Memory cell, memory device and method for manufacturing the same

By adopting the nanosheet transistor structure and internal spacer design in the antifuse memory cell, the contact between the gate stack and the source/drain component is optimized, and the reduction of parasitic capacitance and contact area in the prior art is solved. Problem, improving the performance of memory cells.

CN113363259BActive Publication Date: 2024-12-03TAIWAN SEMICONDUCTOR MANUFACTURING CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202110476796.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2020-05-22
Filing Date
2021-04-29
Publication Date
2024-12-03
Estimated Expiration
2041-04-29

AI Technical Summary

Technical Problem

The existing antifuse memory units have problems of reduced parasitic capacitance and contact area during programming and reading, which affects their performance, especially operating speed and programming yield.

Method used

Using a nanosheet transistor structure, by forming a programming transistor and a reading transistor, the gate stack of the programming transistor is in direct contact with the source/drain component, and an internal spacer is introduced into the reading transistor for electrical isolation, optimizing the gate stack and The contact configuration of the source/drain component and the alternating nanostructure column and gate stack are formed by specific process steps to increase the length and contact area of ​​the conducting channel.

Benefits of technology

Improves the breakdown voltage and time of the programming transistor, reduces the parasitic capacitance of the read transistor, and enhances the performance of the overall memory cell, including operating speed and programming yield.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113363259B_ABST
    Figure CN113363259B_ABST
Patent Text Reader

Abstract

A memory device includes first nanostructures stacked together; a first gate stack, wherein two adjacent ones of the first gate stacks surround a corresponding one of the first nanostructures; a second nanostructure stacked together; a second gate stack, wherein two adjacent ones of the second gate stacks surround a corresponding one of the second nanostructures; a first drain / source component electrically coupled to a first end of the first nanostructure; a second drain / source component electrically coupled to both a second end of the first nanostructure and a first end of the second nanostructure; and a third drain / source component electrically coupled to a second end of the second nanostructure. At least one of the plurality of first gate stacks directly contacts at least one of the first drain / source component and the second drain / source component. Embodiments of the present invention also relate to a memory cell and a method of manufacturing a memory device.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] Embodiments of the present invention relate to a memory cell, a memory device, and a method of manufacturing the same. Background Art

[0002] Integrated circuits (ICs) sometimes include one-time programmable (OTP) memory to provide non-volatile memory (NVM), where data is not lost when the IC is powered off. One type of OTP device includes an anti-fuse memory. The anti-fuse memory includes a number of anti-fuse memory cells (or bit cells), the terminals of which are disconnected before programming and are in a short circuit state (e.g., turned on) after programming. The anti-fuse memory can be based on metal oxide semiconductor (MOS) technology. For example, the anti-fuse memory cell can include a programming MOS transistor (or MOS capacitor) and at least one reading MOS transistor. The gate dielectric of the programming MOS transistor may fail, thereby causing the gate and source or drain regions of the programming MOS transistor to be interconnected. Depending on whether the gate dielectric of the programming MOS transistor fails, different data bits can be presented by the anti-fuse memory cell by reading the composite current flowing through the programming MOS transistor and the reading MOS transistor. Since the programming state of the anti-fuse cell cannot be determined by reverse engineering, the anti-fuse memory has the advantageous feature of reverse engineering inspection. Summary of the invention

[0003] According to one aspect of an embodiment of the present invention, a memory device is provided, comprising: a plurality of first nanostructures stacked together; a plurality of first gate stacks, wherein two adjacent ones of the first gate stacks surround a corresponding one of the plurality of first nanostructures; a plurality of second nanostructures stacked together; a plurality of second gate stacks, wherein two adjacent ones of the second gate stacks surround a corresponding one of the plurality of second nanostructures; a first drain / source component electrically coupled to a first end of the first nanostructure; a second drain / source component electrically coupled to both a second end of the first nanostructure and a first end of the second nanostructure; and a third drain / source component electrically coupled to a second end of the second nanostructure. At least one of the plurality of first gate stacks directly contacts at least one of the first drain / source component and the second drain / source component.

