 101 includes a common source region of a plurality of channel pillars.
[0080] like Figure 6a As shown, the semiconductor substrate 101 includes a first surface and a second surface opposite to each other, and an insulating stacked structure is formed on the first surface of the semiconductor substrate 101 .
[0081] The insulating stacked structure includes a plurality of stacked sacrificial layers 152, and adjacent sacrificial layers 152 are separated from each other by an insulating layer 102. In this embodiment, the insulating layer 102 is composed of silicon oxide, for example, and the sacrificial layers 152 are composed of silicon nitride, for example.
[0082] As described below, the sacrificial layer 152 will be replaced by gate conductors 121 to 123, the gate conductor 121 is connected to the string selection line in one step, the gate conductor 123 is connected to the ground selection line in one step, and the gate conductor 122 is connected to the word line in one step. In order to form a conductive path from the gate conductors 121 to 123 to the word line, the multiple sacrificial layers 152 are patterned into a step shape, that is, the edge portion of each sacrificial layer 152 is exposed relative to the upper sacrificial layer to provide an electrical connection area. After the patterning step of the multiple sacrificial layers 152, an insulating layer can be used to cover the insulating stack structure. Figure 6a In FIG. 1 , the interlayer insulating layer between the plurality of sacrificial layers 152 and the interlayer insulating layer covering the insulating stack structure are collectively shown as the insulating layer 102. However, the present invention is not limited thereto, and the plurality of interlayer insulating layers between and above the plurality of sacrificial layers 152 may be formed by a plurality of independent deposition steps.
[0083] Further, a channel hole is formed in the insulating laminate structure. In this embodiment, for example, a photoresist mask is formed on the surface of the semiconductor structure, and then anisotropic etching is performed to form a channel hole in the insulating laminate structure. Anisotropic etching can be performed by dry etching, such as ion milling etching, plasma etching, reactive ion etching, and laser ablation. For example, by controlling the etching time, the etching stops near the bottom of the common source region, and the etching stops near the bottom of the first insulating region. After etching, the photoresist mask is removed by dissolving in a solvent or ashing.
[0084] Further, a channel pillar 110 is formed in the channel hole. The lower portion of the channel pillar 110 includes a semiconductor layer. Further, the channel pillar 110 includes a channel layer extending from the upper portion thereof to the semiconductor layer. For clarity, Figure 6a The internal structure of the channel pillar 110 is not shown. Figure 1bIn the middle part of the channel column 110, the channel column 110 includes a tunneling dielectric layer, a charge storage layer and a blocking dielectric layer stacked in sequence on the channel layer, and at both ends of the channel column 110, the channel column 110 includes a blocking dielectric layer stacked on the channel layer or the semiconductor layer. The lower end of the channel column 110 is in contact with the common source region in the semiconductor substrate 101. In the final 3D memory device, the upper end of the channel column 110 will be connected to the wiring layer to form an effective memory cell. The structure of the channel column 110 is, for example, ONOP (oxide-nitride-oxide-polysilicon).
[0085] Preferably, a dummy channel column 130 is formed in the channel hole. The internal structure of the dummy channel column 130 may be the same as that of the channel column 110, and at least passes through at least a portion of the gate conductor in the gate stack structure. However, in the final 3D memory device, the upper end of the dummy channel column 130 is not connected to the wiring layer, thereby only providing a mechanical support function, and is not used to form a selection transistor and a storage transistor.
[0086] Preferably, a through hole is formed in the semiconductor substrate 101 and the insulating stacked structure, and a conductive channel 141 and an insulating liner 142 are formed in the through hole. The conductive channel 141 passes through the semiconductor substrate 101 and the insulating stacked structure, and is separated from the semiconductor substrate 101 and the insulating stacked structure by the insulating liner 142. One end of the conductive channel 141 extends to the upper surface of the insulating stacked structure, and the other end extends to the lower surface of the semiconductor substrate 101.
[0087] Furthermore, a through hole is formed in the semiconductor substrate 101 and the insulating laminate structure, and a heat conduction channel 143 and an insulating liner 144 are formed in the through hole. The heat conduction channel 143 passes through the semiconductor substrate 101 and the insulating laminate structure, and is separated from the semiconductor substrate 101 and the insulating laminate structure by the insulating liner 144. One end of the heat conduction channel 143 extends to the upper surface of the insulating laminate structure, and the other end extends to the lower surface of the semiconductor substrate 101.
