Three-dimensional memory device, manufacturing method therefor and electronic apparatus

By using a vertical stacking design and dual-gate electrode structure for three-dimensional memory devices, the leakage current and manufacturing challenges of traditional silicon-based DRAM during the shrinkage process are solved, achieving performance optimization and stability improvement of high-density, low-power memory devices.

WO2026237175A1PCT designated stage Publication Date: 2026-11-19INST OF MICROELECTRONICS CHINESE ACAD OF SCI LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/CN2025/110777
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-05-14
Filing Date
2025-07-26
Publication Date
2026-11-19

AI Technical Summary

Technical Problem

Traditional silicon-based DRAM memory cells face challenges such as increased leakage current, shortened data retention time, and complex and costly manufacturing processes during the miniaturization process, making it difficult to further increase storage density.

Method used

It adopts a vertical stacking design of three-dimensional memory devices, combining a vertical annular channel and a dual-gate electrode structure, using wide bandgap materials such as indium gallium zinc oxide as channel materials, and improving storage density through three-dimensional stacking technology, while optimizing the memory cell structure to reduce leakage current and signal crosstalk.

Benefits of technology

It increases the storage density of storage devices, extends data retention time, reduces power consumption, simplifies the manufacturing process, and enhances the gate control capability and device stability of transistors.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN2025110777_19112026_PF_FP_ABST
    Figure CN2025110777_19112026_PF_FP_ABST
Patent Text Reader

Abstract

The present disclosure provides a three-dimensional memory device, a manufacturing method therefor and an electronic apparatus. The device comprises a substrate, and a plurality of memory cell layers and a plurality of isolation layers alternately stacked on the surface of the substrate in a vertical direction. The orthographic projections of the memory cell layers on the surface of the substrate completely overlap each other; each memory cell layer comprises a common bit line extending in a first horizontal direction, and a plurality of memory cells symmetrically distributed about the common bit line as an axis. In the present disclosure, a three-dimensional stacking technology is introduced to stack the memory cells in the vertical direction, thereby improving the overall density of the memory; and an indium gallium zinc oxide having a wide band gap characteristic is used as a channel material to reduce leakage currents of transistors, thus prolonging the data retention time, decreasing the refresh rate and reducing power consumption of devices. In addition, the structures of the memory cells are optimized, by means of a double-gate vertical ring channel design, the gate control capability of transistors is effectively improved while signal crosstalk of adjacent memory cells is shielded, thereby achieving performance optimization and stability improvement of memory devices.
Need to check novelty before this filing date? Find Prior Art

Description

A three-dimensional storage device and its fabrication method, and an electronic device.

[0001] [Amended according to Rule 26 02.09.2025] This application claims priority to Chinese Patent Application No. 202510620748.X, filed on May 14, 2025, entitled “A Three-Dimensional Storage Device and its Preparation Method, Electronic Device”, the entire contents of which are incorporated herein by reference. Technical Field

[0002] This disclosure relates to the field of semiconductor storage technology, and in particular to a three-dimensional storage device and its fabrication method, as well as an electronic device. Background Technology

[0003] DRAM (Dynamic Random Access Memory) is a core memory technology in modern computer systems, primarily used for main memory and cache storage. Its high speed, low cost, and high density make it a crucial component for data storage in computer systems. The basic working principle of DRAM is to store each bit of data using memory cells composed of transistors and capacitors. Transistors control the charging and discharging state of the capacitors, thus determining the writing and reading of bits; capacitors store electrical charge, and the presence or absence of charge represents the value of the stored data bit. Silicon-based 1T1C (i.e., a memory cell consisting of one transistor and one capacitor) DRAM is currently the most common memory implementation scheme, increasing storage density by continuously shrinking feature sizes. However, with advancements in process technology, traditional silicon-based DRAM is facing a series of physical limits and manufacturing challenges.

[0004] As DRAM cell sizes shrink, leakage current increases in traditional silicon-based 1T1C memory cells, leading to shorter data retention times and requiring more frequent refresh operations. This increases power consumption and reduces efficiency. Further minimizing the feature size of traditional silicon-based 1T1C DRAM still heavily relies on high-precision photolithography, resulting in complex and costly manufacturing processes, and further increases in storage density are difficult to achieve. Summary of the Invention

[0005] The purpose of this disclosure is to provide a three-dimensional storage device and its fabrication method, as well as an electronic device, to solve the performance bottlenecks and process challenges faced by DRAM in the prior art.