[0004] According to another aspect of an embodiment of the present invention, a memory cell is provided, comprising: a first transistor; and a second transistor electrically coupled in series to the first transistor. Wherein, the first transistor comprises: a plurality of first nanosheets, separated from each other in a vertical direction, the plurality of first nanosheets having a first length in a horizontal direction; and a plurality of first fully surrounding gate stacks, operably associated with the plurality of first nanosheets, the plurality of first fully surrounding gate stacks having a second length in a horizontal direction, the second length being equal to or less than the first length. And wherein, the second transistor comprises: a plurality of second nanosheets, separated from each other vertically, the plurality of second nanosheets having a third length in a horizontal direction; and a plurality of second fully surrounding gate stacks, operably associated with the plurality of second nanosheets, the plurality of second fully surrounding gate stacks having a fourth length in a horizontal direction, the fourth length being less than the third length.

[0005] According to another aspect of an embodiment of the present invention, a method for manufacturing a memory device is provided, comprising: forming a first stack above a substrate, the first stack comprising a first nanosheet, a second nanosheet above the first nanosheet, and a third nanosheet above the second nanosheet; forming a second stack above the substrate, the second stack comprising a fourth nanosheet, a fifth nanosheet above the fourth nanosheet, and a sixth nanosheet above the fifth nanosheet; while covering the first stack, removing respective ends of the fourth nanosheet and the sixth nanosheet; and while still covering the first stack, forming a plurality of spacers at respective etched ends of the fourth nanosheet and the sixth nanosheet. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] When with Figure 1 When read together, various aspects of the present invention can be best understood from the following detailed description. It should be noted that, in accordance with standard practice in the industry, the various components are not drawn to scale. In fact, for the sake of clarity of discussion, the size of the various components can be arbitrarily increased or reduced.

[0007] Figure 1 An exemplary circuit diagram of a memory cell according to some embodiments is shown.

[0008] Figure 2 A flow chart of an exemplary method of manufacturing a memory device according to some embodiments is shown.

[0009] Figure 3-1 , Figure 3-2 , Figure 3-3 , Figure 3-4 , Figure 3-5 , Figure 3-6 , Figure 3-7 , Figure 3-8 , Figure 3-9 , Figure 3-10 , Figure 3-11 , Figure 3-12 and Figure 3-13 According to some embodiments Figure 2 sectional views of various manufacturing stages of a memory device manufactured by the method in FIG.

[0010] Figure 4 A cross-sectional view of another exemplary memory device 400 is shown in accordance with some embodiments.

[0011] Figure 5 , Figure 6 and Figure 7 Various exemplary nanostructure transistors including partial internal spacers are shown in accordance with some embodiments. DETAILED DESCRIPTION

[0012] The following disclosure provides many different embodiments or examples for realizing the different features of the provided subject matter. Specific examples of elements and arrangements will be described below to simplify the present invention. Of course, these are merely examples and are not intended to limit the present invention. For example, in the following description, forming a first component above or on a second component may include an embodiment in which the first component and the second component are directly contacted, and may also include an embodiment in which another component may be formed between the first component and the second component so that the first component and the second component may not be in direct contact. In addition, the present invention may repeat reference numerals and / or characters in various examples. This repetition is for the purpose of simplicity and clarity, and does not itself indicate the relationship between the various embodiments and / or configurations discussed.

[0013] Additionally, for ease of description, spatially relative terms such as "below," "beneath," "lower," "above," "upper," etc. may be used herein to describe the relationship of one element or component to another element or component as illustrated 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 therein interpreted accordingly.

[0014] In contemporary semiconductor device manufacturing processes, a large number of semiconductor devices, such as silicon channel n-type field effect transistors (nFETs) and silicon germanium channel p-type field effect transistors (pFETs), are manufactured on a single wafer. Non-planar transistor device architectures, such as fin-based transistors, can provide increased device density and enhanced performance over planar transistors. Due in part to the characteristics of their conductive channels surrounded by respective gate components, some advanced non-planar transistor device architectures, such as nanostructure (e.g., nanosheet, nanowire, or multi-bridge channel) transistors, can further enhance performance over fin-based transistors.

[0015] Such a nanosheet transistor includes multiple semiconductor sheets. Multiple semiconductor sheets can together form a conductive channel for a nanosheet transistor. Each of the semiconductor sheets is separated by a gate stack including a layer of conductive gate and gate dielectric. The gate stack surrounds all sides of the semiconductor sheet, thereby forming a full-surround gate (GAA) structure. Epitaxial growth on the ends of the semiconductor nanosheets forms source / drain components for the nanosheet transistor. Spacers can be formed between the gate stacks from the source / drain components of the nanosheet transistor. Such spacers are generally referred to as "internal spacers". The internal spacers formed of dielectric material can electrically isolate the gate stack from the respective source / drain components, which may reduce the parasitic capacitance (e.g., C gd , C gs ).

