A storage device, a method for forming the same, and an electronic device.

By stacking the core circuit layer and the peripheral circuit layer vertically and connecting them with a hybrid bonding process, the space waste problem caused by the partial overlap between the peripheral circuit and the memory array in the CMOS bonding array architecture is solved, achieving higher integration and storage density.

CN123124352APending Publication Date: 2026-11-17WUHAN SIFANG CHUANGXIN TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611146324.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2026-07-15
Filing Date
2026-07-30
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In existing CMOS bonding array architectures, the peripheral circuits and memory arrays only partially overlap, resulting in wasted chip layout space and limiting the improvement of integration and memory density.

Method used

The core circuit layer and the peripheral circuit layer are stacked vertically, formed separately or together on a single or the same substrate, and vertical interconnection is achieved through a hybrid bonding process to avoid lateral expansion of the circuit area.

Benefits of technology

It increases the area that can be overlapped in the circuit, effectively reduces the overall chip area, improves the integration and storage density of the CBA architecture, and supports differentiated customization of in-memory processors.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123124352A_ABST
    Figure CN123124352A_ABST
Patent Text Reader

Abstract

This disclosure provides a memory device, its fabrication method, and an electronic device. The memory device includes: a core circuit layer configured to control read and write operations of a memory array circuit; and a peripheral circuit layer configured to control data communication, address decoding, and timing management of the memory array circuit. The core circuit layer and the peripheral circuit layer are fabricated separately and stacked vertically. In this disclosure, the core circuit layer and the peripheral circuit layer can be formed on separate substrates or on the same substrate. Different customizable logic processes can be flexibly selected according to performance requirements, supporting differentiated customization of various in-memory processors. Stacking the core circuit layer and the peripheral circuit layer vertically transforms their traditional horizontal side-by-side layout into a vertically stacked structure, thereby increasing the overlapable area of ​​the circuits and thus improving storage density.
Need to check novelty before this filing date? Find Prior Art

Description

Cross-references to related applications

[0001] This disclosure is based on and claims priority to Chinese Patent Application No. 202611054647.1, filed on July 15, 2026, the entire contents of which are incorporated herein by reference. Technical Field

[0002] This disclosure relates to the field of semiconductor technology, and more particularly to a memory device, a method of forming the same, and an electronic device. Background Technology

[0003] Dynamic Random Access Memory (DRAM) devices contain memory array circuitry and Complementary Metal-Oxide-Semiconductor (CMOS) circuitry. The CMOS circuitry primarily consists of core circuitry and peripheral circuitry. Existing CMOS-Bonded Array (CBA) architectures typically employ a two-wafer discrete fabrication approach, where the CMOS circuitry and memory array circuitry are fabricated on two separate wafers, with subsequent interlayer interconnection achieved through bonding processes. Furthermore, in this architecture, the core circuitry and peripheral circuitry are coplanarly arranged.

[0004] While the CBA architecture allows for complete vertical overlap between the core circuitry and the memory array, it is limited by the layout requirements of the core circuitry. The peripheral circuitry can only partially overlap with the memory array, leaving unused chip layout space and resulting in wasted device area. This is not conducive to further improving the integration and storage density of the CBA architecture. Summary of the Invention

[0005] This disclosure provides a memory device, a method for forming the same, and an electronic device. The core circuit layer and the peripheral circuit layer can be formed on separate substrates, or they can be vertically stacked on the same substrate and interconnected. This allows for flexible selection of different customizable logic processes based on performance requirements and avoids lateral expansion of circuit regions, thereby increasing storage density.

[0006] The technical solution of this disclosure embodiment is implemented as follows: This disclosure provides a storage device comprising: a core circuit layer configured to control read and write operations of a storage array circuit; and a peripheral circuit layer configured to control data communication, address decoding, and timing management of the storage array circuit; wherein the core circuit layer and the peripheral circuit layer are stacked vertically.

[0007] In some embodiments of this disclosure, the storage array circuit layer is configured to store data; wherein the storage array circuit layer is stacked along the vertical direction with the core circuit layer and the peripheral circuit layer.

[0008] In some embodiments of this disclosure, along the vertical direction, the memory array circuit layer is bonded to the side of the core circuit layer away from the peripheral circuit layer; or, along the vertical direction, the core circuit layer and the peripheral circuit layer are bonded through the memory array circuit layer; wherein the core circuit layer, the peripheral circuit layer, and the memory array circuit layer are formed on different substrates.

[0009] In some embodiments of this disclosure, the memory array circuit layer and the core circuit layer are formed on the same substrate and interconnected along the vertical direction; the peripheral circuit layer is formed on another substrate; along the vertical direction, the side of the memory array circuit layer away from the core circuit layer is bonded to the peripheral circuit layer; or, along the vertical direction, the side of the core circuit layer away from the memory array circuit layer is bonded to the peripheral circuit layer.

[0010] In some embodiments of this disclosure, the memory array circuit layer and the peripheral circuit layer are formed on the same substrate and interconnected along the vertical direction; the core circuit layer is formed on another substrate; and along the vertical direction, the side of the memory array circuit layer away from the peripheral circuit layer is bonded to the core circuit layer.

[0011] In some embodiments of this disclosure, the core circuit layer and the peripheral circuit layer are formed on the same substrate and interconnected along the vertical direction; the memory array circuit is formed on another substrate; along the vertical direction, the side of the core circuit layer away from the peripheral circuit layer is bonded to the memory array circuit layer.

[0012] In some embodiments of this disclosure, the core circuit layer, the memory array circuit layer, and the peripheral circuit layer are formed on the same substrate; the core circuit layer, the memory array circuit layer, and the peripheral circuit layer are stacked sequentially along the vertical direction; or, the memory array circuit layer, the core circuit layer, and the peripheral circuit layer are stacked sequentially along the vertical direction; or, the peripheral circuit layer, the core circuit layer, and the memory array circuit layer are stacked sequentially along the vertical direction; or, the peripheral circuit layer, the memory array circuit layer, and the core circuit layer are stacked sequentially along the vertical direction.

[0013] In some embodiments of this disclosure, the core circuit layer or the peripheral circuit layer includes: an in-memory computing circuit, which is at least configured to perform computational processing on data read from or written by the memory array circuit layer under the control of the core circuit or the peripheral circuit.

[0014] This disclosure also provides a method for fabricating a memory device, the method comprising: stacking the core circuit layer and the peripheral circuit layer in a vertical direction; wherein the core circuit layer is used to control the read and write operations of the memory array circuit; and the peripheral circuit layer is used to control the data communication, address decoding and timing management of the memory array circuit.