[0006] The embodiments of this disclosure adopt the following technical solution: a three-dimensional memory device, comprising at least: a substrate; a plurality of memory cell layers and a plurality of isolation layers alternately stacked in a vertical direction on the surface of the substrate, wherein the orthographic projections of each memory cell layer on the surface of the substrate completely overlap, and the vertical direction is a direction perpendicular to the surface of the substrate; each memory cell layer comprises at least: a common bit line extending in a first horizontal direction, and a plurality of memory cells symmetrically distributed about the common bit line, the memory cells extending in a second horizontal direction and being mutually insulated, the first horizontal direction and the second horizontal direction being parallel to the surface of the substrate, and the first horizontal direction and the second horizontal direction being perpendicular to each other; each memory cell comprises at least: a transistor on the side near the common bit line and a capacitor on the side away from the common bit line, the transistor comprising at least a first gate electrode, a second gate electrode, a source electrode, a drain electrode, a channel layer, a gate dielectric layer, and an insulating layer; wherein the first gate electrode extends in the vertical direction and penetrates the memory cell to which the transistor belongs. A first common word line is formed by electrically connecting the first gate electrode of a transistor in the same position as the memory cell layer and its adjacent isolation layer. The gate dielectric layer surrounds the first gate electrode with the imaginary line in the vertical direction as its axis, and the channel layer surrounds the gate dielectric layer with the imaginary line as its axis. The drain is located between the channel layer and the common bit line and is electrically connected to the common bit line. The source is located between the channel layer and the capacitor and is electrically connected to one side electrode plate of the capacitor. The insulating layer covers the sidewalls of the channel layer, the source, and the drain. The second gate electrode is located outside the insulating layer. The orthographic projection of the second gate electrode on the imaginary plane perpendicular to the first horizontal direction completely covers the orthographic projection of the channel layer on the imaginary plane. The second gate electrodes of all transistors in the same position of the memory cell layer penetrate the isolation layer adjacent to their respective memory cell layers to achieve electrical connection and form a second common word line. The first common word line and the second common word line are electrically connected.

[0007] In some embodiments, the channel layer is prepared based on a metal oxide material.

[0008] In some embodiments, the second gate electrode includes a first electrode plate and a second electrode plate, the first electrode plate and the second electrode plate being symmetrically distributed on the outer side of the insulating layer, and the orthographic projections of the first electrode plate and the second electrode plate on the imaginary surface coinciding.

[0009] In some embodiments, the capacitor is a cylindrical capacitor.

[0010] In some embodiments, a support material is filled between adjacent storage cells, and the support material is insulating.

[0011] This disclosure also provides a method for fabricating a three-dimensional memory device as described above, comprising: providing a substrate; alternately fabricating an isolation layer and a sacrificial layer on the surface of the substrate to obtain a stacked structure; etching the stacked structure in a vertical direction to obtain a common structure extending in a first horizontal direction and a plurality of independent structures extending in a second horizontal direction; growing an insulating material on the outer surface of the common structure and all the independent structures, and removing the insulating material located at the top of the common structure and all the independent structures to form an insulating layer; and opening a first through-hole penetrating the independent structure at one end of each independent structure near the common structure, wherein the width of the first through-hole in the first horizontal direction is smaller than that of the independent structure in the first horizontal direction. The width of the above; lateral etching is performed on the exposed sacrificial layer in the first via to form a first groove, so that each sacrificial layer in the independent structure is broken at the first groove, forming a drain sacrificial layer near the common structure and a source sacrificial layer away from the common structure; a channel material is grown on the currently exposed surface of the first groove to form a channel layer; a gate dielectric material and a gate material are grown on the currently exposed surface of the stacked structure, and the gate material fills the first via; the gate material is etched to form a first common word line and a second common word line in the vertical direction; the source sacrificial layer on the side of the independent structure away from the common structure is etched to form a second groove, and a capacitor is formed in the second groove.

[0012] In some embodiments, the sacrificial layer and the gate material are each independently a stack of one or more material layers selected from titanium nitride, tungsten, and indium zinc oxide.

[0013] In some embodiments, etching the source sacrificial layer on the side of the independent structure away from the common structure to form a second groove, and forming a capacitor within the second groove, includes: laterally etching the source sacrificial layer on the side of the independent structure away from the common structure to form a second groove, wherein the depth of the second groove in the second horizontal direction is less than the length of the source sacrificial layer in the second horizontal direction; sequentially growing a conductive material and an insulating material on the currently exposed surface of the stacked structure to form one side electrode plate and a capacitor dielectric layer, wherein the sum of the growth thicknesses of the conductive material and the insulating material is less than half the height of the second groove in the vertical direction; and growing a conductive material on the currently exposed surface of the capacitor dielectric layer until the second groove is completely filled to form the other side electrode plate of the capacitor.

[0014] In some embodiments, the method further includes filling the gaps between all the individual structures with a support material, wherein the support material is an insulating material.

[0015] This disclosure also provides an electronic device that includes at least the three-dimensional storage device described above.

[0016] The beneficial effects of the embodiments disclosed herein are as follows: the introduction of three-dimensional stacking technology allows for vertical stacking of memory cells to improve the overall density of the memory; at the same time, the use of wide-bandgap indium gallium zinc oxide as a channel material reduces transistor leakage current, thereby extending data retention time, reducing refresh rate and lowering device power consumption; in addition, the structure of the memory cells is optimized, and the gate control capability of the transistors is effectively improved through a dual-gate combined with a vertical ring channel design, while shielding signal crosstalk between adjacent memory cells, thereby achieving performance optimization and stability improvement of the memory device. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in one or more embodiments of this specification or in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 is a schematic diagram of the structure of the three-dimensional storage device in the first embodiment of this disclosure;

[0019] Figure 2 is a cross-sectional schematic diagram of the three-dimensional storage device shown in Figure 1 at AA';

