Integrated circuit structure and manufacturing method thereof

By forming source lines in the middle of the word string of 3D NAND memory, the problem of reduced current caused by the increase in word string length is solved, realizing increased current and the formation of high-density memory arrays, thus improving the performance of integrated circuits.

CN118900564APending Publication Date: 2024-11-05MACRONIX INTERNATIONAL CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310664232.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2023-05-02
Filing Date
2023-06-06
Publication Date
2024-11-05

AI Technical Summary

Technical Problem

In 3D NAND structures, as the number of stacked layers increases, the word string length becomes longer, resulting in a decrease in current and affecting the performance of integrated circuits.

Method used

Source lines are formed in the middle of the memory word string to reduce the word string length, and a 3D NAND structure is formed through a high-temperature stacking process. Then, top and bottom bit lines are formed in the back-end process to ensure that the performance of the integrated circuit is not affected.

Benefits of technology

By reducing the word string length, increasing the current in NAND, the current performance of integrated circuits is improved, and a high-density memory array can be formed without affecting other components.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118900564A_ABST
    Figure CN118900564A_ABST
Patent Text Reader

Abstract

The invention provides an integrated circuit structure and a method of manufacturing the same. The integrated circuit structure includes a substrate, a first memory string, a source line, and a second memory string. The first memory string is over the substrate and includes a plurality of first memory cells stacked in a vertical direction. The source line extends laterally over the first memory string. The second memory string is over the source line and includes a plurality of second memory cells stacked in a vertical direction.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to an integrated circuit structure, and more particularly to a method for manufacturing the integrated circuit structure. Background Art

[0002] The present disclosure relates generally to semiconductor devices, and more particularly to three-dimensional (3D) memory devices and methods of forming such semiconductor devices.

[0003] The semiconductor industry has experienced rapid growth due to the increasing integration density of various electronic components, such as transistors, diodes, resistors, capacitors, etc. In most cases, the increase in integration density comes from repeated reductions in minimum feature size, which allows more components to be integrated into a given area. Summary of the invention

[0004] The present disclosure provides an integrated circuit structure. The integrated circuit structure includes a substrate, a first memory string, a source line, and a second memory string. The first memory string is located above the substrate and includes a plurality of first memory cells stacked in a vertical direction. The source line extends laterally above the first memory string. The second memory string is located above the source line and includes a plurality of second memory cells stacked in a vertical direction.

[0005] In some embodiments, the first memory string includes a first channel layer. The first channel layer extends in a vertical direction and has an innermost position contacting the first surface of the source line.

[0006] In some embodiments, the second memory string includes a second channel layer. The second channel layer extends in a vertical direction and has an innermost position in contact with a second surface of the source line. The second surface and the first surface are located on opposite sides of the source line.

[0007] In some embodiments, the integrated circuit structure further includes a first bit line and a second bit line. The first bit line extends laterally between the substrate and the first memory string and is electrically connected to the first memory string. The second bit line extends laterally above the second memory string and is electrically connected to the second memory string.

[0008] In some embodiments, the first bit line and the second bit line are made of a same material.

[0009] The present disclosure provides a method for manufacturing an integrated circuit structure. The manufacturing method includes: forming a first multi-layer stacking structure, the first multi-layer stacking structure includes a plurality of first insulating layers stacked in a vertical direction and separated from each other; forming a first storage layer, the first storage layer vertically extending through the first multi-layer stacking structure; forming a first channel layer on the first storage layer; forming a source line extending laterally above the first multi-layer stacking structure; forming a second multi-layer stacking structure, the second multi-layer stacking structure includes a plurality of second insulating layers stacked in a vertical direction and separated from each other; forming a second storage layer, the second storage layer vertically extending through the second multi-layer stacking structure; forming a second channel layer on the second storage layer; forming a plurality of first gate layers alternately stacked with the plurality of first insulating layers to form a first memory string and a plurality of second gate layers alternately stacked with the plurality of second insulating layers to form a second memory string.

[0010] In some embodiments, the manufacturing method further includes: forming a first interconnect structure over the second multi-layer stack structure from a side of the second multi-layer stack structure opposite to the source line.

[0011] In some embodiments, the manufacturing method further includes: forming a second interconnect structure over the first multi-layer stack structure from a side of the first multi-layer stack structure opposite to the source line.

[0012] In some embodiments, the manufacturing method further includes bonding the complementary metal oxide semiconductor wafer to the first interconnect structure.

[0013] In some embodiments, the manufacturing method further includes: forming a deep through hole, the deep through hole vertically extending through the first interconnect structure, the first multi-layer stacking structure, the second multi-layer stacking structure and the source line, and electrically connecting the first memory string and the second memory string to the complementary metal oxide semiconductor wafer. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] One aspect of the present disclosure can be best understood from the following detailed description when read together with the accompanying drawings. It should be noted that, in accordance with standard practices in the industry, various features are not drawn to scale. In fact, the sizes of various features may be arbitrarily increased or reduced for clarity of discussion.

[0015] Figure 1A is a memory circuit according to some embodiments of the present disclosure;

[0016] Figure 1B is a cross-sectional view of an integrated circuit structure according to some embodiments of the present disclosure;

[0017] FIG. 2A to FIG. 2C is a flow chart of a method for manufacturing an integrated circuit structure according to some embodiments of the present disclosure;

[0018] Figures 3 to 24Bare cross-sectional views of a method for manufacturing an integrated circuit structure at various stages according to some embodiments of the present disclosure.