[0016] By adopting such a nanosheet transistor configuration to form a read transistor of a memory cell (e.g., an anti-fuse memory cell), the performance (e.g., switching speed) of the read transistor can be enhanced. However, the internal spacers coupled between the gate stack and the source / drain features may increase the time and voltage (commonly referred to as “T BD ” and “V BD ”) to break through the gate dielectric. This is partly due to the presence of the internal spacer which may reduce the contact area between the gate stack and the nanosheet, thereby reducing the contact area between each of the gate stack and the source / drain features. As a result, the overall performance (e.g., operating speed, programming yield, etc.) of the anti-fuse memory cell may be adversely affected.

[0017] The present invention provides various embodiments of memory cells in nanosheet transistors. In some embodiments, the disclosed memory cell includes an anti-fuse memory cell consisting of a programming transistor and one or more read transistors. Each of the programming transistor and the one or more read transistors includes a nanosheet transistor. The programming transistor can have less dielectric material coupled between the respective gate stack and the source / drain components, and the read transistor can have more dielectric material coupled between the respective gate stack and the source / drain components. For example, the programming transistor of the disclosed memory cell can include one or more gate stacks in direct contact with at least one of its respective source / drain components, while the read transistor can include one or more gate stacks electrically isolated from its respective source / drain components by internal spacers.

[0018] Figure 1An exemplary circuit diagram of a memory cell 100 according to some embodiments is shown. As shown, the memory cell (or sometimes referred to as a storage bit cell, storage bit or bit) 100 includes a first transistor 110 and a second transistor 120. Each of the first transistor 110 and the second transistor 120 can include an n-type metal oxide semiconductor field effect transistor (MOSFET). The transistor 110 and the transistor 120 can each include another type of MOSFET, for example, a p-type MOSFET. In some other embodiments, at least one of the transistor 110 or the transistor 120 can be replaced by another type of electronic device (e.g., a MOS capacitor), while still remaining within the scope of the present invention. The first transistor 110 and the second transistor 120 are electrically coupled in series with each other. For example, the source 110S of the first transistor is connected to the drain 120D of the second transistor.

[0019] The memory cell 100 may be configured as a one-time programmable (OTP) memory cell, such as an anti-fuse cell, for example. It should be understood that the memory cell 100 may be configured as any type of memory cell including two transistors electrically coupled in series with each other (e.g., a NOR type non-volatile memory cell, a dynamic random access memory (DRAM) cell, a two-transistor static random access memory (SRMA) cell, etc.).

[0020] When the memory cell 100 is configured as an anti-fuse cell, the first transistor 110 can function as a programming transistor, and the second transistor 120 can function as a reading transistor. Thus, the drain 110D of the first transistor is floating (e.g., without any coupling), and the gate 110G of the first transistor is coupled to a programming word line (WLP) 130; and the gate 120G of the second transistor is coupled to a reading word line (WLR) 132, and the source 120S of the second transistor is coupled to a bit line (BL) 134.

[0021] To program the memory cell 100, the read transistor 120 is turned on by providing a high voltage (e.g., a positive voltage corresponding to a logic high state) to the gate 120G via the WLR 132. Before, during, or after turning on the read transistor 120, a sufficiently high voltage (e.g., a breakdown voltage (V BD )) is applied to WLP 130, and a low voltage (eg, a positive voltage corresponding to a logic low state) is applied to BL 134. The low voltage (applied to BL 134) can be transferred to source 110S, thereby generating V across source 110S and across gate 110G. BDThe portion of the programming transistor 110 that is used to break down the gate dielectric (e.g., the portion between the source 110S and the gate 110G) of the programming transistor 110. After the breakdown of the gate dielectric of the programming transistor 110, the behavior of the portion interconnecting the gate 110G and the source 110S belongs to an equivalent resistance. For example, such a portion can play the role of the resistor 136. Before programming (before the gate dielectric of the programming transistor 110 is broken down), when the read transistor 120 is turned on, there is no conduction path between the BL 134 and the WLP 130; and after programming, when the read transistor 120 is turned on, there is a conduction path between the BL 134 and the WLP 130 (e.g., via the resistor 136).