[0020] Figure 3 is a cross-sectional schematic diagram of the three-dimensional storage device shown in Figure 1 at BB';

[0021] Figure 4 is a flowchart of the fabrication method of the three-dimensional storage device in the second embodiment of this disclosure;

[0022] Figure 5 is a three-dimensional schematic diagram of the stacked structure after step S2 is completed in the second embodiment of this disclosure;

[0023] Figure 6 is a cross-sectional schematic diagram of the stacked structure in the YZ plane after step S2 is completed in the second embodiment of this disclosure;

[0024] Figure 7 is a three-dimensional schematic diagram of the remaining structure of the stacked structure after step S3 is completed in the second embodiment of this disclosure;

[0025] Figure 8 is a top view of Figure 7;

[0026] Figure 9 is a three-dimensional schematic diagram of the insulating material after growth in step S4 of the second embodiment of this disclosure;

[0027] Figure 10 is a three-dimensional schematic diagram of step S3 after completion in the second embodiment of this disclosure;

[0028] Figure 11 is a cross-sectional view of Figure 10 in the YZ plane;

[0029] Figure 12 is a three-dimensional schematic diagram of the stacked structure after etching the first through hole in the second embodiment of this disclosure;

[0030] Figure 13 is a schematic diagram of the cross section at AA' in Figure 12;

[0031] Figure 14 is a cross-sectional schematic diagram of the stacked structure after etching the first groove in the second embodiment of this disclosure in the YZ plane;

[0032] Figure 15 is a schematic diagram of the growth of the channel material in the second embodiment of this disclosure;

[0033] Figure 16 is a cross-sectional schematic diagram of the stacked structure in the YZ plane after the channel layer is completed in the second embodiment of this disclosure;

[0034] Figure 17 is a cross-sectional schematic diagram of the stacked structure in the YZ plane after step S8 is completed in the second embodiment of this disclosure;

[0035] Figure 18 is a three-dimensional schematic diagram of the stacked structure after step S9 is completed in the second embodiment of this disclosure;

[0036] Figure 19 is a schematic diagram of the cross section at AA' in Figure 18;

[0037] Figure 20 is a three-dimensional schematic diagram of the stacked structure after step S10 is completed in the second embodiment of this disclosure;

[0038] Figure 21 is a cross-sectional schematic diagram of the stacked structure in the YZ plane after the capacitor is completed in the second embodiment of this disclosure;

[0039] Figure 22 is a cross-sectional schematic diagram of the CMP-fed stacked structure in the YZ plane in the second embodiment of this disclosure. Detailed Implementation

[0040] To enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the technical solutions in one or more embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, and not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of this document.

[0041] DRAM (Dynamic Random Access Memory) is a core memory technology in modern computer systems, primarily used for main memory and cache storage. Its high speed, low cost, and high density make it a crucial component for data storage in computer systems. The basic working principle of DRAM is to store each bit of data using memory cells composed of transistors and capacitors. Transistors control the charging and discharging state of the capacitors, thus determining the writing and reading of bits; capacitors store electrical charge, and the presence or absence of charge represents the value of the stored data bit. Silicon-based 1T1C (i.e., a memory cell consisting of one transistor and one capacitor) DRAM is currently the most common memory implementation scheme, increasing storage density by continuously shrinking feature sizes. However, with advancements in process technology, traditional silicon-based DRAM is facing a series of physical limits and manufacturing challenges.

[0042] As DRAM cell sizes shrink, leakage current increases in traditional silicon-based 1T1C memory cells, leading to shorter data retention times and requiring more frequent refresh operations. This increases power consumption and reduces efficiency. Further minimizing the feature size of traditional silicon-based 1T1C DRAM still heavily relies on high-precision photolithography, resulting in complex and costly manufacturing processes, and further increases in storage density are difficult to achieve.

[0043] To address the aforementioned issues, the first embodiment of this disclosure provides a three-dimensional storage device that utilizes a stackable vertical design to increase the density of storage cells. Simultaneously, the DRAM storage cells are implemented based on transistors and capacitors. Combining the vertical storage cell stacking method, the transistors in this embodiment employ a vertical annular channel (CAA) and a horizontal bit line design, and further incorporate an additional second gate electrode configuration, providing the storage cells with faster read / write speeds and better signal crosstalk shielding capabilities.

[0044] Figure 1 shows a schematic diagram of the structure of the three-dimensional memory device of this embodiment, with a partial cross-sectional layer diagram. As shown in Figure 1, the three-dimensional memory device of this embodiment includes a substrate and multiple memory cell layers 1 and isolation layers 2 arranged alternately in a vertical direction on the surface of the substrate. The orthographic projection of each memory cell layer on the substrate surface completely overlaps, that is, the structure of each memory cell layer 1 is completely identical. In this embodiment, the vertical direction is the direction perpendicular to the substrate surface, corresponding to the Z direction marked in the coordinate system of Figure 1. In this embodiment, the memory cell layer 1 mainly refers to the layer for fabricating memory cells, while the isolation layer 2 is mainly used to achieve insulation isolation between the memory cell layers 1. For example, Figure 1 shows a three-dimensional memory device based on the stacking of five memory cell layers 1, and the light blue filling layer between adjacent memory cell layers 1 is the isolation layer 2.