[0088] like Figure 6b As shown, in the insulating stack structure, the sacrificial layer 152 is replaced by gate conductors 121 to 123 to form a gate stack structure.
[0089] In this step, a gate line gap 151 is formed in the insulating stack structure (see Figure 2a ), the insulating layer 102 is used as an etching stop layer, the sacrificial layer 152 is removed by etching through the gate line gap 151 to form a cavity, and a metal layer is used to fill the cavity to form gate conductors 121 to 123, wherein a plurality of gate conductors 121 to 123 and the insulating layer 102 are alternately stacked. Accordingly, a plurality of channel pillars 110 penetrate the gate stack structure.
[0090] When forming the gate line gap 151, anisotropic etching can be used, for example, dry etching, such as ion milling etching, plasma etching, reactive ion etching, laser ablation. For example, by controlling the etching time, the etching stops near the surface of the semiconductor substrate 101. In this embodiment, the gate line gap 151 divides the gate conductors 121 to 123 into a plurality of gate lines.
[0091] When forming the cavity, the gate line gap 151 is used as an etchant channel, and the sacrificial layer 152 in the insulating stack structure is removed by isotropic etching to form the cavity. Isotropic etching can be selective wet etching or vapor etching. In wet etching, an etching solution is used as an etchant, wherein the semiconductor structure is immersed in the etching solution. In vapor etching, an etching gas is used as an etchant, wherein the semiconductor structure is exposed to the etching gas.
[0092] In the case where the insulating layer 102 and the sacrificial layer 152 in the insulating stacked structure are respectively composed of silicon oxide and silicon nitride, a phosphoric acid solution can be used as an etchant in wet etching, and one or more of C4F8, C4F6, CH2F2 and O2 can be used in vapor phase etching. In the etching step, the etchant fills the gate line gap 151. The end of the sacrificial layer 152 in the insulating stacked structure is exposed to the opening of the gate line gap 151, so the sacrificial layer 152 contacts the etchant. The etchant gradually etches the sacrificial layer 152 from the opening of the gate line gap 151 to the inside of the insulating stacked structure. Due to the selectivity of the etchant, the etching removes the sacrificial layer 152 relative to the insulating layer 102 in the insulating stacked structure.
[0093] When forming the gate conductors 121 to 123 , the gate line gaps 151 are used as deposition channels, and atomic layer deposition (ALD) is adopted to fill the gate line gaps 151 and the cavities with metal layers.
[0094] In this embodiment, the metal layer is composed of tungsten, for example. The precursor source used in the atomic layer deposition is, for example, tungsten hexafluoride WF6, and the reducing gas used is, for example, silane SiH4 or diborane B2H6. In the step of atomic layer deposition, the tungsten material is obtained by chemical adsorption of the reaction product of tungsten hexafluoride WF6 and silane SiH4 to realize the deposition process.
[0095] like Figure 6c As shown, an interconnect structure is formed above the gate stack structure.
[0096] The interconnect structure includes a plurality of conductive channels 161 located above the gate stack structure, a plurality of contact pads 162 respectively contacting the plurality of conductive channels 161, a plurality of wiring layers 164 located on the plurality of contact pads 162, a plurality of external pads 165 located on the plurality of wiring layers 164, and a conductive channel 163 providing interconnection in a direction perpendicular to the surface of the semiconductor substrate 101.
[0097] The semiconductor structure formed in this step is a memory cell array 220, in which the gate stack structure forms a selection transistor and a storage transistor together with the channel pillar. In the middle part of the channel pillar 110, the gate conductors 121 to 123 form a storage transistor together with the channel layer, tunneling dielectric layer, charge storage layer and blocking dielectric layer inside the channel pillar 110. At both ends of the channel pillar 110, the gate conductors 121 to 123 form a selection transistor together with the channel layer (or semiconductor layer) and blocking dielectric layer inside the channel pillar 110.
[0098] The gate conductors 121, 122 and 123 in the gate stack structure are formed in a step shape, for example, to provide a space for the conductive channel 161 to extend to the corresponding gate conductor. The conductive channel and wiring layer of the memory cell array 220 are located in at least one insulating layer 102. As described above, the insulating layer 102 is shown as a single layer in the figure, however, the insulating layer 102 can actually be composed of a plurality of interlayer insulating layers, including a plurality of interlayer insulating layers for separating the gate conductors 121, 122 and 123 and a plurality of interlayer insulating layers for separating different wiring layers. In addition, the contact pad 162 and the external pad 165 can also be located on a separate interlayer insulating layer.