[0019] Description of Reference Numerals

[0020] 101: Substrate

[0021] 101b: Dorsal surface

[0022] 112: Semiconductor layer

[0023] 116: First multi-layer stacking structure

[0024] 116a: Insulation layer

[0025] 116b: Sacrificial layer

[0026] 116c: Dorsal surface

[0027] 116f: front surface

[0028] 136: Second multi-layer stacking structure

[0029] 136a: Insulation layer

[0030] 136b: Sacrificial layer

[0031] 136f: front surface

[0032] 150: Storage layer

[0033] 152: Channel layer

[0034] 152b: Dorsal surface

[0035] 154: Dielectric Materials

[0036] 155: Air gap

[0037] 156: Bonding pad

[0038] 170: Storage layer

[0039] 172: Channel layer

[0040] 174: Dielectric Materials

[0041] 175: Air gap

[0042] 176: Bonding pad

[0043] 180: Gate layer

[0044] 182: Storage unit

[0045] 184: Storage unit

[0046] 190: Interconnection structure

[0047] 191: Metallization layer

[0048] 193: Metalized through hole

[0049] 195: Intermetallic Dielectric Structure

[0050] 196: Interconnection Structure

[0051] 197: Metallization layer

[0052] 198: Metallized through hole

[0053] 199: Intermetallic Dielectric Structure

[0054] 200: Integrated Circuit Structure

[0055] 210: Semiconductor substrate

[0056] 211: Metal Oxide Semiconductor Devices

[0057] 220: Connected structure

[0058] 221: Metallization layer

[0059] 222: Metallized through hole

[0060] 223: Intermetallic Dielectric Structure

[0061] 330: Dielectric layer

[0062] 330b: back surface

[0063] 340: Interconnection Structure

[0064] 342: Deep through hole

[0065] 344: Metallization layer

[0066] 346: Intermetallic Dielectric Structure

[0067] 502: Metal plug

[0068] 504: Dielectric materials

[0069] C1: Space

[0070] C2: Space

[0071] M: Method

[0072] O1: Through hole

[0073] O2: Through hole

[0074] P1: Etching process

[0075] P2: Polishing process

[0076] P3: Etching process

[0077] P4: Polishing process

[0078] P5: Etching process

[0079] S1: Sub memory string

[0080] S2: Sub-memory string

[0081] S101-S124: Steps

[0082] X: Direction

[0083] Z: Direction DETAILED DESCRIPTION

[0084] The following disclosure provides many different embodiments or examples for implementing the different features of the provided target. Specific examples of components and configurations are described below to simplify the present disclosure. Of course, these are only examples and are not intended to be restrictive. For example, in the following description, the formation of a first feature above or on a second feature may include an embodiment in which the first feature is directly contacted with the second feature, and may also include an embodiment in which an additional feature may be formed between the first feature and the second feature so that the first feature and the second feature may not be directly contacted. In addition, the present disclosure may repeat reference numbers and / or letters in various examples. This repetition is for the purpose of simplicity and clarity, and itself does not specify the relationship between the various embodiments and / or configurations discussed.

[0085] Additionally, for ease of description, spatially relative terms, such as "below," "beneath," "lower," "above," "upper," and the like, may be used herein to describe the relationship of one element or feature to another element or feature of the illustrations in the figures. The spatially relative terms are intended to encompass different orientations of the device in use or operation in addition to the orientation depicted in the figures. The device may be otherwise oriented (rotated 90 degrees or at other orientations), and the spatially relative descriptors used herein should be similarly interpreted accordingly.

[0086] As used herein, "approximately", "about", "roughly", or "substantially" shall generally mean within 20%, or within 10%, or within 5% of a given value or range. However, those skilled in the art will recognize that the values ​​or ranges cited throughout the description are merely examples and may decrease as the scale of integrated circuits decreases. The quantities given herein are approximate, meaning that the terms "approximately", "about", "roughly", or "substantially" may be inferred where not explicitly stated.

[0087] For the development of semiconductor devices, high storage density memory devices are a development direction. Therefore, three-dimensional (3D) integrated circuit (IC) memory devices, such as 3D NAND, can provide high-density storage through its multi-layer structure, in which complementary metal oxide semiconductor (CMOS) can be placed under the memory array (CMOS under smemory array, CuA). However, the more layers stacked in 3D NAND, the longer the string length will be, which may cause the current in NAND to decrease.

[0088] Therefore, various embodiments of the present disclosure provide a 3D NAND that forms a source line in the middle of a memory string, thereby reducing the string length, so that the current in the NAND can be increased. For example, if the string length is halved, the current in the NAND can be doubled. In other words, if the number of strings is divided into n in the same column of the NAND array, the current can become n times the original. In addition, once the memory array is formed, the top bit line, the bottom bit line, and the CMOS chip can be bonded to the bottom and / or top of the memory array. Therefore, before bonding, the 3D NAND can be formed in a high thermal budged process without affecting the top bit line, the bottom bit line, and the CMOS chip, thereby helping to improve the performance of the integrated circuit structure. On the other hand, after the 3D NAND is formed, the top bit line and the bottom bit line can be formed by a backend of line (BEOL) process, thereby improving the RC value of the integrated circuit structure.

[0089] Please refer to Figure 1A and Figure 1B . Figure 1A Memory circuits according to some embodiments of the present disclosure. Figure 1B 1 is a cross-sectional view of an integrated circuit structure according to some embodiments of the present disclosure. A plurality of memory cells 182 may be defined at the intersection of the gate layer 180, the memory layer 150, and the channel layer 152, and a plurality of memory cells 184 may be defined at the intersection of the gate layer 180, the memory layer 170, and the channel layer 172, so that a memory cell array is formed in the first and second multilayer stacks 116 and 136. The gate layers 180 are spaced apart from each other by insulating layers 136a and 116a. The semiconductor layer 112 is located between the memory cells 182 and 184 and is electrically connected to the memory cells 182 and 184. In some embodiments, the semiconductor layer 112 may also be interchangeably referred to as a shared semiconductor layer or a shared source line.

[0090] In some embodiments, the memory cells 182 in the same column may be referred to as sub-memory strings S1, the memory cells 184 in the same column may be referred to as sub-memory strings S2, and the sub-memory strings S1 and S2 may be collectively referred to as memory strings. That is, the source line (i.e., semiconductor layer 112) is formed in the middle of the memory string, thereby shortening the string length, so that the current in the NAND may be increased. For example, if the string length is halved, the current in the NAND may be doubled. In other words, if the strings in the same column of the NAND array are divided into n (e.g., Figure 1B As shown in the figure, if there are two sub-memory strings S1 and S2), the current can become n times (eg, twice) the original current.

[0091] like Figure 1B As shown, an interconnect structure 190 is formed on the front side surface 136f of the multilayer stack structure 136. In some embodiments, the interconnect structure 190 may include three metallization layers 191 and three metallization through holes 193. Other embodiments may include more or fewer metallization layers and corresponding numbers of through holes. In some embodiments, the metallization layer 191 can be interchangeably referred to as a bottom bit line (BL). The interconnect structure 190 also includes a plurality of inter-metal dielectric (IMD) layers. In some embodiments, the plurality of inter-metal dielectric layers can be collectively referred to as an inter-metal dielectric structure 195. The inter-metal dielectric structure 195 can provide electrical insulation and structural support for various features of the integrated circuit during many process steps. In some embodiments, the topmost metal layer in the interconnect structure 190 can be interchangeably referred to as a bonding layer.