[0045] Each memory cell layer 1 in this embodiment includes at least a common bit line 10 extending along a first horizontal direction, and a plurality of memory cells 20 symmetrically distributed about the common bit line 10. The memory cells 20 all extend along a second horizontal direction and are mutually insulated. Both the first and second horizontal directions are parallel to the substrate surface and perpendicular to each other, corresponding to the X and Y directions marked in the coordinate system of Figure 1. Figure 2 shows a cross-sectional schematic diagram of the three-dimensional memory device shown in Figure 1 at AA', i.e., a hierarchical schematic diagram of the memory cell layer 1 in the XY plane. Figure 3 shows a cross-sectional schematic diagram of the three-dimensional memory device shown in Figure 1 at BB', i.e., a hierarchical schematic diagram of the three-dimensional memory device in the YZ plane.

[0046] As shown in Figures 2 and 3, the memory cell 20 includes at least a transistor 100 and a capacitor 200. The transistor 1 is close to the common bit line 10, while the capacitor 2 is far from the common bit line 10. In this embodiment, the transistor 1 includes at least a first gate electrode 110, a second gate electrode 120, a source 130, a drain 140, a channel layer 150, a gate dielectric layer 160, and an insulating layer 170. The first gate electrode 100 extends vertically and penetrates the memory cell layer 1 to which the transistor belongs and its adjacent isolation layer 2, and is electrically connected to the first gate electrode 110 of transistors at the same position as those in other memory cell layers 1. The first common word line WL1 shown in Figure 3 is formed; the gate dielectric layer 160 surrounds the first gate electrode 110 with an imaginary line in the vertical direction as the axis, and the channel layer 150 surrounds the gate dielectric layer 160 with an imaginary line as the axis, forming a CAA architecture transistor; the drain 140 is located between the channel layer 150 and the common bit line 10, and is electrically connected to the common bit line 10. It should be noted that in this embodiment, both the drain 140 and the common bit line 10 are made of conductive materials, and they can be fabricated based on the same process steps in actual fabrication. Therefore, the construction of the drain 140 and the common bit line 10 is not explicitly shown in the figure. The boundary division can also be understood as the common bit line 10 actually serving as the drain 140 of each transistor 100; the source 130 is located between the channel layer 150 and the capacitor 200, and is electrically connected to one side electrode plate of the capacitor 200; the insulating layer 170 covers the sidewalls of the channel layer 150, the source 130, and the drain 140, and at the same time can also cover the exposed sidewalls of the common bit line 10 and the capacitor 200 to provide insulation; the second gate electrode 120 is located outside the insulating layer 170, and the orthogonal projection of the second gate electrode 120 on the imaginary plane perpendicular to the first horizontal direction completely covers the channel. The orthographic projection of layer 150 onto the imaginary surface allows its electric field to fully act on the channel layer when energized. Furthermore, the second gate electrode 120 of transistors located at the same position in all memory cell layers 1 penetrates the adjacent isolation layer 2 of their respective memory cell layers to achieve electrical connection and form a second common word line WL2. In actual implementation, the first common word line WL1 and the second common word line WL2 are electrically connected (the specific electrical connection method is not shown in the figure) so that the first common word line WL1 and the second common word line WL2 can act simultaneously when controlling the transistor, thereby improving the controllability and performance of the transistor.

[0047] In this embodiment, the first gate electrode 110, the second gate electrode 120, the source electrode 130, the drain electrode 140, and the common bit line 10 are all made of conductive materials. Typically, the materials used for these layers can be stacked independently of one or more of titanium nitride, tungsten, and indium zinc oxide, or other conductive materials. Different materials can be selected based on the desired function during fabrication. For example, indium zinc oxide (IZO) can be used as the gate electrode material, and titanium nitride (TiN) can be used for the common bit line 10 and the source / drain electrodes. The channel layer 150 is fabricated using semiconductor materials. This embodiment primarily uses metal oxide materials, including but not limited to indium gallium zinc oxide (IGZO) and indium tin zinc oxide (ITZO), utilizing their wide bandgap characteristics to reduce transistor leakage current, thereby extending the data retention time of the memory cell, reducing reliance on frequent refresh operations, and lowering overall power consumption. The isolation layer 2, the gate dielectric layer 160, and the insulating layer 170 are fabricated using insulating materials, including but not limited to silicon oxide (SiO) and hafnium oxide (HfO). x It is important to note that for both gate and capacitor dielectric materials, high-k dielectric materials (High-K materials) should be primarily selected. Using High-K materials for the gate dielectric can further reduce the gate leakage current of the transistor and improve transistor performance. Using High-K materials for the capacitor dielectric can increase the capacitance and mitigate capacitor discharge. It is also crucial to ensure a high etch selectivity between adjacent layers during the fabrication process to guarantee effective etching across different layers.