[0015] In some embodiments of this disclosure, before stacking the core circuit layer and the peripheral circuit layer in a vertical direction, the method for fabricating the memory device further includes: forming a memory array circuit layer on another substrate; or forming a memory array circuit layer on the core circuit layer; or forming a memory array circuit layer on the peripheral circuit layer; wherein the memory array circuit layer is used to store data.

[0016] In some embodiments of this disclosure, stacking the core circuit layer and the peripheral circuit layer in a vertical direction includes: stacking the core circuit layer and the peripheral circuit layer in a vertical direction through a memory array circuit layer; or, bonding the side of the core circuit layer away from the memory array circuit layer to the peripheral circuit layer in the vertical direction; or, stacking the memory array circuit layer, the core circuit layer, and the peripheral circuit layer sequentially in the vertical direction.

[0017] In some embodiments of this disclosure, after stacking the core circuit layer and the peripheral circuit layer in a vertical direction, the method for fabricating the memory device further includes: bonding the side of the core circuit layer away from the peripheral circuit layer to the memory array circuit layer in the vertical direction; or, forming the memory array circuit layer on the side of the core circuit layer away from the peripheral circuit layer.

[0018] In some embodiments of this disclosure, all bonding methods in the storage device include hybrid bonding.

[0019] This disclosure also provides an electronic device, including a storage device as described in any of the above embodiments, or a storage device formed according to the method for preparing the storage device according to any of the above embodiments.

[0020] The embodiments disclosed herein offer the following advantages: The core circuit layer and the peripheral circuit layer can be formed on separate substrates or on the same substrate. This allows for flexible selection of different customizable logic processes based on performance requirements. This flexibility facilitates the construction of an open ecosystem and supports differentiated customization of various in-memory processors. Stacking the core circuit layer and the peripheral circuit layer vertically transforms their traditional horizontal side-by-side layout into a vertically stacked structure. This avoids lateral expansion of the circuit area, significantly increases the overlapable area of ​​the circuit, effectively reduces the overall chip area, and thus improves the integration and memory density of the CBA architecture. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this disclosure, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort, wherein: Figure 1 This is a schematic diagram of the structure of the storage device provided in the embodiments of this disclosure. Figure 1 ; Figure 2 This is a schematic diagram of the structure of the storage device provided in the embodiments of this disclosure. Figure 2 ; Figure 3 This is a schematic diagram of the structure of the storage device provided in the embodiments of this disclosure. Figure 3 ; Figure 4 This is a schematic diagram of the structure of the storage device provided in the embodiments of this disclosure. Figure 4 ; Figure 5 This is a schematic diagram of the structure of the storage device provided in the embodiments of this disclosure. Figure 5 ; Figure 6 This is a schematic diagram of the structure of the storage device provided in the embodiments of this disclosure. Figure 6 ; Figure 7 Schematic diagram of the structure of the method for fabricating the storage device provided in the embodiments of this disclosure Figure 1 ; Figure 8 Schematic diagram of the structure of the method for fabricating the storage device provided in the embodiments of this disclosure Figure 2 ; Figure 9 Schematic diagram of the structure of the method for fabricating the storage device provided in the embodiments of this disclosure Figure 3 ; Figure 10 Schematic diagram of the structure of the method for fabricating the storage device provided in the embodiments of this disclosure Figure 4 ; Figure 11 Schematic diagram of the structure of the method for fabricating the storage device provided in the embodiments of this disclosure Figure 5 ; Figure 12 Schematic diagram of the structure of the method for fabricating the storage device provided in the embodiments of this disclosure Figure 6 ; Figure 13 Schematic diagram of the structure of the method for fabricating the storage device provided in the embodiments of this disclosure Figure 7 ; Figure 14 Schematic diagram of the structure of the method for fabricating the storage device provided in the embodiments of this disclosure Figure 8 ; Figure 15 Schematic diagram of the structure of the method for fabricating the storage device provided in the embodiments of this disclosure Figure 9 ; Figure 16 Schematic diagram of the structure of the method for fabricating the storage device provided in the embodiments of this disclosure Figure 10 ; Figure 17 Schematic diagram of the structure of the method for fabricating the storage device provided in the embodiments of this disclosure Figure 10 one; Figure 18 Schematic diagram of the structure of the method for fabricating the storage device provided in the embodiments of this disclosure Figure 10 two; Figure 19 Schematic diagram of the structure of the method for fabricating the storage device provided in the embodiments of this disclosure Figure 10 three; Figure 20 Schematic diagram of the structure of the method for fabricating the storage device provided in the embodiments of this disclosure Figure 10 Four; Figure 21 Schematic diagram of the structure of the method for fabricating the storage device provided in the embodiments of this disclosure Figure 10 five. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the specific technical solutions of the invention will be further described in detail below with reference to the accompanying drawings of the embodiments of this disclosure. The following embodiments are used to illustrate the embodiments of this disclosure, but are not intended to limit the scope of the embodiments of this disclosure.

[0023] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0024] In the following description, the terms “first / second / third” are used merely to distinguish similar objects and do not represent a specific ordering of objects. It is understood that “first / second / third” may be interchanged in a specific order or sequence where permitted, so that the embodiments of this disclosure described herein can be implemented in an order other than that illustrated or described herein.

[0025] In this document, when a layer / component is referred to as being "above" another layer / component, the layer / component may be directly above the other layer / component, or there may be an intermediate layer / component between them. Furthermore, in one orientation, a layer / component is "above" another layer / component; when the orientation is reversed, the layer / component may be "below" the other layer / component.

[0026] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which embodiments of this disclosure belong. The terminology used herein is for descriptive purposes only and is not intended to limit the scope of this disclosure.

[0027] Complementary Metal-Oxide-Semiconductor (CMOS) circuits in Dynamic Random Access Memory (DRAM) devices typically consist of core circuits and peripheral circuits, and represent the most fundamental and core manufacturing technology in integrated circuits. In a CMOS Bonded Array (CBA) architecture, the CMOS circuits and the memory array circuits are usually fabricated on two separate wafers and then connected via bonding. The core circuits and peripheral circuits are positioned on the same horizontal plane. While this arrangement allows for complete overlap between the core and memory array circuits, the peripheral circuits only partially overlap, resulting in wasted device area.