[0022] To read the memory cell 100, similar to programming, the read transistor 120 is turned on and the BL 134 is coupled to a voltage corresponding to a logic low state. In response, a positive voltage is applied to the gate 110G of the programming transistor. As discussed above, if the gate dielectric of the programming transistor 110 is not broken down, there is no conduction path between the BL 134 and the WLP 130. Therefore, a relatively low current is conducted from the WLP 130 to the BL 134 via the transistor 110 and the transistor 120. If the gate dielectric of the programming transistor 110 is broken down, there is a conduction path between the BL 134 and the WLP 130. Therefore, a relatively high current is conducted from the WLP 130 to the BL 134 via the transistor 110 (now equivalent to the resistor 136) and the transistor 120. Such low current and high current may sometimes be referred to as the I and I of the memory cell 100, respectively. off and I on A circuit element (eg, a sense amplifier) ​​coupled to BL 134 can convert I off with I on The memory cell 100 can be distinguished from the memory cell 100 by a logic high (“1”) or a logic low (“0”). on When 1 is read, the memory cell 100 may present a 1; and when 1 is read off , memory cell 100 may present a 0.

[0023] Figure 2 1 is a flowchart of a method 200 for forming a memory device according to one or more embodiments of the present invention. The method 200 may be used to form an anti-fuse memory cell, the anti-fuse memory cell comprising a programming transistor and a reading transistor coupled in series. It should be noted that the method 200 is merely an example and is not intended to limit the present invention. Therefore, it should be understood that the method 200 may be used to form an anti-fuse memory cell. Figure 2 Additional operations are provided before, during, and after method 200, and only some of the other operations are briefly described herein.

[0024] The operations of method 200 may be performed with Figure 3-1 , Figure 3-2 , Figure 3-3 , Figure 3-4 , Figure 3-5 , Figure 3-6 , Figure 3-7 , Figure 3-8 , Figure 3-9 , Figure 3-10 , Figure 3-11 , Figure 3-12 and Figure 3-13 . In some embodiments, the memory device may be included in or coupled to a microprocessor, another memory device, and / or an integrated circuit (IC). Similarly, in order to better understand the concept of the present invention, Figure 3-1 to Figure 3-13 Although the figures show memory devices, it should be understood that the IC may include several other devices, such as inductors, resistors, capacitors, transistors, etc., which are not shown in the figure for the sake of simplicity. Figure 3-1 to Figure 3-13 Shown in.

[0025] First refer to Figure 2 In a brief overview, the method 200 begins at operation 202, where a substrate is provided. The method 200 proceeds to operation 204, where an alternating series of first nanostructures and second nanostructures are formed. The method 200 proceeds to operation 206, where a number of dummy gate stacks are formed. The method 200 proceeds to operation 208, where a first alternating column of nanostructures and a second alternating column of nanostructures are defined. The method 200 proceeds to operation 210, where the first alternating column of nanostructures is covered. The method 200 proceeds to operation 212, where respective ends of the first nanostructures of the second alternating column of nanostructures are removed. The method 200 proceeds to operation 214, where internal spacers are formed in the second alternating column of nanostructures. The method 200 proceeds to operation 216, where source features and drain features are formed. The method 200 proceeds to operation 218, where an interlayer dielectric is deposited. The method 200 proceeds to operation 220, where the dummy gate stacks are removed. The method 200 proceeds to operation 222, where the first nanostructures of the first alternating column of nanostructures and the second alternating column of nanostructures are removed. The method 200 proceeds to operation 224 where a gate dielectric is deposited. The method 200 proceeds to operation 226 where a gate metal is deposited.

[0026] Corresponding to operation 202, Figure 3-13 is a cross-sectional view of a memory device 300 cut by a plane perpendicular to the Y direction at one of the various stages of manufacturing, and the memory device 300 includes a substrate 302. The substrate 302 includes a semiconductor material substrate, such as silicon. Alternatively, the substrate 302 may include, for example, another basic semiconductor material, such as germanium. The substrate 302 may also include a compound semiconductor, such as silicon carbide, gallium arsenide, indium arsenide, and indium phosphide. The substrate 302 may include an alloy semiconductor, such as silicon germanium, silicon germanium carbide, gallium arsenic phosphide, and gallium indium phosphide. In an embodiment, the substrate 302 includes an epitaxial layer. For example, the substrate may have an epitaxial layer covering a bulk semiconductor. For example, the substrate 302 may include a semiconductor on an insulator (SOI) structure. For example, the substrate 302 may include a buried oxide (BOX) layer formed by a process such as isolation by implantation of oxygen (SIMOX) or other suitable techniques such as wafer bonding and grinding.