[0048] Referring to the three-dimensional device illustrations shown in Figures 1 to 3, in this embodiment, the second gate electrode 120 of each transistor 100 includes a first electrode plate 121 and a second electrode plate 122. These two electrode plates are symmetrically distributed on the outside of the insulating layer 170, and the orthographic projections of the first electrode plate 121 and the second electrode plate 122 on the imaginary plane perpendicular to the first horizontal direction coincide, meaning they are the same size and simultaneously act on the channel layer 150. This further enhances the gate control capability of the transistor using a dual-gate design on the CAA architecture transistor, improving transistor performance. Simultaneously, the dual-gate design shields signal crosstalk between adjacent memory cells, improving the overall operational stability of the device. It should be noted that the insulating layer 170 serves as the dielectric layer between the second gate electrode 120 and the channel layer 150; therefore, its fabrication material can also be a High-K material. In addition, the second common word line WL2, formed by the electrical connection of the second gate electrode of the transistor at the same position in different memory cell layers, is actually placed on both sides of the channel layer. It can be electrically connected to the first common word line WL1 on the top of the three-dimensional memory device through the design of a sacrificial layer. The specific electrical connection method can be made according to the design position requirements of other layers of the device. It is not shown in the figure. This embodiment does not impose any restrictions. As long as it can be achieved that the electric fields of the first common word line WL1 and the second common word line WL2 can act on the channel layer at the same time when a certain transistor is controlled, it is acceptable.

[0049] In some embodiments, capacitor 200 is a cylindrical capacitor, as shown in Figures 2 and 3. Capacitor 200 mainly includes a first electrode 210, a second electrode 220, and a capacitor dielectric layer 230. The cross-sectional shape of the first electrode 210 is a laterally extending U-shape. The opening direction of the U-shaped electrode is away from the common electrode line 10, and its bottom (i.e., the closed end of the U-shaped electrode) is electrically connected to the source electrode 130. Since both the first electrode 210 and the source electrode 130 are made of conductive material, it can be considered that the source electrode 130 and the first electrode 210 together complete the function of one side electrode of the capacitor. The capacitor dielectric layer 230 is placed inside the U-shaped plate of the first electrode 210, and its thickness should be the same at any position. The second electrode 220 completely fills the remaining space of the U-shaped plate of the first electrode 210, serving as the other side electrode of the capacitor and also playing a supporting role. In actual implementation, the second plate 220 of capacitor 200 is grounded. The specific method by which the second plate 220 is grounded is not shown in the figure; it can be implemented according to the actual structure of the three-dimensional storage device and conventional grounding settings. This embodiment does not impose any limitations. It should be noted that the structure of capacitor 200 described above is merely an example. In actual fabrication, it can be implemented using other capacitor structures within the structural constraints of the storage cell.

[0050] Corresponding to the air gap between adjacent storage cells 20, after the fabrication of each device layer inside the storage cell 20 is completed, the air gap between the storage cells can be filled with a support material. The support material should be insulating to ensure that crosstalk between storage cells 20 is minimized as much as possible and to isolate each layer inside the storage cell 20 from the external environment.

[0051] This embodiment introduces three-dimensional stacking technology to vertically stack memory cells to increase the overall density of the memory. At the same time, it uses wide-bandgap indium gallium zinc oxide as the channel material to reduce transistor leakage current, thereby extending data retention time, reducing refresh rate and lowering device power consumption. In addition, the structure of the memory cells is optimized. Through a dual-gate combined with a vertical ring channel design, the gate control capability of the transistors is effectively improved, while shielding signal crosstalk between adjacent memory cells, thereby optimizing the performance and improving the stability of the memory device.

[0052] The second embodiment of this disclosure provides a method for fabricating a three-dimensional storage device according to the first embodiment. The flowchart of the method is shown in Figure 4. The method mainly includes steps S1 to S10. The fabrication process of the three-dimensional storage device will be described in detail below with reference to Figures 4 to 22.

[0053] S1 provides a substrate.

[0054] A substrate is a base plate used to support a memory chip, and one or more film layers can be fabricated on the substrate. The substrate type can be an insulating substrate, a semiconductor substrate, a conductive substrate, etc. Insulating substrates can include glass substrates, quartz substrates, sapphire substrates, zirconium oxide substrates, resin substrates, etc. Semiconductor substrates can include semiconductor substrates made of materials such as silicon or germanium, or compound semiconductor substrates made of materials such as silicon carbide, silicon germanium, gallium arsenide, indium phosphide, zinc oxide, or gallium oxide. Conductive substrates can include graphite substrates, metal substrates, alloy substrates, conductive resin substrates, etc. In some embodiments, the substrate can also be a flexible substrate, such as a polyimide (PI) substrate, a polyethylene terephthalate (PET) substrate, or a surface-treated polymer soft film substrate. In some embodiments, the substrate can also be referred to as a wafer.

[0055] S2, an isolation layer and a sacrificial layer are alternately prepared on the surface of the substrate to obtain a stacked structure.

[0056] In this embodiment, a stacked structure can be obtained by alternately depositing an isolation layer and a sacrificial layer on a substrate, and the memory is fabricated by processing this stacked structure. The isolation layer is used to isolate different memory cell layers, and the sacrificial layer is partially or completely sacrificed during device fabrication to provide a foundation for subsequent etching and fabrication. This embodiment does not limit the method of fabricating the isolation layer and the sacrificial layer; they can be deposited in the same way or in different ways. Exemplarily, the deposition methods include, but are not limited to, sputtering, evaporation, chemical vapor deposition, and atomic layer deposition.