[0099] Further, the first ends of the channel pillars 110 are commonly connected to the common source region in the semiconductor substrate 101, and the second ends of the channel pillars 110 are connected to the contact pads 162 via the conductive channels 161, and then connected to the corresponding external pads 165 via the conductive channels and wiring. The first ends of the conductive channels 141 extend to the common source region in the semiconductor substrate 101, and the second ends are connected to the contact pads 162 via the conductive channels 161, and then connected to the corresponding external pads 165 via the conductive channels and wiring.
[0100] The bonding surface of the memory cell array 220 is the first surface of the insulating layer 102. In this step, the first surface is an exposed free surface. The contact surface of the external pad 165 is exposed on the first surface.
[0101] In this embodiment, the conductive channel 141 is extended from above the gate stack structure to the bonding surface of the memory cell array 220 using a process similar to that of the conductive channels 161 and 163 .
[0102] like Figure 6dAs shown, an interlayer insulating layer 104 and a heat dissipation structure 105 are sequentially formed on the second surface of the semiconductor substrate 101 .
[0103] In this step, for example, the semiconductor structure is turned over so that the second surface faces upward, so as to form an interlayer insulating layer 104 and a heat dissipation structure 105. The interlayer insulating layer 104 is used to separate the semiconductor substrate 101 and the heat dissipation structure 105 from each other. In this embodiment, the interlayer insulating layer 104 is composed of silicon oxide, for example, and the heat dissipation structure 105 is composed of a metal or resin with good thermal conductivity, such as copper, silver, aluminum thermal conductive silicone.
[0104] Between the steps of forming the interlayer insulating layer 104 and the heat dissipation structure 105, it also includes: forming a through hole aligned with the conductive channel 141 in the interlayer insulating layer 104, and filling it with a conductive material, thereby extending the conductive channel 141 to the surface of the interlayer insulating layer 104; and forming a through hole aligned with the heat conduction channel 143 in the interlayer insulating layer 104, and filling it with a thermal conductive material, thereby extending the heat conduction channel 143 to the surface of the interlayer insulating layer 104.
[0105] The first end and the second end of the conductive channel 141 extend to two opposite surfaces (i.e., the free surface and the bonding surface) of the memory cell array 220 to form external pads. The conductive channel 141 is a part of a through contact via (TAC) for providing a conductive path between the CMOS circuit and the external circuit.
[0106] The first end and the second end of the heat conduction channel 143 extend to the free surface of the memory cell array 220 and the wiring layer 164 respectively.
[0107] The heat dissipation structure 105 is, for example, patterned and includes an opening area for exposing the first end of the conductive channel 141. The heat dissipation structure 105 is connected to the first end of the thermal conductive channel 143, thereby providing a heat dissipation path.
[0108] like Figure 6e As shown, transistors (not shown) of a CMOS circuit are formed in a semiconductor substrate 201 , and an interconnection structure is formed on the semiconductor substrate 201 .
[0109] In this embodiment, the semiconductor substrate 201 is, for example, a single crystal silicon substrate. In order to form transistors, a plurality of doped regions are formed in the semiconductor substrate 201. For example, the semiconductor substrate 201 includes source regions and drain regions of a plurality of transistors.
[0110] The semiconductor structure formed at this step is a CMOS circuit 210 , in which doped regions of a plurality of transistors formed in a semiconductor substrate 201 provide external electrical connections via an interconnect structure.
[0111] The interconnect structure includes a plurality of contact pads 261 on the semiconductor substrate 201, a plurality of wiring layers 263 on the plurality of contact pads 261, a plurality of external pads 264 on the plurality of wiring layers 263, and conductive channels 262 providing interconnections in a direction perpendicular to the surface of the semiconductor substrate 201. The plurality of wiring layers 260 are separated from each other, and the plurality of wiring layers 260 and the contact pads 261 and the external pads 264 are separated from each other by an insulating layer 202, and are electrically connected to each other by conductive channels 262 in the insulating layer 202.
[0112] like Figure 6f As shown, a CMOS circuit 210 and a memory cell array 220 are bonded to each other into a 3D memory device 200 .