[0028] Based on this, embodiments of this disclosure provide a storage device, referring to... Figures 1 to 4 As shown, the storage device includes: a core circuit layer 10, which is configured at least to control read and write operations of the storage array circuit.

[0029] It should be noted that, referring to Figures 1 to 4As shown, the core circuit layer 10 includes core circuitry, which mainly comprises modules such as bit line sense amplifiers (BLSA) and word line drivers (SWD). The BLSA senses and amplifies the weak charge signals in the memory cells, converting them into clear logic levels; the SWD selects and drives specific word lines to open the corresponding memory cell rows. Through these circuits closely integrated with the memory array, the core circuitry enables fast and accurate access to the memory cells, making it the most critical front-end control component in the data read / write process.

[0030] In this embodiment of the disclosure, reference is made to Figures 1 to 4 As shown, the peripheral circuit layer 20 is configured at least to control data communication, address decoding, and timing management of the memory array circuitry.

[0031] It should be noted that, referring to Figures 1 to 4 As shown, the peripheral circuit layer 20 includes peripheral circuits, which mainly include modules such as an address decoder, a data buffer, a clock generator, a voltage pump (charge pump), and command control logic. The decoder converts externally input addresses into row / column selection signals; the buffer temporarily stores data to be read or written and matches the external bus speed; the clock generator provides synchronization timing for the internal modules; the voltage pump generates the high voltage required for read, write, and erase operations (especially in flash memory); and the command control logic parses instructions from the host and coordinates the core circuitry, memory array, and other modules to operate sequentially. The peripheral circuits ensure that the memory chip can interact correctly with the external system and guarantee the orderly execution of internal operations.

[0032] In this embodiment of the disclosure, reference is made to Figures 1 to 4 As shown, the core circuit layer 10 and the peripheral circuit layer 20 are stacked in the vertical direction.

[0033] It should be noted that, referring to Figures 1 to 4 As shown, the core circuit layer and the peripheral circuit layer can be formed on separate substrates or on the same substrate. This allows for flexible selection of different customizable logic processes based on performance requirements. This flexibility helps build an open ecosystem and supports differentiated customization of various in-memory processors. The substrate can include silicon-based substrates, such as wafers, or substrates formed from other semiconductor materials.

[0034] Furthermore, the core circuit layer 10 and the peripheral circuit layer 20 are bonded vertically, transforming their traditional horizontal side-by-side layout into a vertically stacked structure. This avoids lateral expansion of the circuit area, significantly increases the overlapable area of ​​the circuit, effectively reduces the overall chip area, and thus improves the integration and storage density of the CBA architecture.

[0035] In some embodiments of this disclosure, reference is made to Figures 1 to 4 As shown, the storage array circuit layer 30 is configured to store at least data; wherein the storage array circuit layer 30 is stacked vertically with the core circuit layer 10 and the peripheral circuit layer 20.

[0036] It should be noted that, referring to Figures 1 to 4 As shown, the memory array circuit layer 30 is the core functional layer in a semiconductor memory chip used for actual data storage. It consists of a large number of repeating memory cells (such as transistors or capacitor cells) arranged in a matrix of rows (word lines) and columns (bit lines), responsible for storing binary information in a high-density manner. In the CBA architecture, this layer is independently fabricated on a dedicated "array substrate," does not contain any CMOS logic circuits, and is vertically interconnected with the CMOS circuit layer through subsequent bonding processes.

[0037] It should also be noted that, referring to Figures 1 to 4 As shown, the storage array circuit layer 30, the core circuit layer 10, and the peripheral circuit layer 20 are stacked vertically. This allows for an increase in the overlap area between the peripheral circuits and the storage array circuits without affecting the overlap area between the core circuit layer 10 and the storage array circuits, thereby significantly increasing storage density.

[0038] In some embodiments of this disclosure, reference is made to Figure 1 and Figure 2 As shown, along the vertical direction, the memory array circuit layer 30 is bonded to the side of the core circuit layer 10 away from the peripheral circuit layer 20; or, along the vertical direction, the core circuit layer 10 and the peripheral circuit layer 20 are bonded through the memory array circuit layer 30; wherein, the core circuit layer 10, the peripheral circuit layer 20 and the memory array circuit layer 30 are formed on different substrates.

[0039] It should be noted that, referring to Figure 1 and Figure 2As shown, the memory array circuit layer 30, core circuit layer 10, and peripheral circuit layer 20 are fabricated on three independent substrates and interconnected. For example, the core circuit layer 10 is formed on the first substrate, the peripheral circuit layer 20 is formed on the second substrate, and the memory array circuit layer 30 is formed on the third substrate. This independent design offers good flexibility, allowing for free combination with CMOS substrates of different generations, forming an open ecosystem and meeting customized needs such as in-memory computing. The connection methods for the memory array circuit layer 30, core circuit layer 10, and peripheral circuit layer 20 include, but are not limited to, three-dimensional interconnection processes using multiple substrates; among them, hybrid bonding is a preferred bonding process. Hybrid bonding is a bumpless bonding process where dielectric layers and embedded copper pads are fabricated on two substrates respectively. After planarization and activation, they are aligned, bonded, and annealed at low temperature. The dielectric and copper pads are bonded simultaneously, completing the mechanical fixation and electrical interconnection of the substrates at the same interface, achieving high-density vertical interconnection. The transistor channels in the memory array circuit layer 30 can use single-crystal silicon or indium gallium zinc oxide (IGZO) materials. The transistor channels in the core circuit layer 10 and the peripheral circuit layer 20 can use monocrystalline silicon or new materials. Specific new materials may include: thin-film monocrystalline silicon, molybdenum disulfide (MoS2), bismuth oxyselenide (Bi2O2Se), indium selenide (InSe), and tungsten diselenide (WSe2).

[0040] In some embodiments of this disclosure, reference is made to Figure 3 and Figure 4 As shown, the memory array circuit layer 30 and the core circuit layer 10 are formed on the same substrate (such as the first substrate) and are interconnected in the vertical direction; the peripheral circuit layer 20 is formed on another substrate (such as the second substrate); in the vertical direction, the side of the memory array circuit layer 30 away from the core circuit layer 10 is bonded to the peripheral circuit layer 20; or, in the vertical direction, the side of the core circuit layer 10 away from the memory array circuit layer 30 is bonded to the peripheral circuit layer 20.