[0057] In this embodiment, the isolation layer can be obtained by depositing an isolation material (e.g., an insulating material), and the sacrificial layer can be obtained by depositing a material with an etching selectivity ratio to the isolation layer. For example, the isolation layer is made of silicon oxide, and the sacrificial layer is made of titanium nitride, tungsten, or indium zinc oxide, or other materials with conductive properties, so that the remaining portion of the sacrificial layer can be directly used as a partial conductive layer of the memory cell. Figure 5 shows a three-dimensional schematic diagram of the stacked structure of this embodiment, and Figure 6 shows a cross-sectional schematic diagram of the stacked structure in the YZ plane, wherein the isolation layer is prepared using silicon oxide (SiO), and the sacrificial layer is prepared using titanium nitride (TiN). It should be noted that the number of isolation layers and sacrificial layers shown in Figures 5 and 6 is only illustrative. More can be stacked as needed, for example, hundreds of isolation layers and hundreds of sacrificial layers can be stacked alternately to meet the requirement of completing more memory cell fabrication processes in the same process step.

[0058] S3, etch the stacked structure in the vertical direction to obtain a common structure extending in the first horizontal direction and multiple independent structures extending in the second horizontal direction.

[0059] When etching the stacked structure, the stacked structure can be etched vertically using a dry etching method until the substrate is finished. The common structure is the part of the etched structure that extends along the first horizontal direction. At this time, the remaining sacrificial layer part in the common structure can be used as the common bit line in the memory device. The independent structures extend along the second horizontal direction and are used to fabricate memory cells. Their number is set according to the device size, process level and fabrication density requirements. There are gaps between adjacent independent structures for electrical isolation between different memory cells.

[0060] Figure 7 is a three-dimensional schematic diagram of the remaining structure of the stacked structure after step S3 is completed, and Figure 8 is a top view of Figure 7. As shown in Figure 8, the critical dimension of the independent structure is 400nm, which mainly refers to the width of the independent structure. During the etching process, the gap between adjacent independent structures is the same as the width of the independent structure.

[0061] S4. Insulating material is grown on the outer surface of the common structure and all independent structures, and the insulating material located on top of the common structure and all independent structures is removed to form an insulating layer.

[0062] The insulating material can be grown on the outer surfaces of the common structure and all individual structures using atomic layer deposition (ALD), forming the structural schematic shown in Figure 9. Specifically, the insulating material can be 10 nm hafnium-based oxide (HfO). Subsequently, the insulating material on the top of the common structure and all individual structures is removed by dry etching, leaving the remaining insulating material forming an insulating layer on the sidewalls of the common structure and all individual structures, as shown in Figure 10. Figure 11 is a cross-sectional schematic diagram of Figure 10 in the YZ plane.

[0063] S5, a first through hole is opened at the end of each independent structure that is close to the common structure, penetrating the independent structure.

[0064] The first via in this embodiment can be fabricated using a dry etching method. The cross-sectional shape of the via can be square or circular. The width of the first via in the first horizontal direction is smaller than the width (i.e., feature size) of the independent structure in the first horizontal direction. In this embodiment, this is limited to half the feature size of the independent structure, for example, 200 nm. Furthermore, the location of the first via in the second horizontal direction should be maintained at the end of the independent structure closest to the common structure to ensure sufficient space for capacitor fabrication at the end furthest from the common structure. Its location in the first horizontal direction should be at the center of the width of the independent structure so that the remaining width of the corresponding independent structure portion after etching the first via is consistent, facilitating subsequent layer fabrication. Figure 12 shows a three-dimensional schematic diagram of the stacked structure after etching the first via, and Figure 13 is a cross-sectional schematic diagram at AA' in Figure 12.

[0065] S6, the exposed sacrificial layer is laterally etched in the first through hole to form a first groove, so that each sacrificial layer in the independent structure is broken at the first groove, forming a drain sacrificial layer on the side close to the common structure and a source sacrificial layer on the side away from the common structure.

[0066] In the first via, the sacrificial layer is etched laterally in the horizontal plane using lateral wet etching to form a first groove. The depth of the first groove should be greater than half the remaining width after the independent structure is etched into the first via, for example, 50 nm. This ensures that the remaining portion of the sacrificial layer after the independent structure is etched into the first via during the lateral etching of the first groove, thereby breaking each sacrificial layer in the independent structure at the first groove, forming a drain sacrificial layer near the common structure and a source sacrificial layer away from the common structure, as shown in Figure 14. It should be noted that in Figure 14, since the sacrificial layers are all stacked in step S2, the drain sacrificial layer formed after the first groove is actually an integral hierarchical structure with the sacrificial layer in the common structure of the same layer, which is used as the drain of the transistor and the common bit line in the memory cell.

[0067] S7, Growing channel material on the currently exposed surface of the first groove to form a channel layer.