[0027] Corresponding to operation 204, Figure 3-2 3 is a cross-sectional view of a memory device 300 cut by a plane perpendicular to the Y direction at one of the various stages of fabrication, the memory device 300 including an alternating series of first semiconductor layers 304, 308, and 312 and second semiconductor layers 306, 310, and 314. The first semiconductor layers 304, 308, and 312 may include SiGe nanostructures (hereinafter referred to as "SiGe nanostructures 304, 308, and 312"), and the second semiconductor layers 306, 310, and 314 may include Si nanostructures (hereinafter referred to as "Si nanostructures 306, 310, and 314"). In an embodiment, each of the SiGe nanostructures 304, 308, and 312 may include a SiGe layer (e.g., a SiGe nanosheet) having a thickness in the nanometer range; and each of the Si nanostructures 306, 310, and 314 may include a Si layer (e.g., a Si nanosheet) having a thickness in the nanometer range. It should be understood that the thickness of SiGe nanostructures 304, 308 and 312 and Si nanostructures 306, 310 and 314 can be reduced to the sub-nanometer range (e.g., angstroms) while still remaining within the scope of the present invention. The alternating series of SiGe nanostructures 304, 308 and 312 and Si nanostructures 306, 310 and 314 can be formed as a stack above substrate 302, wherein the nanostructures 304-314 are arranged on top of each other along a vertical direction (e.g., Z direction). Such stacks can sometimes be referred to as superlattices. In a non-limiting example, SiGe nanostructures 304, 308 and 312 are in the range from 10% to 40%. It should be understood that the percentage of Ge in each of SiGe nanostructures 304, 308 and 312 can be any value between 0 and 100 (excluding 0 and 100) while still remaining within the scope of the present invention.

[0028] An alternating series of nanostructures can be formed by epitaxially growing one layer and then the next until the desired number and desired thickness of the nanostructures are reached. The epitaxial material can be grown from a vapor phase precursor or a liquid phase precursor. The epitaxial material can be grown using vapor phase epitaxy (VPE), molecular beam epitaxy (MBE), liquid phase epitaxy (LPE), or other suitable processes. Depending on the type of transistor, epitaxial silicon, silicon germanium, and / or carbon-doped silicon (Si:C) silicon can be doped during deposition (in-situ doping) by adding dopants, n-type dopants (e.g., phosphorus or arsenic) or p-type dopants (e.g., boron or gallium).

[0029] Corresponding to operation 206, Figure 3-3 is a cross-sectional view of a memory device 300 cut by a plane perpendicular to the Y direction at one of the various stages of manufacturing, the memory device 300 including a first dummy gate stack 316a and a second dummy gate stack 316b. Each of the dummy gate stacks 316a-b includes a dummy gate and a hard mask. For example, in Figure 3-3 In the embodiment, the first dummy gate stack 316a includes a dummy gate 318a formed above the Si nanostructure 314 and a hard mask 320a formed above the dummy gate 318a; and the second dummy gate stack 316b includes a dummy gate 318b formed above the Si nanostructure 314 and a hard mask 320b formed above the dummy gate 318b.

[0030] In some embodiments, the dummy gate stacks 316a-b may correspond to regions where gate features of the programming transistor and the reading transistor of the memory device 300 will be formed. Figure 3-3 In the embodiment of the present invention, each of the dummy gate stacks 316a-b is shown as a two-dimensional structure, but it should be understood that each of the dummy gate stacks 316a-b is formed as a three-dimensional structure to span the first nanostructures 304, 308 and 312 and the second nanostructures 306, 310 and 314. For example, each of the dummy gate stacks 316a-b can be formed above and around the sidewalls of the first nanostructures 304, 308 and 312 and the second nanostructures 306, 310 and 314. The dummy gates 318a-b can be formed by depositing amorphous silicon (a-Si) above and around the alternating series of first nanostructures 304, 308 and 312 and second nanostructures 306, 310 and 314. The a-Si is then planarized to a desired level. A hard mask (not shown) is deposited over the planarized a-Si and patterned to form hard masks 320a-b. The hard masks 320a-b can be formed by a nitride layer or an oxide layer. An etching process (eg, a reactive ion etching (RIE) process) is applied to the a-Si to form the dummy gate stacks 316 a - b .