[0092] like Figure 1B As shown, the integrated circuit structure 200 is formed on the interconnect structure 190. The interconnect structure 190 includes a semiconductor substrate 210 and an interconnect structure 220 formed on the semiconductor substrate 210. In some embodiments, the semiconductor substrate 210 may include a complementary metal oxide semiconductor (CMOS) element 211 located therein. In some embodiments, the interconnect structure 220 may include three metallization layers 221 and three metallized vias 222. Other embodiments may include more or fewer metallization layers and a corresponding number of vias. The interconnect structure 220 also includes multiple intermetallic dielectric layers. In some embodiments, these intermetallic dielectric layers may be collectively referred to as intermetallic dielectric structures 223. The intermetallic dielectric structures 223 can provide electrical insulation and structural support to support various features of the integrated circuit in many process steps.

[0093] like Figure 1BAs shown, the dielectric layer 330 may be formed to laterally surround the end of the channel layer 152, and the backside surface 330b of the dielectric layer 330 may be flush with the backside surface 152b of the channel layer 152. An interconnect structure 196 is formed on the backside surface 116c of the multi-layer stack structure 116, and the interconnect structure 196 is electrically connected to the channel layer 152 of the memory cell 182. In some embodiments, the interconnect structure 196 may include three metallization layers 197 and three metallization vias 198. Other embodiments may include more or fewer metallization layers and a corresponding number of vias. In some embodiments, the metallization layer 197 may be interchangeably referred to as a top bit line (BL).

[0094] The interconnect structure 196 also includes a plurality of intermetallic dielectric layers. In some embodiments, these intermetallic dielectric layers may be collectively referred to as an intermetallic dielectric structure 199. The intermetallic dielectric structure 199 may provide electrical insulation and structural support to support various features of the integrated circuit during many process steps. In some embodiments, the top metal layer in the interconnect structure 196 may be interchangeably referred to as a bonding layer. In some embodiments, the interconnect structure 196 may be formed according to a back-end process.

[0095] The interconnect structure 340 is formed on the interconnect structure 196 and is electrically connected to the interconnect structure 196. The interconnect structure 340 may include a deep via. Fig.24A The deep via 342 shown is formed to pass through the interconnect structure 196 and the multilayer stack structures 116 and 136 from the outermost metallization layer 344 in the interconnect structure 340 to the metallization layer 191 in the interconnect structure 190, so that the deep via can electrically connect the memory cells 182 and 184 to the semiconductor elements in the integrated circuit structure 200. A plurality of intermetallic dielectric layers are also included in the interconnect structure 340. In some embodiments, the plurality of intermetallic dielectric layers can be collectively referred to as an intermetallic dielectric structure 346. The intermetallic dielectric structure 346 can provide electrical insulation and structural support for various integrated circuit features in many manufacturing process steps. In some embodiments, the topmost metal layer in the interconnect structure 346 can be interchangeably referred to as a bonding layer.

[0096] Now please refer to FIG. 2A to FIG. 2C , describes a flow chart of an exemplary method M for manufacturing an integrated circuit structure according to some embodiments. The method M includes relevant parts of the entire manufacturing process. It should be understood that the method may be FIG. 2A to FIG. 2CAdditional operations are provided before, during, and after the operations shown in, and some operations described below may be replaced or eliminated for additional embodiments of the foregoing methods. The order of operations / processes may be interchangeable. Method M includes manufacturing a memory element (e.g., a NAND memory element). However, the manufacture of a memory element is only one example of a manufacturing process for describing some embodiments of the present disclosure.

[0097] Figures 3 to 24A The method M is a method for manufacturing an integrated circuit structure at various stages according to some embodiments of the present disclosure. In some embodiments, the integrated circuit structure includes a vertical channel flash memory device. The method M starts at step S101. Please refer to Figure 3 In some embodiments of step S101, a multilayer stack structure 116 including alternately stacked insulating layers 116a and sacrificial layers 116b is formed on a substrate 101. The substrate 101 may be a semiconductor substrate, such as a single crystal silicon crystal material, a semiconductor-on-insulator (SOI) substrate, or other similar materials, which may be doped (e.g., using p-type or n-type impurities) or undoped. The semiconductor-on-insulator substrate may be a semiconductor material layer formed on an insulating layer. The insulating layer may be a buried oxide (BOX) layer, a silicon dioxide layer, or other similar materials. The insulating layer is typically located on a substrate, which may typically be a silicon or glass substrate. Other substrates, such as multilayer or gradient substrates, may also be used. In some embodiments, the substrate 101 may be a carrier wafer, such as a lower cost wafer or a reclaim wafer. The substrate 101 may have a circular top view shape or a rectangular top view shape.

[0098] In some embodiments, the semiconductor material of the substrate 101 may include silicon; germanium; a compound semiconductor including silicon carbide, gallium arsenide, indium phosphide, indium phosphide, indium arsenide, indium arsenide, and / or indium antimonide; an alloy semiconductor including silicon germanium, gallium arsenide phosphide, aluminum indium arsenide, aluminum gallium arsenide, gallium indium arsenide, gallium indium phosphide, and / or gallium indium arsenide phosphide; a similar material; or a combination thereof.

[0099] In some embodiments, the insulating layer 116a and the sacrificial layer 116b are arranged to extend parallel to each other and are alternately stacked on the substrate 101 along the direction Z. In some embodiments, the multi-layer stack structure 116 can be interchangeably referred to as a film stack structure (film stack) or a plurality of silicon oxide / silicon nitride pairs (deck of ON pairs). However, it should be noted that in the embodiments of the present disclosure, the material of the sacrificial layer 116b is different from the material of the insulating layer 116a. For example, the material of the sacrificial layer 116b can be silicon nitride, and the material of the insulating layer 116a can be silicon oxide. In some embodiments, the uppermost layer of the plurality of insulating layers 116a can be interchangeably referred to as a hard mask oxide layer. In some embodiments, the sacrificial layer 116b and the insulating layer 116a can be formed by low pressure chemical vapor deposition (LPCVD).