[0041] It should be noted that, referring to Figure 3 and Figure 4As shown, the memory array circuit layer 30 and the core circuit layer 10 are formed on a first substrate. This integrated manufacturing method eliminates the need for complex substrate bonding processes, resulting in a relatively simple process flow and lower mass production costs. Specifically, the memory array circuit layer 30 is formed on the surface of the core circuit layer 10 through processes such as deposition, etching, and filling. In this case, the transistor channels in the memory array circuit layer 30 can use indium gallium zinc oxide (IGZO) material, and the transistor channels in the core circuit layer 10 can use monocrystalline silicon or new materials (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2). Alternatively, the core circuit layer 10 can be formed on the surface of the memory array circuit layer 30 using semiconductor manufacturing processes. In this case, the transistor channels in the memory array circuit layer 30 can use monocrystalline silicon or indium gallium zinc oxide (IGZO) material, and the transistor channels in the core circuit layer 10 can use new materials (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2). The transistor channels in the peripheral circuit layer 20 can use monocrystalline silicon or new materials (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2).

[0042] In some embodiments of this disclosure, reference is made to Figure 2 and Figure 3 As shown, the memory array circuit layer 30 and the peripheral circuit layer 20 are formed on the same substrate (such as the second substrate) and are interconnected in the vertical direction; the core circuit layer 10 is formed on another substrate (such as the first substrate); in the vertical direction, the side of the memory array circuit layer 30 away from the peripheral circuit layer 20 is bonded to the core circuit layer 10.

[0043] It should be noted that, referring to Figure 2 and Figure 3 As shown, a memory array circuit layer 30 and a peripheral circuit layer 20 are formed on a second substrate. When the memory array circuit layer 30 and the peripheral circuit layer 20 are sequentially formed on the second substrate, the transistor channels in the memory array circuit layer 30 can use single-crystal silicon or indium gallium zinc oxide (IGZO) material; the transistor channels in the peripheral circuit layer 20 can use new materials (e.g., thin-film single-crystal silicon, MoS2, Bi2O2Se, InSe, WSe2). When the peripheral circuit layer 20 and the memory array circuit layer 30 are sequentially formed on the second substrate, the transistor channels in the memory array circuit layer 30 can use indium gallium zinc oxide (IGZO) material; the transistor channels in the peripheral circuit layer 20 can use single-crystal silicon or new materials (e.g., thin-film single-crystal silicon, MoS2, Bi2O2Se, InSe, WSe2). A core circuit layer is formed on a first substrate, and the transistor channels in the core circuit layer 10 can use single-crystal silicon or new materials (e.g., thin-film single-crystal silicon, MoS2, Bi2O2Se, InSe, WSe2).

[0044] In some embodiments of this disclosure, reference is made to Figure 1 and Figure 4 As shown, the core circuit layer 10 and the peripheral circuit layer 20 are formed on the same substrate (such as the first substrate) and are interconnected in the vertical direction; the memory array circuit layer 30 is formed on another substrate (such as the second substrate); in the vertical direction, the side of the core circuit layer 10 away from the peripheral circuit layer 20 is bonded to the memory array circuit layer 30.

[0045] It should be noted that, referring to Figure 1 and Figure 4 As shown, the core circuit layer 10 and the peripheral circuit layer 20 are formed on the first substrate. (Referring to...) Figure 1 As shown, when a core circuit layer 10 and a peripheral circuit layer 20 are sequentially formed on a first substrate, the transistor channel in the core circuit layer 10 can use monocrystalline silicon or new materials (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2); the transistor channel in the peripheral circuit layer 20 can use new materials (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2). (Refer to...) Figure 4 As shown, when the peripheral circuit layer 20 and the core circuit layer 10 are sequentially formed on the second substrate, the transistor channel in the core circuit layer 10 can use a new material (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2); the transistor channel in the peripheral circuit layer 20 can use monocrystalline silicon or a new material (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2). The memory array circuit layer is formed on the third substrate, and the transistor channel in the memory array circuit layer 30 can use monocrystalline silicon or indium gallium zinc oxide (IGZO) material.

[0046] In some embodiments of this disclosure, reference is made to Figures 1 to 4 As shown, the core circuit layer 10, the memory array circuit layer 30, and the peripheral circuit layer 20 are formed on the same substrate; the core circuit layer 10, the memory array circuit layer 30, and the peripheral circuit layer 20 are stacked sequentially in the vertical direction; or, the memory array circuit layer 30, the core circuit layer 10, and the peripheral circuit layer 20 are stacked sequentially in the vertical direction; or, the peripheral circuit layer 20, the core circuit layer 10, and the memory array circuit layer 30 are stacked sequentially in the vertical direction; or, the peripheral circuit layer 20, the memory array circuit layer 30, and the core circuit layer 10 are stacked sequentially in the vertical direction.

[0047] It should be noted that, referring to Figures 1 to 4As shown, the core circuit layer 10, the memory array circuit layer 30, and the peripheral circuit layer 20 can be formed on the same substrate. The connection between the core circuit layer 10, the memory array circuit layer 30, and the peripheral circuit layer 20 can utilize a single-substrate manufacturing process. For example, a semiconductor thin film (such as IGZO) is deposited on the surface of the core circuit layer 10, and then the memory array circuit layer 30 is formed on the IGZO surface using a semiconductor manufacturing process. A new material thin film (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2) is then transferred onto the memory array circuit layer 30, and the peripheral circuit layer 20 is formed using a semiconductor manufacturing process. Thus, a stacked memory device consisting of the core circuit layer 10, the memory array circuit layer 30, and the peripheral circuit layer 20 is formed on a single substrate. In this case, the transistor channel of the core circuit layer 10 can use monocrystalline silicon or a new material (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2). For example, based on the memory circuit layer 30, a new material thin film (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2) is transferred onto the memory array circuit layer 30 to form a core circuit layer 10 through semiconductor manufacturing processes; then, a new material thin film (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2) is transferred onto the core circuit layer 10 to form a peripheral circuit layer 20 through semiconductor manufacturing processes; finally, a stacked memory device consisting of the memory circuit layer 30, the core circuit layer 10, and the peripheral circuit layer 20 is formed on a substrate; at this time, the transistor channel of the memory array circuit layer 30 can use monocrystalline silicon or indium gallium zinc oxide (IGZO) material.