[0068] In actual fabrication, a channel material, such as 6nm IGZO, can be grown on the exposed surface of the current stacked structure, as shown in Figure 15. Then, the current channel material is dry-etched to retain only the channel material on the surface of the first groove in each memory cell layer, so as to form an independent annular channel layer for each layer, as shown in Figure 16.

[0069] S8, grow gate dielectric material and gate material on the currently exposed surface of the stacked structure, and fill the first via with gate material.

[0070] Subsequently, gate dielectric material and gate material, such as 10nm HfO and 50nm IZO, are sequentially grown on the exposed surface of the current stacked structure, so that the IZO material fills the first via and the first groove, as shown in Figure 17. In actual fabrication, the gate material can also be one or more of titanium nitride, tungsten, and indium zinc oxide stacked layers; this embodiment does not impose specific limitations. It should be noted that the gate dielectric material and the insulating material in this embodiment are the same, both being HfO. Therefore, when growing the gate dielectric material, for the portion of the structure sidewall, the newly grown gate dielectric material forms a new insulating layer with the already formed insulating layer. Therefore, this embodiment limits the HfO layer located at the first via and the first groove to be the gate dielectric layer, and the HfO layer located on the structure sidewall to be the insulating layer.

[0071] S9, etching the gate material to form the first and second common word lines in the vertical direction.

[0072] After step S8 is executed, the first via is filled with gate material, forming a conductive structure that penetrates each independent structure. In step S9, the gate material is wet-etched, retaining only the gate material inside the first via and the gate material outside the insulating layer corresponding to the position of the first via, as shown in Figure 18. This allows the gate material inside the first via to form the first gate electrode of the transistor in each memory cell layer, and a first common word line is formed based on the integrated structure between multiple first gate electrodes. Similarly, the gate material outside the insulating layer corresponding to the position of the first via forms the second gate electrode of the transistor in each memory cell layer, and a second common word line is formed based on the integrated structure between multiple second gate electrodes. Figure 19 is a cross-sectional schematic diagram at AA' in Figure 18, showing only the formation of the first common word line.

[0073] S10, the source sacrificial layer on the side of the independent structure away from the common structure is etched to form a second groove, and a capacitor is formed in the second groove.

[0074] After step S9 is executed, the transistor fabrication of each memory cell in the three-dimensional memory device is completed. Following capacitor fabrication, the fabrication of the three-dimensional memory device is finished. It should be noted that before fabricating the capacitors, the connection portion of the first and second common word lines fabricated in S9 can be etched to form the first and second common word line structure shown in Figure 20. This facilitates capacitor fabrication, and the electrical connections of the first and second common word lines of the corresponding memory cells can then be made in conjunction with subsequent hierarchical design.

[0075] During capacitor fabrication, the source sacrificial layer on the side of the independent structure furthest from the common structure is first etched laterally using a wet etching process to form a second groove. The depth of this second groove in the second horizontal direction should be less than the length of the source sacrificial layer in the second horizontal direction, but its depth should not be too small. Sufficient space should be etched to facilitate the layer fabrication of the capacitor and maximize the capacitor size. Subsequently, conductive and insulating materials are grown sequentially on the currently exposed surface of the stacked structure to serve as one side electrode plate and capacitor dielectric layer. However, at this point, the sum of the growth thicknesses of the conductive and insulating materials is less than half the height of the second groove in the vertical direction to reserve sufficient space for the fabrication of the other side electrode plate of the capacitor. Then, conductive material is grown on the currently exposed surface of the capacitor dielectric layer until the second groove is completely filled to form the other side electrode plate of the capacitor, as shown in Figure 21. It should be noted that the two conductive material layers in Figure 21 are represented by pink and orange, respectively, to characterize the two plates of the capacitor. The dark gray layer between the pink and orange layers represents the capacitor dielectric layer. In some embodiments, the structure shown in Figure 21 can be chemically mechanically polished to remove excess conductive and insulating materials at the top, forming a hierarchical structure as shown in Figure 22. Supporting materials are then used to fill the gaps between the storage cells. Finally, insulating materials are used to encapsulate the overall storage device structure and connect it to external circuits. This completes the fabrication process of the three-dimensional storage device. It should be noted that both the supporting and insulating materials can be insulating materials.

[0076] This embodiment introduces three-dimensional stacking technology to vertically stack memory cells to increase the overall density of the memory, meeting the needs of future high-density storage applications. It also simplifies the manufacturing process, reduces costs during memory device fabrication, and utilizes lateral etching and thin-film deposition techniques to fabricate the memory cell structure, effectively avoiding high-precision photolithography and further simplifying the process. Simultaneously, it uses wide-bandgap indium gallium zinc oxide as the channel material to reduce transistor leakage current, thereby extending data retention time, reducing refresh rate, and lowering device power consumption. Furthermore, the memory cell structure is optimized through a dual-gate design with a vertical ring channel, effectively improving the gate control capability of the transistors while shielding signal crosstalk between adjacent memory cells, achieving performance optimization and stability improvement of the memory device.

[0077] Based on the same inventive concept, the third embodiment of this disclosure provides an electronic device that includes at least the three-dimensional storage device provided in the first embodiment of this disclosure, so as to realize high-density, low-power data storage and help the electronic device to operate stably and with low power consumption.