[0100] Please refer to the reply Figure 2A The method M then proceeds to step S102, where a first etching process P1 is performed to form a first through hole O1 passing through the first multi-layer stacked structure 116, wherein the first through hole O1 exposes the substrate 101. Figure 4In some embodiments of step S102, an etching process P1 (e.g., a hole etching process) is performed on the multilayer stacked structure 116 to form a plurality of through holes O1 penetrating the multilayer stacked structure 116, and the through holes O1 expose the substrate 101. In some embodiments, the etching process P1 may be an anisotropic etching process, such as a reactive ion etching (RIE) etching process, and the multilayer stacked structure 116 is etched using a patterned hard mask layer formed on the multilayer stacked structure 116 as an etching mask. In some embodiments, the through holes O1 may be a plurality of circular through holes passing through the multilayer stacked structure 116 along the Z direction and terminated at the substrate 101. The through holes O1 expose multiple portions of the sacrificial layer 116b and the insulating layer 116a. The multiple portions exposed by the sacrificial layer 116b and the insulating layer 116a may serve as sidewalls of the through holes O1. In some embodiments, the through hole O1 may be interchangeably referred to as a vertical channel opening.

[0101] Please refer to the reply Figure 2A The method M then proceeds to step S103, which is to form a first storage layer 150 on the sidewall of the first through hole O1. Figure 5 In some embodiments of step S103, the memory layer 150 is conformally deposited above the multi-layer stack structure 116 and on the sidewalls and bottom of the through hole O1. In some embodiments, the memory layer 150 may include a composite layer. The composite layer has but is not limited to an oxide layer-nitride layer-oxide layer (oxide-nitride-oxide; ONO) structure, an oxide layer-nitride layer-oxide layer-nitride layer-oxide layer (oxide-nitride-oxide; ONONO) structure, or an oxide layer-nitride layer-oxide layer-nitride layer-oxide layer-nitride layer-oxide layer-nitride layer-oxide layer (oxide-nitride-oxide; ONONONO) structure. Next, an etching process is performed to remove the composite layer located on the front side surface 116f of the multi-layer stack structure 116 and at the bottom of the through hole O1, thereby forming the memory layer 150.

[0102] Please refer to the reply Figure 2A The method M then proceeds to step S104, which is to conformally form a first channel layer 152 on the first storage layer 150. Figure 6In some embodiments of step S104, the channel layer 152 is conformally deposited above the storage layer 150 and has a U-shaped cross-section in the through hole O1, so the integrated circuit structure may include a vertical channel flash memory device. The storage layer 150 is located between the channel layer 152 and the sacrificial layer 116b. In some embodiments, the storage layer 150 may be made of a semiconductor material, such as polysilicon (Si), germanium, or other doped / undoped semiconductor materials. For example, the channel layer 152 may be made of doped polysilicon.

[0103] Please refer to the reply Figure 2A The method M then proceeds to step S105, which is to deposit a first dielectric material on the first channel layer and fill the first through hole. Figure 6 In some embodiments of step S105, a dielectric material 154 is deposited on the channel layer 152 and fills the first through hole O1. In some embodiments, the material of the dielectric material 154 includes, for example, silicon dioxide (SiO2). In some embodiments, at least one air gap 155 is formed in the through hole O1. In some embodiments, the material of the dielectric material 154 may include the same material as the insulating layer 116a. In some embodiments, the material of the dielectric material 154 may include a material different from the insulating layer 116a.

[0104] Please refer to the reply Figure 2A The method M then proceeds to step S106, which is to perform a first polishing process to remove excess first dielectric material, first channel layer, and first storage layer above the top surface of the first multi-layer stack structure 116. Figure 7 In some embodiments of step S106, a polishing process P2 (e.g., CMP) is performed to remove excess dielectric material 154, channel layer 152, and storage layer 150 above the front surface 116f of the multi-layer stack structure 116. After this step, the channel layer 152 surrounds the dielectric material 154 in the through hole O1, and the storage layer 150 surrounds the channel layer 152 in the through hole O1.

[0105] Please refer to the reply Figure 2A, method M then proceeds to step S107, which is to form a first bonding pad on the first dielectric material and contact the first channel layer. Please refer to step 8. In some embodiments of step S107, the dielectric material 154 is etched back to reproduce the upper part of the through hole O1. In some embodiments, the etch back process can be a dry etching process. For example, the dry etching process can be performed using hydrofluoric acid / ammonia (HF / NH3) or nitrogen trifluoride / ammonia (NF3 / NH3) as an etching gas. Then, a bonding pad 156 is formed in the upper part of the through hole O1 and is located on the etched dielectric material 154 to form an electrical contact with the channel layer 152. In some embodiments, the bonding pad 156 is formed by depositing polysilicon, germanium or a doped semiconductor material above the multilayer stack structure 116. Generally speaking, the aforementioned doped semiconductor material can use an n-type dopant (N+), such as phosphorus or arsenic. Next, a polishing process is performed to remove excess semiconductor material above the front side surface 116f of the multi-layer stack structure 116. After this step is performed, the bonding pad 156 may be formed as follows: Figure 8 In some embodiments, the bonding pad 156 may be an N-type doped (N+) polysilicon bonding pad.

[0106] Please refer to the reply Figure 2A The method M then proceeds to step S108, which is to form at least one semiconductor layer on the first multi-layer stack structure 116. Fig. 9 In some embodiments of step S108, a semiconductor layer 112 is formed above the multilayer stack structure 116. In some embodiments, the semiconductor layer 112 may be formed by depositing polysilicon and may be interchangeably referred to as a polysilicon layer. In some embodiments, the semiconductor layer 112 may be highly doped, including an n-type dopant (N+) such as phosphorus or arsenic. In some embodiments, the semiconductor layer 112 may be made of a semiconductor material such as polysilicon, germanium, silicon germanium, germanium arsenide, or the like, other suitable semiconductor materials, or combinations thereof. In some embodiments, the semiconductor layer 112 may be interchangeably referred to as a shared semiconductor layer or a shared source line.