[0048] Specifically, refer to Figure 1 and Figure 3 As shown, a peripheral circuit layer 20 is formed on a second substrate, and a core circuit layer 10 and a memory array circuit layer 30 are sequentially formed on the peripheral circuit layer 20. Alternatively, when the memory array circuit layer 30 and the core circuit layer 10 are sequentially formed on the peripheral circuit layer 20, the transistor channel in the peripheral circuit layer 20 can use monocrystalline silicon or new materials (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2), the transistor channel in the memory array circuit layer 30 uses indium gallium zinc oxide (IGZO) material, and the transistor channel in the core circuit layer 10 uses new materials (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2).

[0049] Furthermore, referring to Figure 2As shown, when a core circuit layer 10 is formed on a first substrate, and a memory array circuit layer 30 and a peripheral circuit layer 20 are sequentially formed on the core circuit layer 10, the transistor channel in the core circuit layer 10 can use single-crystal silicon or new materials (e.g., thin-film single-crystal silicon, MoS2, Bi2O2Se, InSe, WSe2), the transistor channel in the memory array circuit layer 30 uses indium gallium zinc oxide (IGZO) material, and the transistor channel in the peripheral circuit layer 20 uses new materials (e.g., thin-film single-crystal silicon, MoS2, Bi2O2Se, InSe, WSe2).

[0050] Furthermore, referring to Figure 4 As shown, when a memory array circuit layer 30 is formed on a third substrate, and a core circuit layer 10 and a peripheral circuit layer 20 are sequentially formed on the memory array circuit layer 30, the transistor channels in the memory array circuit layer 30 use single-crystal silicon or indium gallium zinc oxide (IGZO) material, while the transistor channels in the core circuit layer 10 and the peripheral circuit layer 20 use new materials (e.g., thin-film single-crystal silicon, MoS2, Bi2O2Se, InSe, WSe2).

[0051] It should also be noted that when the core circuit layer 10, the peripheral circuit layer 20, and the memory array circuit layer 30 are not formed through a separate substrate, new materials must be used instead of single-crystal silicon to form the transistor channels in the core circuit layer 10, the peripheral circuit layer 20, and the memory array circuit layer 30. This can avoid the problem of low carrier mobility and ensure the conductivity of the device.

[0052] In some embodiments of this disclosure, reference is made to Figure 5 and Figure 6 As shown, the core circuit layer 10 or the peripheral circuit layer 20 includes: in-memory computing (PIM) circuitry 40, which is configured at least to perform computational processing on data read from or written to the memory array circuit layer 30 under the control of the core circuitry or the peripheral circuitry.

[0053] It should be noted that, referring to Figure 5 and Figure 6 As shown, by introducing PIM circuit 40 into the core circuit layer 10 or the peripheral circuit layer 20, computational operations can be completed directly inside the CMOS circuit, greatly reducing the round-trip data transfer between the processor and the memory array circuit, thereby significantly reducing access latency and data path power consumption.

[0054] This disclosure also provides a method for fabricating a storage device, the method comprising step S101.

[0055] Step S101: Stack the core circuit layer and the peripheral circuit layer vertically.

[0056] In this embodiment of the disclosure, reference is made to Figure 7 As shown, the core circuit layer 10 and the peripheral circuit layer 20 are stacked vertically, transforming their traditional horizontal side-by-side layout into a vertical stacking structure. This avoids lateral expansion of the circuit area, significantly increases the overlapable area of ​​the circuit, effectively reduces the overall chip area, and thus improves the integration and storage density of the CBA architecture.

[0057] It should be noted that the core circuit layer 10 is used to control the read and write operations of the memory array circuit, while the peripheral circuit layer 20 is used to control data communication, address decoding, and timing management of the memory array circuit. The core circuit layer 10 and the peripheral circuit layer 20 can be formed on separate substrates or on the same substrate. This allows for flexible selection of different customizable logic processes based on performance requirements. This flexibility helps build an open ecosystem and supports differentiated customization of various in-memory processors. The substrate is the basic substrate material for semiconductor manufacturing, typically high-purity single-crystal silicon, in the form of a circular thin film. It should be noted that when the core circuit layer 10 and the peripheral circuit layer 20 are not formed on separate substrates, new materials (e.g., thin-film single-crystal silicon, MoS2, Bi2O2Se, InSe, WSe2) must be used instead of single-crystal silicon to form the transistor channels in the core circuit layer 10 and the peripheral circuit layer 20. This avoids the problem of low carrier mobility and ensures the conductivity of the device.

[0058] It should also be noted that the core circuit layer and the peripheral circuit layer can be connected vertically by bonding or by single-substrate manufacturing technology. For example, a semiconductor thin film (such as IGZO) is deposited on the surface of the core circuit layer 10, and then the memory array circuit layer 30 is formed on the surface of the IGZO semiconductor thin film by semiconductor manufacturing process. Then, a new material thin film (e.g., thin film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2) is transferred onto the memory array circuit layer 30, and the peripheral circuit layer 20 is formed by semiconductor manufacturing process. Thus, a stacked memory device consisting of core circuit layer 10, memory circuit layer 30, and peripheral circuit layer 20 is formed on a single substrate. At this time, the transistor channel of the core circuit layer 10 can use monocrystalline silicon or new materials (e.g., thin film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2). For example, based on the memory circuit layer 30, a new material thin film (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2) is transferred onto the memory array circuit layer 30 to form a core circuit layer 10 through semiconductor manufacturing processes; then, a new material thin film (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2) is transferred onto the core circuit layer 10 to form a peripheral circuit layer 20 through semiconductor manufacturing processes; finally, a stacked memory device consisting of the memory circuit layer 30, the core circuit layer 10, and the peripheral circuit layer 20 is formed on a substrate; at this time, the transistor channel of the memory array circuit layer 30 can use monocrystalline silicon or indium gallium zinc oxide (IGZO) material.

[0059] In some embodiments of this disclosure, prior to S101, the method for fabricating the storage device may further include steps S1011-S1012.

[0060] Step S1011: Form a core circuit layer on the first substrate.

[0061] In this embodiment of the present disclosure, a core circuit layer 10 is formed on a first substrate. Thus, the transistor channels in the core circuit layer 10 can use monocrystalline silicon or new materials (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2) to avoid the problem of low carrier mobility and ensure the conductivity of the device.

[0062] In some embodiments of this disclosure, forming the peripheral circuit layer 20 may specifically include step S1012.

[0063] Step S1012: Form a peripheral circuit layer on the second substrate.

[0064] In this embodiment of the disclosure, when the peripheral circuit layer 20 is formed on the second substrate, the transistor channel in the peripheral circuit layer 20 can use single crystal silicon or new materials (e.g., thin film single crystal silicon, MoS2, Bi2O2Se, InSe, WSe2), which can also avoid the problem of low carrier mobility and ensure the conductivity of the device.