[0113] When the CMOS circuit 210 and the memory cell array 220 are bonded to each other to form the 3D memory device 200, a first group of pads among the external pads 264 of the CMOS circuit 210 and the external pads 165 of the memory cell array 220 are in contact with each other, thereby realizing electrical connection between the CMOS circuit 210 and the memory cell array 220, and a second group of pads among the external pads 264 of the CMOS circuit 210 and the second end of the conductive channel 141 of the memory cell array 220 are in contact with each other, thereby using the conductive channel 141 as part of a through contact via (TAC) to realize electrical connection between the CMOS circuit 210 and the external circuit.
[0114] In the above-mentioned embodiment, it is described that the 3D memory device is a bonding component including a CMOS circuit and a memory cell array, and a heat conduction channel and a heat dissipation structure are formed in the memory cell array to provide a heat dissipation path. In an alternative embodiment, the present invention can also be applied to a 3D memory device including only a memory cell array but not a CMOS circuit, and a heat conduction channel and a heat dissipation structure are formed in the memory cell array to provide a heat dissipation path.
[0115] In the above description, the technical details of the patterning and etching of each layer are not described in detail. However, it should be understood by those skilled in the art that various technical means can be used to form layers, regions, etc. of the desired shape. In addition, in order to form the same structure, those skilled in the art can also design methods that are not completely the same as the methods described above. In addition, although the various embodiments are described above separately, this does not mean that the measures in the various embodiments cannot be used in combination to advantage.
[0116] The embodiments of the present invention are described above. However, these embodiments are only for illustrative purposes and are not intended to limit the scope of the present invention. The scope of the present invention is defined by the appended claims and their equivalents. Without departing from the scope of the present invention, those skilled in the art may make various substitutions and modifications, which should all fall within the scope of the present invention.
Claims
1. A semiconductor device, comprising: A memory array, comprising a first side and a second side distributed along a first direction, and a first pad layer located on the first side and a first pad located on the first pad layer; a heat dissipation structure, located at the second side; The heat conduction channel is located in the storage array and includes a first end and a second end arranged along the first direction, the first end is connected to the first pad, and the second end is connected to the heat dissipation structure.
2. The semiconductor device according to claim 1, The storage array further comprises a stacked structure, wherein the stacked structure is located between the first pad layer and the heat dissipation structure; The heat conduction channel runs through the stacked structure.
3. The semiconductor device according to claim 1, The memory array further includes a first channel structure and a second channel structure penetrating the stacked structure along the first direction; the stacked structure includes a plurality of overlapping conductive layers and interlayer insulating layers.
4. The semiconductor device according to claim 3, The first channel structure includes a channel layer, a tunneling dielectric layer, a charge storage layer, and a blocking dielectric layer arranged from the inside to the outside in a second direction, and the second direction is perpendicular to the first direction.
5. The semiconductor device according to claim 3, The heat conduction channel includes the second channel structure.
6. The semiconductor device according to claim 3, The storage array further includes an insulating liner, which is disposed around the heat-conducting channel and is located between the heat-conducting channel and the stacked structure.
7. The semiconductor device according to claim 2, The memory array further comprises a first semiconductor layer, wherein the first semiconductor layer is located between the heat dissipation structure and the stacked structure; The heat conduction channel passes through the first semiconductor layer and is insulated from the first semiconductor layer.
8. The semiconductor device according to claim 2, The storage array further includes a wiring layer, wherein the wiring layer is located between the stacked structure and the first pad layer, and the second end of the heat-conducting structure is connected to the first pad through the wiring layer.
9. The semiconductor device according to claim 1, further comprising: A CMOS circuit is located on the first side and includes a second pad layer located on a side of the first pad layer away from the heat dissipation structure, and a second pad located on the second pad layer; The second pad is connected to the first pad.
10. The semiconductor device according to claim 9, The CMOS circuit further includes a second semiconductor layer, and the second semiconductor layer is located on a side of the second pad layer away from the first pad layer.
11. The semiconductor device according to claim 9, further comprising: A conductive channel runs through the storage array and is connected to the second pad.
12. The semiconductor device according to claim 1, The heat dissipation structure and / or the heat conduction channel include at least one of metal and resin materials.
13. The semiconductor device according to claim 7, The heat dissipation structure and / or the heat conduction channel include at least one of copper, silver and aluminum thermal conductive silicone.