[0107] Please refer to the reply Figure 2A The method M then proceeds to step S109, which is to form a second multi-layer stack structure 136 including alternately stacked second insulating layers and second sacrificial layers on the semiconductor layer. Fig.10In some embodiments of step S109, a multilayer stack structure 136 including alternately stacked insulating layers 136a and sacrificial layers 136b is formed on the semiconductor layer 112. That is, the semiconductor layer 112 is sandwiched between the multilayer stack structure 116 and the multilayer stack structure 136. The insulating layer 136a and the sacrificial layer 136b are arranged to extend parallel to each other and are alternately stacked on the semiconductor layer 112 along the direction Z. In some embodiments, the multilayer stack structure 136 can be interchangeably referred to as a film stack structure (filmstack) or multiple sets of silicon oxide / silicon nitride pairs (deck of ON pairs). However, it should be noted that in the embodiments of the present disclosure, the material of the sacrificial layer 136b is different from the material of the insulating layer 136a. For example, the material of the sacrificial layer 136b can be silicon nitride, and the material of the insulating layer 136a can be silicon oxide. In some embodiments, the uppermost layer of the multiple insulating layers 136a can be interchangeably referred to as a hard mask oxide layer. In some embodiments, the sacrificial layer 136 b and the insulating layer 136 a may be formed by low pressure chemical vapor deposition (LPCVD).

[0108] Please refer to the reply Figure 2A The method M then proceeds to step S110, where a second etching process is performed to form a second through hole through the second multi-layer stacked structure 136, wherein the second through hole exposes the semiconductor layer. Fig.11In some embodiments of step S110, an etching process P3 (e.g., a hole etching process) is performed on the multilayer stack structure 136 to form a plurality of through holes O2 penetrating the multilayer stack structure 136, and the through holes O2 expose the semiconductor layer 112. In some embodiments, the etching process P3 may be an anisotropic etching process, such as a reactive ion etching (RIE) etching process, and the multilayer stack structure 136 is etched using a patterned hard mask layer formed on the multilayer stack structure 136 as an etching mask. In some embodiments, the through holes O2 may be a plurality of circular through holes passing through the multilayer stack structure 136 along the Z direction and terminated at the semiconductor layer 112. The through holes O2 expose multiple portions of the sacrificial layer 136b and the insulating layer 136a. The multiple portions exposed by the sacrificial layer 136b and the insulating layer 126a may serve as sidewalls of the through holes O2. In some embodiments, the through hole O2 may be interchangeably referred to as a vertical channel opening. In some embodiments, the through hole O2 of the multilayer stack structure 136 may cover the through hole O1 of the multilayer stack structure 116 located below. In some embodiments, the through hole O2 of the multilayer stack structure 136 may not cover the through hole O1 of the multilayer stack structure 116 located below.

[0109] Please refer to the reply Figure 2B The method M then proceeds to step S111, which is to form a second storage layer on the sidewall of the second through hole. Fig.12 In some embodiments of step S111, the memory layer 170 is conformally deposited above the multilayer stack structure 136 and on the sidewalls and bottom of the through hole O2. In some embodiments, the memory layer 170 may include a composite layer. The composite layer has but is not limited to an oxide layer-nitride layer-oxide layer (oxide-nitride-oxide; ONO) structure, an oxide layer-nitride layer-oxide layer-nitride layer-oxide layer (oxide-nitride-oxide; ONONO) structure, or an oxide layer-nitride layer-oxide layer-nitride layer-oxide layer-nitride layer-oxide layer-nitride layer-oxide layer (oxide-nitride-oxide; ONONONO) structure. Next, an etching process is performed to remove the composite layer located on the front side surface 136f of the multilayer stack structure 116 and at the bottom of the through hole O3, thereby forming the memory layer 170.

[0110] Please refer to the reply Figure 2B The method M then proceeds to step S112, which is to conformally form a second channel layer above the second storage layer. Fig.13 In some embodiments of step S112, the channel layer 172 is conformally deposited above the storage layer 170 and has a U-shaped cross-section in the through hole O2, so that the integrated circuit structure may include a vertical channel flash memory device. The storage layer 170 is located between the channel layer 172 and the sacrificial layer 136b. In some embodiments, the storage layer 170 may be made of a semiconductor material, such as polysilicon (Si), germanium, or other doped / undoped semiconductor materials. For example, the channel layer 172 may be made of doped polysilicon.

[0111] Please refer to the reply Figure 2B The method M then proceeds to step S113, which is to deposit a second dielectric material on the second channel layer and fill the second through hole. Fig.13 In some embodiments of step S113, a dielectric material 174 is deposited on the channel layer 172 and filled in the second through hole O2. In some embodiments, the material of the dielectric material 174 includes, for example, silicon dioxide (SiO2). In some embodiments, at least one air gap 175 is formed and filled in the through hole O2. In some embodiments, the material of the dielectric material 174 may include the same material as the insulating layer 136a. In some embodiments, the material of the dielectric material 174 may include a material different from the insulating layer 136a.

[0112] Please refer to the reply Figure 2B The method M then proceeds to step S114, which is to perform a second polishing process to remove excess second dielectric material, second channel layer, and second storage layer above the top surface of the second multi-layer stack structure 136. Fig.14 In some embodiments of step S114, a polishing process P4 (e.g., CMP) is performed to remove excess dielectric material 174, channel layer 172, and memory layer 170 above the front surface 136f of the multilayer stack structure 136. After this step, the channel layer 172 surrounds the dielectric material 174 in the through hole O2, and the memory layer 170 surrounds the channel layer 172 in the through hole O2.

[0113] Please refer to the reply Figure 2B The method M then proceeds to step S115, which is to form a second bonding pad on the second dielectric material and contact the second channel layer. Fig.15In some embodiments of step S115, an etching back process is performed on the dielectric material 174 to reproduce the upper portion of the through hole O2. In some embodiments, the etching back process may be a dry etching process. For example, the dry etching process may be performed using hydrofluoric acid / ammonia (HF / NH3) or nitrogen trifluoride / ammonia (NF3 / NH3) as etching gas. Next, a bonding pad 176 is formed in the upper portion of the through hole O2 and is located on the etched dielectric material 174 to form an electrical contact with the channel layer 172. In some embodiments, the bonding pad 176 is formed by depositing polysilicon, germanium or a doped semiconductor material above the multilayer stack structure 136. Generally speaking, the aforementioned doped semiconductor material may use an n-type dopant (N+), such as phosphorus or arsenic. Next, a polishing process is performed to remove excess semiconductor material above the front side surface 136f of the multilayer stack structure 136. After this step is performed, the bonding pad 176 may be formed as follows. Fig.15 In some embodiments, the bonding pad 176 can be a p-type doped (P+) polysilicon bonding pad.