[0065] In some embodiments of this disclosure, the specific steps of the method for fabricating the storage device prior to step S101 may further include step S201.

[0066] Step S201: Form a memory array circuit layer on another substrate; or, form a memory array circuit layer on a core circuit layer; or, form a memory array circuit layer on a peripheral circuit layer.

[0067] In this embodiment of the disclosure, reference is made to Figure 10 , Figure 11 and Figure 21 As shown, a memory array circuit layer 30 is formed on another substrate (such as a third substrate); or, referring to... Figure 12 and Figure 13 As shown, a memory array circuit layer 30 is formed on the core circuit layer 10; or, referring to... Figure 14 and Figure 15 As shown, a memory array circuit layer 30 is formed on the peripheral circuit layer 20; wherein, the memory array circuit layer 30 is used to store data. The memory array circuit layer 30 can be formed on a third substrate, or on a first substrate or a second substrate, which can be flexibly selected according to actual process requirements.

[0068] In some embodiments of this disclosure, step S101 may include step S301.

[0069] Step S301: Stack the core circuit layer and the peripheral circuit layer vertically through the storage array circuit layer.

[0070] In this embodiment of the disclosure, one selectable stacking method is described below. Figure 11As shown, when a core circuit layer 10 is formed on a first substrate, a peripheral circuit layer 20 is formed on a second substrate, and a memory array circuit layer 30 is formed on a third substrate, the memory array circuit layer 30 is bonded to the core circuit layer 10 in a vertical direction via the memory array circuit layer 30. That is, the memory array circuit layer 30 is formed separately on the third substrate, and then the memory array circuit layer 30 is bonded to the core circuit layer 10 and the peripheral circuit layer 20 on both sides in the vertical direction, respectively. This independent design also has good flexibility and can meet customized needs such as in-memory computing. The transistor channels in the memory array circuit layer 30 can use single-crystal silicon or indium gallium zinc oxide (IGZO) materials. The transistor channels in the core circuit layer 10 and the peripheral circuit layer 20 can use single-crystal silicon or new materials (e.g., thin-film single-crystal silicon, MoS2, Bi2O2Se, InSe, WSe2).

[0071] Furthermore, another alternative stacking method is described in reference... Figure 12 As shown, a peripheral circuit layer 20 is formed on a second substrate, and a memory array circuit layer 30 and a core circuit layer 10 are sequentially formed on a third substrate; then, the peripheral circuit layer 20 is bonded to the side of the memory array circuit layer 30 away from the core circuit layer 10. The transistor channels in the memory array circuit layer 30 can use single-crystal silicon or indium gallium zinc oxide (IGZO) material. The transistor channels in the core circuit layer 10 use novel materials (e.g., thin-film single-crystal silicon, MoS2, Bi2O2Se, InSe, WSe2). The transistor channels in the peripheral circuit layer 20 can use single-crystal silicon or novel materials (e.g., thin-film single-crystal silicon, MoS2, Bi2O2Se, InSe, WSe2).

[0072] Furthermore, another alternative stacking method is described in reference... Figure 15 As shown, a core circuit layer 10 and a memory array circuit layer 30 are sequentially formed on a first substrate, and a peripheral circuit layer 20 is formed on a second substrate; then, the peripheral circuit layer 20 is bonded to the side of the memory array circuit layer 30 away from the core circuit layer 10. The transistor channels in the memory array circuit layer 30 can use indium gallium zinc oxide (IGZO) material. The transistor channels in the core circuit layer 10 and the peripheral circuit layer 20 can use monocrystalline silicon or new materials (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2).

[0073] Furthermore, another alternative stacking method is described in reference... Figure 16As shown, a core circuit layer 10 is formed on a first substrate, and a memory array circuit layer 30 and a peripheral circuit layer 20 are sequentially formed on a third substrate. The core circuit layer 10 is then bonded to the side of the memory array circuit layer 30 away from the peripheral circuit layer 20. The transistor channels in the memory array circuit layer 30 can use single-crystal silicon or indium gallium zinc oxide (IGZO) materials; the transistor channels in the peripheral circuit layer 20 must use new materials (e.g., thin-film single-crystal silicon, MoS2, Bi2O2Se, InSe, WSe2); the transistor channels in the core circuit layer 10 can use single-crystal silicon or new materials (e.g., thin-film single-crystal silicon, MoS2, Bi2O2Se, InSe, WSe2).

[0074] Furthermore, another alternative stacking method is described in reference... Figure 17 As shown, a core circuit layer 10 is formed on a first substrate, and a peripheral circuit layer 20 and a memory array circuit layer 30 are sequentially formed on a second substrate. The core circuit layer 10 is then bonded to the side of the memory array circuit layer 30 away from the peripheral circuit layer 20. The transistor channels in the memory array circuit layer 30 use indium gallium zinc oxide (IGZO) material; the transistor channels in the core circuit layer 10 and the peripheral circuit layer 20 can use monocrystalline silicon or new materials (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2).

[0075] Furthermore, another alternative stacking method is described in reference... Figure 19 As shown, when the core circuit layer 10 is formed on the first substrate, a memory array circuit layer 30 and a peripheral circuit layer 20 are then sequentially formed on the core circuit layer 10 using a single-substrate manufacturing process. The transistor channels in the memory array circuit layer 30 use indium gallium zinc oxide (IGZO) material; the transistor channels in the core circuit layer 10 can use monocrystalline silicon or new materials (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2); the transistor channels in the peripheral circuit layer 20 use new materials (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2).

[0076] Furthermore, another alternative stacking method is described in reference... Figure 20As shown, when the peripheral circuit layer 20 is sequentially formed on the second substrate, the memory array circuit layer 30 and the core circuit layer 10 are then sequentially formed on the peripheral circuit layer 20 using a single-substrate manufacturing process. The transistor channel in the memory array circuit layer 30 uses indium gallium zinc oxide (IGZO) material; the transistor channel in the core circuit layer 10 must use a new material (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2); the transistor channel in the peripheral circuit layer 20 can use monocrystalline silicon or a new material (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2).

[0077] In some embodiments of this disclosure, step S101 may further include step S302.

[0078] Step S302: Bond the side of the core circuit layer away from the memory array circuit layer to the peripheral circuit layer in the vertical direction.