[0078] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this disclosure, and are not intended to limit them. Although this disclosure has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this disclosure.

Claims

1. A three-dimensional memory device, comprising: At least including: Substrate; Multiple memory cell layers and multiple isolation layers are alternately stacked in a vertical direction on the substrate surface, and the orthographic projection of each memory cell layer on the substrate surface completely overlaps, wherein the vertical direction is perpendicular to the substrate surface; Each of the memory cell layers includes at least: a common bit line extending along a first horizontal direction, and a plurality of memory cells symmetrically distributed about the common bit line, the memory cells extending along a second horizontal direction and being insulated from each other, the first horizontal direction and the second horizontal direction being parallel to the substrate surface, and the first horizontal direction and the second horizontal direction being perpendicular to each other; The memory cell includes at least a transistor on the side closest to the common bit line and a capacitor on the side furthest from the common bit line. The transistor includes at least a first gate electrode, a second gate electrode, a source electrode, a drain electrode, a channel layer, a gate dielectric layer, and an insulating layer. The first gate electrode extends vertically and penetrates the memory cell layer to which the transistor belongs and its adjacent isolation layer, and is electrically connected to the first gate electrodes of transistors at the same position in other memory cell layers, forming a first common word line. The gate dielectric layer surrounds the first gate electrode with an imaginary line in the vertical direction as its axis, and the channel layer surrounds the gate dielectric layer with the imaginary line as its axis. The drain electrode is located between the channel layer and the... The first common word line is electrically connected to the common bit line. The source is located between the channel layer and the capacitor and is electrically connected to one side electrode plate of the capacitor. The insulating layer covers the sidewalls of the channel layer, the source, and the drain. The second gate electrode is located outside the insulating layer. The orthographic projection of the second gate electrode on an imaginary plane perpendicular to the first horizontal direction completely covers the orthographic projection of the channel layer on the imaginary plane. The second gate electrodes of transistors in the same position in all memory cell layers penetrate the adjacent isolation layer of their respective memory cell layers to achieve electrical connection and form a second common word line. The first common word line and the second common word line are electrically connected.

2. The three-dimensional memory device of claim 1, wherein, The channel layer is prepared based on a metal oxide material.

3. The three-dimensional memory device of claim 1, wherein, The second gate electrode includes a first electrode plate and a second electrode plate, which are symmetrically distributed on the outside of the insulating layer, and the orthographic projections of the first electrode plate and the second electrode plate on the imaginary surface coincide.

4. The three-dimensional memory device of claim 1, wherein, The capacitor is a cylindrical capacitor.

5. The three-dimensional memory device of any one of claims 1-4, wherein, The adjacent storage cells are filled with a support material, which is insulated.

6. A method for manufacturing the three-dimensional memory device according to any one of claims 1 to 5, wherein include: Provide a substrate; An isolation layer and a sacrificial layer are alternately fabricated on the surface of the substrate to obtain a stacked structure; The stacked structure is etched along the vertical direction to obtain a common structure extending along the first horizontal direction and multiple independent structures extending along the second horizontal direction. An insulating material is grown on the outer surface of the common structure and all the independent structures, and the insulating material located on top of the common structure and all the independent structures is removed to form an insulating layer; A first through hole is opened at one end of each independent structure near the common structure, and the width of the first through hole in the first horizontal direction is smaller than the width of the independent structure in the first horizontal direction. The exposed sacrificial layer is laterally etched in the first through hole to form a first groove, so that each sacrificial layer in the independent structure is broken at the first groove, forming a drain sacrificial layer on the side close to the common structure and a source sacrificial layer on the side away from the common structure. Growing channel material on the currently exposed surface of the first groove to form a channel layer; A gate dielectric material and a gate material are grown on the currently exposed surface of the stacked structure, and the gate material is made to fill the first via. The gate material is etched to form a first common word line and a second common word line in the vertical direction; The source sacrificial layer on the side of the independent structure away from the common structure is etched to form a second groove, and a capacitor is formed in the second groove.

7. The production method according to claim 6, characterized by, The sacrificial layer and the gate material are each independently a stack of one or more material layers selected from titanium nitride, tungsten, and indium zinc oxide.

8. The preparation method according to claim 6, characterized in that, The step of etching the source sacrificial layer on the side of the independent structure away from the common structure to form a second groove, and forming a capacitor within the second groove, includes: A second groove is formed by lateral etching of the source sacrificial layer on the side of the independent structure away from the common structure. The depth of the second groove in the second horizontal direction is less than the length of the source sacrificial layer in the second horizontal direction. Conductive and insulating materials are sequentially grown on the currently exposed surface of the stacked structure to form one side electrode plate and capacitor dielectric layer of the capacitor. The sum of the growth thicknesses of the conductive and insulating materials is less than half the height of the second groove in the vertical direction. A conductive material is grown on the currently exposed surface of the capacitor dielectric layer until the second groove is completely filled to form the other electrode plate of the capacitor.

9. The production method according to any one of claims 6 to 8, characterized by, Also includes: The gaps between all the individual structures are filled with a supporting material, which is an insulating material.

10. An electronic device, comprising: It includes at least the three-dimensional storage device as described in any one of claims 1 to 5.