[0114] Please refer to the reply Figure 2B The method M then proceeds to step S116, which is to remove the first and second sacrificial layers in the first and second multi-layer stack structures 116 and 136. Fig.16 In some embodiments of step S116, an etching process is performed to form a through hole (not shown) passing through the first and second multilayer stacks 116 and 136 along the Z direction, the through hole terminating at the substrate 101, so as to partially expose the sidewalls of the sacrificial layers 116b, 136b and the sidewalls of the insulating layers 116a, 136a. In some embodiments, the etching process may be an anisotropic etching process, such as a reactive ion etching (RIE) etching process, and the multilayer stack structures 116, 136 are etched using a patterned hard mask layer formed on the multilayer stack structure 136 as an etching mask.

[0115] Next, the sacrificial layer 116b and the sacrificial layer 136b are removed by a solution, such as phosphoric acid (H3PO4), through a through hole (not shown) to expose a portion of the storage layer 150 and the storage layer 170. Therefore, the space C1 is formed to inherit the shape of the sacrificial layer 116b, and the space C2 is formed to inherit the shape of the sacrificial layer 136b.

[0116] Please refer to the reply Figure 2BThe method M then proceeds to step S117, where a plurality of gate layers are formed in the spaces originally occupied by the first and second sacrificial layers to form a plurality of first memory cells with the first memory layer and the first channel layer, and a plurality of second memory cells with the second memory layer and the second channel layer. Fig.17 In some embodiments of step S117, the plurality of gate layers 180 are formed at Fig.16 The through holes (not shown) in the space C1 and the space C2 are formed. Therefore, a plurality of memory cells 182 can be defined at the intersection of the gate layer 180, the memory layer 150 and the channel layer 152, and a plurality of memory cells 184 can be defined at the intersection of the gate layer 180, the memory layer 170 and the channel layer 172 to form a memory cell array in the first multi-layer stacking structure 116 and the second multi-layer stacking structure 136, thereby forming a three-dimensional NAND architecture. The semiconductor layer 112 is located between the memory cells 182 and the memory cells 184, and is electrically connected to the memory cells 182 and the memory cells 184.

[0117] In some embodiments, the memory cell 182 in the same column may be referred to as a sub-memory string S1, the memory cell 184 in the same column may be referred to as a sub-memory string S2, and the sub-memory strings S1 and S2 may be collectively referred to as a memory string. That is, the source line (i.e., the semiconductor layer 112) is formed in the middle of the memory string, thereby shortening the string length, so that the current in the NAND may be increased. For example, if the string length is halved, the current in the NAND may be doubled. In other words, if the strings in the same column of the NAND array are divided into n (e.g., sub-memory strings S1 and S2), the current may become n times (e.g., twice) the original.

[0118] In NAND flash memory, a memory string (e.g., memory strings S1, S2) is a column of connected memory cells (e.g., memory cells 182, 184) in a NAND memory array. NAND flash memory is organized into blocks, pages, and strings. Each block contains multiple pages, and each page contains multiple memory strings. In a memory string, the memory cells are connected in series and controlled by a single select gate transistor (e.g., gate layer 180). To access a specific memory cell in a memory string, a column voltage is applied to the control gate and select gate transistor of each memory cell in the string. This allows the data stored in the memory cell to be read or written. Memory strings are used in NAND flash memory arrays because they allow a high density of memory cells in a small area of ​​the chip.

[0119] In some embodiments, the material of the gate layer 180 may include polysilicon, metal or other suitable conductive materials. In some embodiments, the gate layer 180 may include multiple metal layers, such as TiN / W, TaN / W, TaN / Mo or other suitable metal layers. In some embodiments, the gate layer 180 may include a dielectric layer, such as AlOx. For example, each gate layer 180 may be a multi-layer structure including a high dielectric constant material layer (such as an HfOx layer or an AlOx layer), a TiN layer and a tungsten layer.

[0120] Please refer to the reply Figure 2B The method M then proceeds to step S118, which is to form a first interconnect structure on the front surface of the second multi-layer stack structure 136 and electrically connect the first memory cell and the second memory cell. Fig.18 In some embodiments of step S118, an interconnect structure 190 is formed above the front side surface 136f of the second multi-layer stack structure 136. In some embodiments, the interconnect structure 190 may include three metallization layers 191 and three metallization vias 193. Other embodiments may include more or fewer metallization layers and corresponding numbers of vias. In some embodiments, the metallization layer 191 may be interchangeably referred to as a bottom bit line (BL).

[0121] In some embodiments, the metallization layer 191 and the metallization through-hole 193 can be made of TiN, TaN, Ti, Ta, Cu, Al, Ag, W, Ir, Ru, Pt, combinations of the foregoing materials, or other suitable conductive materials.

[0122] The interconnect structure 190 also includes a plurality of inter-metal dielectric (IMD) layers. In some embodiments, the plurality of IMD layers may be collectively referred to as an IMD structure 195. The IMD structure 195 may provide electrical insulation and structural support for various features of the integrated circuit during many process steps. In some embodiments, the topmost metal layer in the interconnect structure 190 may be interchangeably referred to as a bonding layer. For example, in Fig.19 In the subsequent process shown in , the uppermost metal layer in the interconnect structure 190 may be a copper layer to achieve copper-to-copper bonding. In some embodiments, the interconnect structure 190 may be formed by a back end of line (BEOL) process.

[0123] Please refer to the reply Figure 2B The method M then proceeds to step S119, which is to bond the integrated circuit structure including the semiconductor device to the first interconnect structure to form a stacked wafer structure. Fig.19In some embodiments of step S119, an integrated circuit structure 200 (e.g., complementary metal oxide semiconductor, CMOS wafer) is provided. The integrated circuit structure 200 includes a semiconductor substrate 210 and an interconnect structure 220 formed on the semiconductor substrate 210. In some embodiments, the semiconductor substrate 210 may include a complementary metal oxide semiconductor element 211. In some embodiments, the interconnect structure 220 may include, for example, three metallization layers 221 and three metallized vias 222. Other embodiments may include more or fewer metallization layers and correspondingly more or fewer vias. In some embodiments, the metallization layer 221 and the metallized via 222 may be made of TiN, TaN, Ti, Ta, Cu, Al, Ag, W, Ir, Ru, Pt, combinations thereof, or other suitable conductive materials. The interconnect structure 220 also includes a plurality of intermetallic dielectric layers. In some embodiments, the plurality of intermetallic dielectric layers may be collectively referred to as an intermetallic dielectric structure 223. The IMD structure 223 may provide electrical insulation and structural support to various features of the integrated circuit during many fabrication process steps.