[0079] In this embodiment of the disclosure, reference is made to Figure 13 As shown, a peripheral circuit layer 20 is formed on a second substrate, and a memory array circuit layer 30 and a core circuit layer 10 are sequentially formed on a third substrate; then, the peripheral circuit layer 20 is bonded to the side of the core circuit layer 10 away from the memory array circuit layer 30. The transistor channels in the memory array circuit layer 30 can use single-crystal silicon or indium gallium zinc oxide (IGZO) materials. The transistor channels in the core circuit layer 10 must use new materials (e.g., thin-film single-crystal silicon, MoS2, Bi2O2Se, InSe, WSe2). The transistor channels in the peripheral circuit layer 20 can use single-crystal silicon or new materials (e.g., thin-film single-crystal silicon, MoS2, Bi2O2Se, InSe, WSe2).

[0080] Furthermore, another alternative stacking method is described in reference... Figure 14 As shown, a core circuit layer 10 and a memory array circuit layer 30 are sequentially formed on a first substrate, and a peripheral circuit layer 20 is formed on a second substrate. The peripheral circuit layer 20 is then bonded to the side of the core circuit layer 10 away from the memory array circuit layer 30. The transistor channels in the memory array circuit layer 30 can use indium gallium zinc oxide (IGZO) material. The transistor channels in the core circuit layer 10 and the peripheral circuit layer 20 can use monocrystalline silicon or new materials (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2). This allows for an increase in the overlap area between the peripheral circuit and the memory array circuit without affecting the overlap area of ​​the core circuit and the memory array circuit, thereby significantly increasing the memory density.

[0081] In some embodiments of this disclosure, step S101 may further include step S303.

[0082] Step S303: Stack the storage array circuit layer, core circuit layer and peripheral circuit layer in sequence along the vertical direction.

[0083] In this embodiment of the disclosure, reference is made to Figure 21 As shown, when the memory array circuit layer 30 is formed on the first substrate, a core circuit layer 10 and a peripheral circuit layer 20 are then sequentially formed on the memory array circuit layer 30 using a single-substrate manufacturing process. The transistor channels in the memory array circuit layer 30 use single-crystal silicon or indium gallium zinc oxide (IGZO) material; the transistor channels in both the core circuit layer 10 and the peripheral circuit layer 20 use new materials (e.g., thin-film single-crystal silicon, MoS2, Bi2O2Se, InSe, WSe2).

[0084] In some embodiments of this disclosure, after step S101, the specific steps of the method for fabricating the storage device may further include step S401.

[0085] Step 401: Bond the side of the core circuit layer furthest from the peripheral circuit layer to the memory array circuit layer in the vertical direction.

[0086] In this embodiment of the disclosure, one selectable stacking method is described below. Figure 10 As shown, with a core circuit layer 10 formed on a first substrate, a peripheral circuit layer 20 formed on a second substrate, and a memory array circuit layer 30 formed on a third substrate, the side of the core circuit layer 10 furthest from the peripheral circuit layer 20 is bonded to the memory array circuit layer 30 in a vertical direction. The memory array circuit layer 30, the core circuit layer 10, and the peripheral circuit layer 20 are fabricated on three independent substrates and interconnected. The connection methods of the memory array circuit layer 30, the core circuit layer 10, and the peripheral circuit layer 20 include, but are not limited to, three-dimensional interconnection techniques using multiple wafers (such as hybrid bonding techniques). This independent design offers good flexibility, allowing for free combination with CMOS substrates of different generations, forming an open ecosystem, and meeting customized needs such as in-memory computing. Furthermore, it avoids lateral expansion of the circuit area, significantly increasing the overlapable area of ​​the circuit, effectively reducing the overall chip area, and thus increasing storage density. The transistor channels in the memory array circuit layer 30 can use single-crystal silicon or indium gallium zinc oxide (IGZO) materials. The transistor channels in the core circuit layer 10 and the peripheral circuit layer 20 can use monocrystalline silicon or new materials (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2).

[0087] Furthermore, another alternative stacking method is described in reference... Figure 8As shown, a core circuit layer 10 and a peripheral circuit layer 20 are sequentially formed on a first substrate, and a memory array circuit layer 30 is formed on a second substrate. Then, the side of the core circuit layer 10 away from the peripheral circuit layer 20 is bonded to the memory array circuit layer 30. At this time, the transistor channel of the core circuit layer 10 can use single-crystal silicon or new materials (e.g., thin-film single-crystal silicon, MoS2, Bi2O2Se, InSe, WSe2), the transistor channel of the peripheral circuit layer 20 can use new materials (e.g., thin-film single-crystal silicon, MoS2, Bi2O2Se, InSe, WSe2), and the transistor channel of the memory array circuit 30 can use single-crystal silicon or indium gallium zinc oxide (IGZO) material.

[0088] Furthermore, another alternative stacking method is described in reference... Figure 9 As shown, a peripheral circuit layer 20 and a core circuit layer 10 are sequentially formed on a second substrate, and a memory array circuit layer 30 is formed on the second substrate. Then, the side of the core circuit layer 10 away from the peripheral circuit layer 20 is bonded to the memory array circuit layer 30. At this time, the transistor channel of the core circuit layer 10 can use new materials (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2), the transistor channel of the peripheral circuit layer 20 can use monocrystalline silicon or new materials (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2), and the transistor channel of the memory array circuit 30 can use monocrystalline silicon or indium gallium zinc oxide (IGZO) material.

[0089] In some embodiments of this disclosure, after step S101, the specific steps of the method for fabricating the storage device may further include step S402.

[0090] Step 402: Form a memory array circuit layer on the side of the core circuit layer 10 away from the peripheral circuit layer 20.

[0091] In this embodiment of the disclosure, reference is made to Figure 18 As shown, when the peripheral circuit layer 20 is formed on the second substrate, the core circuit layer 10 and the memory array circuit layer 30 are then sequentially formed on the peripheral circuit layer 20 using a single-substrate manufacturing process. The transistor channels in the memory array circuit layer 30 use indium gallium zinc oxide (IGZO) material. The transistor channels in the peripheral circuit layer 20 can use monocrystalline silicon or new materials (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2). The transistor channels in the core circuit layer 10 use new materials (e.g., thin-film monocrystalline silicon, MoS2, Bi2O2Se, InSe, WSe2).

[0092] In some embodiments of this disclosure, all bonding methods in the storage device include hybrid bonding.

[0093] It should be noted that hybrid bonding is a bumpless bonding process. Two substrates are fabricated with dielectric layers and embedded copper pads respectively. After planarization and activation, they are aligned, bonded, and subjected to low-temperature annealing. The dielectric and copper pads are bonded simultaneously, achieving mechanical fixation and electrical interconnection of the substrates at the same interface, realizing high-density vertical interconnection. Hybrid bonding enables bumpless, high-density direct interconnection, significantly reducing interconnect spacing and signal delay between chips, thereby significantly improving integration density and transmission performance.

[0094] Based on the same inventive concept, this disclosure also provides an electronic device, which includes a storage device as provided in any of the above embodiments, or a storage device formed according to the manufacturing method of the storage device provided in any of the above embodiments. The electronic device may be a storage device, a smartphone, a computer, a tablet computer, an artificial intelligence device, a wearable device, or a power bank, etc. The storage device may include memory in a computer, etc., and is not limited thereto.

[0095] It should be understood that in the various embodiments of this disclosure, the sequence numbers of the processes described above do not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this disclosure. The sequence numbers of the embodiments of this disclosure are merely for description and do not represent the superiority or inferiority of the embodiments. It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0096] The sequence numbers of the embodiments disclosed above are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments. The methods disclosed in the several method embodiments provided in this disclosure can be arbitrarily combined to obtain new method embodiments without conflict. The features disclosed in the several product embodiments provided in this disclosure can be arbitrarily combined to obtain new product embodiments without conflict. The features disclosed in the several method or device embodiments provided in this disclosure can be arbitrarily combined to obtain new method embodiments or device embodiments without conflict.

[0097] The above are merely specific implementations of the embodiments of this disclosure, but the protection scope of the embodiments of this disclosure is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the embodiments of this disclosure should be included within the protection scope of the embodiments of this disclosure.

Claims

1. A storage device, characterized in that, The storage device includes: The core circuit layer is configured at least to control the read and write operations of the memory array circuitry; The peripheral circuit layer is configured at least to control data communication, address decoding, and timing management of the memory array circuitry. The core circuit layer and the peripheral circuit layer are stacked vertically.

2. The storage device according to claim 1, characterized in that, The storage device further includes: The storage array circuit layer is configured at least for storing data; The storage array circuit layer is stacked along the vertical direction with the core circuit layer and the peripheral circuit layer.

3. The storage device according to claim 2, characterized in that, Along the vertical direction, the memory array circuit layer is bonded to the side of the core circuit layer away from the peripheral circuit layer; or, along the vertical direction, the core circuit layer and the peripheral circuit layer are bonded through the memory array circuit layer. The core circuit layer, the peripheral circuit layer, and the memory array circuit layer are formed on different substrates.

4. The storage device according to claim 2, characterized in that, The storage array circuit layer and the core circuit layer are formed on the same substrate and are interconnected along the vertical direction; the peripheral circuit layer is formed on another substrate. Along the vertical direction, the side of the memory array circuit layer away from the core circuit layer is bonded to the peripheral circuit layer; Alternatively, along the vertical direction, the side of the core circuit layer away from the memory array circuit layer is bonded to the peripheral circuit layer.

5. The storage device according to claim 2, characterized in that, The storage array circuit layer and the peripheral circuit layer are formed on the same substrate and are interconnected along the vertical direction; the core circuit layer is formed on another substrate. Along the vertical direction, the side of the storage array circuit layer away from the peripheral circuit layer is bonded to the core circuit layer.

6. The storage device according to claim 2, characterized in that, The core circuit layer and the peripheral circuit layer are formed on the same substrate and interconnected along the vertical direction; the memory array circuit is formed on another substrate. Along the vertical direction, the side of the core circuit layer away from the peripheral circuit layer is bonded to the memory array circuit layer.

7. The storage device according to claim 2, characterized in that, The core circuit layer, the memory array circuit layer, and the peripheral circuit layer are formed on the same substrate; Along the vertical direction, the core circuit layer, the memory array circuit layer, and the peripheral circuit layer are stacked sequentially; or, Along the vertical direction, the storage array circuit layer, the core circuit layer, and the peripheral circuit layer are stacked sequentially; or, along the vertical direction, the peripheral circuit layer, the core circuit layer, and the storage array circuit layer are stacked sequentially; or, along the vertical direction, the peripheral circuit layer, the storage array circuit layer, and the core circuit layer are stacked sequentially.

8. The storage device according to claim 1, characterized in that, The core circuit layer or the peripheral circuit layer includes: an in-memory computing circuit, which is configured to perform computational processing on the data read from or written by the memory array circuit layer under the control of the core circuit or the peripheral circuit.

9. A method for manufacturing a storage device, characterized in that, The method for manufacturing the storage device includes: The core circuit layer and the peripheral circuit layer are stacked vertically; wherein, the core circuit layer is used to control the read and write operations of the memory array circuit; and the peripheral circuit layer is used to control the data communication, address decoding and timing management of the memory array circuit.

10. The method for manufacturing a storage device according to claim 9, characterized in that, Before stacking the core circuit layer and the peripheral circuit layer vertically, the method for fabricating the memory device further includes: A memory array circuit layer is formed on another substrate; or, a memory array circuit layer is stacked on the core circuit layer; or, a memory array circuit layer is stacked on the peripheral circuit layer. The storage array circuit layer is used to store data.

11. The method for manufacturing the storage device according to claim 10, characterized in that, The stacking of the core circuit layer and the peripheral circuit layer in the vertical direction includes: The core circuit layer and the peripheral circuit layer are stacked along the vertical direction through the memory array circuit layer; or, the side of the core circuit layer away from the memory array circuit layer is bonded to the peripheral circuit layer along the vertical direction; or, the memory array circuit layer, the core circuit layer and the peripheral circuit layer are stacked sequentially along the vertical direction.

12. The method for manufacturing the storage device according to claim 10, characterized in that, After stacking the core circuit layer and the peripheral circuit layer vertically, the method for fabricating the memory device further includes: The core circuit layer is bonded to the memory array circuit layer along the vertical direction on the side of the core circuit layer away from the peripheral circuit layer; or, the memory array circuit layer is formed by stacking on the side of the core circuit layer away from the peripheral circuit layer.

13. The method for manufacturing the storage device according to any one of claims 11-12, characterized in that, All bonding methods in the storage device include hybrid bonding.

14. An electronic device, characterized in that, This includes the memory device as described in any one of claims 1 to 8, or the memory device formed by the method of manufacturing the memory device according to any one of claims 9 to 13.