[0124] Next, the interconnect structure 190 with the memory elements 182 and 184 (e.g., 3D NAND) thereon is bonded to the interconnect structure 220 to electrically connect to the CMOS element 211, thereby forming a stacked wafer structure 300. Therefore, before bonding, the CMOS element 211 is not affected by the higher thermal budget in the memory process steps. The CMOS element 211 can be formed by other processes (advanced process) to provide higher speed and lower current operation and have a smaller chip area. Specifically, the integrated circuit structure 200 is bonded to the outermost metal layer in the interconnect structure 190 through the outermost metal layer (e.g., bonding layer) in its interconnect structure 220. In some embodiments, the combination of the interconnect structures 190 and 220 can be referred to as an interconnect stack.

[0125] Please refer to the reply Figure 2B The method M then proceeds to step S120, which is to flip the stacked wafer structure upside down. Fig. 20 In some embodiments of step S120, after inversion, the backside surface 101b of the substrate 101 faces upward, and the memory elements 182 and 184 are supported by the integrated circuit structure 200 for subsequent processes. In some embodiments, the inversion process may be optional.

[0126] Please refer to the reply Figure 2C The method M then proceeds to step S121, which is to remove the substrate so that the first channel layer of the first memory element is exposed from the back side of the first multi-layer stack structure 116. Fig.21 In some embodiments of step S121, an etching process P5 is performed on the back surface 101b of the substrate 101, including a silicon etching process and a polishing process (eg, CMP), and the etching process P5 is terminated at the outermost insulating layer 116a in the multi-layer stacked structure 116. The structure obtained after the etching process P5 is completed is as shown in FIG. Fig.21 As shown, the channel layer 152 is exposed from the backside surface 116c of the multi-layer stack structure 116. In some embodiments, the backside surface 152b of the channel layer 152 protrudes from the backside surface 116c of the multi-layer stack structure 116.

[0127] Please refer to the reply Figure 2C The method M then proceeds to step S122, which is to form a dielectric layer on the back side of the first multi-layer stack structure 116 and laterally surround the end portion of the first channel layer. Fig. 22 In some embodiments of step S122, a dielectric layer 330 is deposited on the back surface 116c of the first multi-layer stacked structure 116. In some embodiments, the dielectric layer 330 can be made of a material such as silicon dioxide (SiO2). In some embodiments, the dielectric layer 330 can be made of the same material as the insulating layer 116a and / or the insulating layer 136a. In some embodiments, the dielectric layer 330 can be made of a different material from the insulating layer 116a and / or the insulating layer 136a.

[0128] Next, a polishing process is performed to remove excess dielectric layer 330 above the channel layer 152 until the channel layer 152 is exposed, so that the dielectric layer 330 can laterally surround the end portion of the channel layer 152, and the back surface 330b of the dielectric layer 330 can be in a position flush with the back surface 152b of the channel layer 152.

[0129] Please refer to the reply Figure 2C The method M then proceeds to step S123, which is to form a second interconnect structure on the back surface of the first multilayer stack 116, the second interconnect structure being electrically connected to the first channel layer of the first memory cell. Fig.23In some embodiments of step S123, an interconnect structure 196 is formed on the backside surface 116c of the multilayer stack structure 116 and is electrically connected to the channel layer 152 of the memory cell 182. In some embodiments, the interconnect structure 196 may include three metallization layers 197 and have three metallization vias 198. Other embodiments may include more or fewer metallization layers and corresponding numbers of vias. In some embodiments, the metallization layer 197 may be interchangeably referred to as a top bitline (BL).

[0130] Furthermore, once the memory array is formed, the top bit lines, bottom bit lines, and CMOS wafer can be bonded to the bottom and / or top of the memory array. Therefore, before bonding, a high heat accumulation process can be used to form the 3D NAND without affecting the top bit lines, bottom bit lines, and CMOS wafer, thereby improving the performance of the integrated circuit structure.

[0131] In some embodiments, the metallization layer 197 and the metallized via 198 may be made of TiN, TaN, Ti, Ta, Cu, Al, Ag, W, Ir, Ru, Pt, combinations thereof, or other suitable conductive materials. The interconnect structure 196 also includes a plurality of intermetallic dielectric layers. In some embodiments, the plurality of intermetallic dielectric layers may be collectively referred to as an intermetallic dielectric structure 199. The intermetallic dielectric structure 199 may provide electrical insulation and structural support for various features of the integrated circuit in many manufacturing process steps. In some embodiments, the topmost metal layer in the interconnect structure 196 may be interchangeably referred to as a bonding layer. In some embodiments, the interconnect structure 196 may be formed according to a back end of line (BEOL) process.

[0132] Please refer to the reply Figure 2C The method M then proceeds to step S124, where a third interconnect structure including a deep through hole is formed above the second interconnect structure to electrically connect the first and second memory cells to the semiconductor device in the integrated circuit structure. Fig.24A In some embodiments of step S124, interconnect structure 340 is formed above interconnect structure 196 and electrically connected to interconnect structure 196. Interconnect structure 340 includes deep via 342. Deep via 342 is formed to extend from outermost metallization layer 344 in interconnect structure 340 through interconnect structure 196, multi-layer stack structures 116 and 136, and interconnect structure 190 to metallization layer 221 in interconnect structure 220, so that deep via 342 can electrically connect memory cells 182 and 184 to semiconductor device 211 in integrated circuit structure 200.

[0133] In some embodiments, the metallization layer 344 and / or the deep via 342 may be made of TiN, TaN, Ti, Ta, Cu, Al, Ag, W, Ir, Ru, Pt or other suitable conductive materials. In some embodiments, the deep via 342 may include one or more barrier / adhesion layers (not shown) surrounding it to protect the deep via 342 from metal diffusion (e.g., copper diffusion). The one or more barrier / adhesion layers may include titanium, titanium nitride, tantalum, tantalum nitride or other similar materials and may be formed using physical vapor deposition (PVD), CVD, ALD or other similar processes. A plurality of intermetallic dielectric layers are also included in the interconnect structure 340. In some embodiments, the plurality of intermetallic dielectric layers may be collectively referred to as an intermetallic dielectric structure 346. The intermetallic dielectric structure 346 may provide electrical insulation and structural support for various features of the integrated circuit during many manufacturing process steps. In some embodiments, the topmost metal layer in the interconnect structure 346 may be interchangeably referred to as a bonding layer.

[0134] Please refer to Fig. 24B . Fig. 24B Some alternative embodiments according to the present disclosure are shown. Fig.24A In some embodiments, the materials and manufacturing methods of the components of this embodiment are similar to those of the Fig.24A The materials and manufacturing methods of the components shown are basically the same. For other relevant details, please refer to the previous paragraphs, which will not be repeated here. In addition, the same reference symbols and / or letters may be reused in different examples in this embodiment. This repetition is for simplicity and clarity and is not used to determine the relationship between the various embodiments and / or configurations discussed.

[0135] This embodiment and Fig.24A The difference between the embodiment of the present invention and the embodiment of the present invention is that the present embodiment forms a metal plug 502, which penetrates the multi-layer stacked structures 116 and 136 and the common source line (i.e., the semiconductor layer 112) and electrically connects the interconnect structure 196 to the semiconductor element 211 in the integrated circuit structure 200. The present embodiment also forms a dielectric material 504, which laterally surrounds the metal plug 502 to electrically isolate the metal plug 502 from the gate layer 180. In some embodiments, the metal plug 502 can be made of TiN, TaN, Ti, Ta, Cu, Al, Ag, W, Ir, Ru, Pt, combinations thereof, or other suitable conductive materials. In some embodiments, the dielectric material 504 can be made of materials such as silicon dioxide (SiO2) or silicon nitride (SiN).

[0136] Although the present invention has described some of its embodiments in detail, other embodiments are possible. Therefore, the spirit and scope of the attached claims should not be limited to the description of the embodiments contained herein. The present invention provides a 3D NAND with a source line located in the middle of a memory string in various embodiments, thereby reducing the string length and increasing the current in the NAND. For example, if the string length is halved, the current in the NAND can be doubled. In other words, if the number of strings in the same column of the NAND array is divided into n, the current can be n times the original. In addition, once the memory array is formed, the top bit line, the bottom bit line and the complementary metal oxide wafer can be bonded to the bottom and / or top of the memory array. Therefore, before bonding, the 3D NAND can be formed in a high thermal budget process without affecting the top bit line, the bottom bit line and the complementary metal oxide wafer, thereby helping to improve the performance of the integrated circuit structure. On the other hand, after the 3D NAND is formed, the top bit line and the bottom bit line can be formed by a back-end (BEOL) process, thereby improving the RC of the integrated circuit structure.

[0137] The foregoing summarizes the features of several embodiments so that those skilled in the art can better understand aspects of the present disclosure. Those skilled in the art will appreciate that they can readily use the present disclosure as a basis for designing or modifying other processes and structures for implementing the same purposes and / or achieving the same advantages of the embodiments introduced herein. Those skilled in the art will also recognize that such equivalent constructions do not depart from the spirit and scope of the present disclosure, and such equivalent constructions may be variously modified, substituted, and replaced herein without departing from the spirit and scope of the present disclosure.

Claims

1. An integrated circuit structure, comprising: a substrate; a first memory string located above the substrate and comprising a plurality of first memory cells stacked in a vertical direction; a source line extending laterally above the first memory string; as well as A second memory string is located above the source line and includes a plurality of second memory cells stacked in the vertical direction. 2 . The integrated circuit structure according to claim 1 , wherein the first memory string comprises a first channel layer extending in the vertical direction and having an innermost position contacting the first surface of the source line.

3. The integrated circuit structure according to claim 2, wherein the second memory string comprises a second channel layer, the second channel layer extends in the vertical direction and has an innermost position in contact with the second surface of the source line, and the second surface and the first surface are located on opposite sides of the source line.

4. The integrated circuit structure according to claim 1, further comprising: a first bit line extending laterally between the substrate and the first memory string and electrically connected to the first memory string; and A second bit line extends laterally above the second memory string and is electrically connected to the second memory string.

5. The integrated circuit structure according to claim 1, wherein the substrate is a complementary metal oxide semiconductor wafer, and the integrated circuit structure further comprises: a first interconnect structure located between the complementary metal oxide semiconductor wafer and the first memory string; a second interconnect structure located on a side of the second memory string relative to the first memory string and electrically connected to the second memory string; and A deep through hole vertically extends through the second interconnect structure, the source line and the first interconnect structure to electrically connect the first memory string and the second memory string to the complementary metal oxide semiconductor wafer.

6. A method for manufacturing an integrated circuit structure, comprising: forming a first multi-layer stacking structure, the first multi-layer stacking structure comprising a plurality of first insulating layers stacked in a vertical direction and separated from each other; forming a first storage layer, the first storage layer extending vertically through the first multi-layer stack structure; forming a first channel layer on the first storage layer; forming a source line extending laterally above the first multi-layer stack structure; forming a second multi-layer stacking structure, the second multi-layer stacking structure comprising a plurality of second insulating layers stacked in the vertical direction and separated from each other; forming a second storage layer extending vertically through the second multi-layer stack structure; forming a second channel layer on the second storage layer; as well as A plurality of first gate layers alternately stacked with the first insulating layers are formed to form a first memory string, and a plurality of second gate layers alternately stacked with the second insulating layers are formed to form a second memory string.

7. The manufacturing method according to claim 6, further comprising: A first interconnection structure is formed above the second multi-layer stack structure from a side of the second multi-layer stack structure opposite to the source line.

8. The manufacturing method according to claim 7, further comprising: A second interconnection structure is formed above the first multi-layer stack structure from a side of the first multi-layer stack structure opposite to the source line.

9. The manufacturing method according to claim 7, further comprising: A complementary metal oxide semiconductor wafer is bonded to the first interconnect structure.

10. The manufacturing method according to claim 9, further comprising: A deep via is formed, the deep via vertically extending through the first interconnect structure, the first multi-layer stack structure, the second multi-layer stack structure and the source line, and electrically connecting the first memory string and the second memory string to the complementary metal oxide semiconductor wafer.