[0031] After forming the dummy gate stacks 316a-b, gate spacers 322a and 322b may be formed to extend along respective sidewalls of the dummy gate stacks 316a and 316b, such as Figure 3-3 As shown. The gate spacers 322a-b can be formed using a spacer pull-down formation process. The gate spacers 322a-b can also be formed by conformal deposition of a dielectric material (e.g., silicon oxide, silicon nitride, silicon oxynitride, SiBCN, SiOCN, SiOC, or any suitable combination of those materials) followed by directional etching (e.g., RIE). Such gate spacers may sometimes be referred to as external spacers. In some embodiments, the dummy gate stack 316a together with the gate spacer 322a can extend a first distance D1 along a horizontal direction (e.g., the X direction); and the dummy gate stack 316b together with the gate spacer 322b can extend a second distance D2 along the X direction, as shown. Figure 3-3 The first distance D1 and the second distance D2 may be the same as or different from each other.

[0032] Corresponding to operation 208, Figure 3-4 3 is a cross-sectional view of a memory device 300 cut through a plane perpendicular to the Y direction at one of the various stages of fabrication, the memory device 300 including alternating nanostructure columns 324a and 324b. After forming the gate spacers 322a-b, the alternating nanostructure columns 324a and 324b may be formed by at least some of the following processes: using the gate spacers 322a-b, the dummy gates 318a-b, and the hard masks 320a-b used as masks to define the footprints of the alternating nanostructure columns 324a and 324b, and etching the alternating series of first nanostructures 304, 308, and 312 and second nanostructures 306, 310, and 314 (e.g., Figure 3-3 ) to form alternating nanostructure columns 324a and 324b. As such, each of alternating nanostructure columns 324a and 324b includes a stack of alternating etched SiGe / Si nanostructures. For example, alternating nanostructure column 324a includes alternating stacks of etched SiGe nanostructures 325a, etched Si nanostructures 326a, etched SiGe nanostructures 327a, etched Si nanostructures 328a, etched SiGe nanostructures 329a, and etched Si nanostructures 330a; and alternating nanostructure column 324b includes alternating stacks of etched SiGe nanostructures 325b, etched Si nanostructures 326b, etched SiGe nanostructures 327b, etched Si nanostructures 328b, etched SiGe nanostructures 329b, and etched Si nanostructures 330b.

[0033] In some embodiments, each of the etched Si nanostructures and the etched SiGe nanostructures of the alternating nanostructure column 324a can follow the horizontal dimensions of the dummy gate stack 316a and the gate spacer 322a; and each of the etched Si nanostructures and the etched SiGe nanostructures of the alternating nanostructure column 324b can follow the horizontal dimensions of the dummy gate stack 316b and the gate spacer 322b. Therefore, each of the etched Si nanostructures and the etched SiGe nanostructures of the alternating nanostructure column 324a can extend D1 along the X direction; and each of the etched Si nanostructures and the etched SiGe nanostructures of the alternating nanostructure column 324b can extend D2 along the X direction.

[0034] Corresponding to operation 210, Figure 3-5 3 is a cross-sectional view of a memory device 300 cut by a plane perpendicular to the Y direction at one of the various stages of manufacturing, and the memory device 300 includes a shielding mask 334. In some embodiments, the shielding mask 334 is formed to cover the alternating nanostructure columns 324a, while the alternating nanostructure columns 324b remain exposed. The shielding mask 334 can be formed to have a sufficiently large thickness (or height) so that the respective sidewalls of each of the etched SiGe nanostructures 325a, 327a, and 329a are completely covered. The formation of the shielding mask 334 can allow one or more processes (which will be discussed below) to be performed on the alternating nanostructure columns 324b. The shielding mask 334 can be formed of a material that is relatively resistant to an etchant that can, for example, etch SiGe, such as silicon oxide, silicon nitride, silicon oxynitride, SiBCN, SiOCN, SiOC, or any suitable combination of those materials.

[0035] Corresponding to operation 212, Figure 3-6 3 is a cross-sectional view of the memory device 300 cut by a plane perpendicular to the Y direction at one of the various stages of manufacturing, wherein the etched SiGe nanostructure 325b, the etched SiGe nanostructure 327b and the etched SiGe nanostructure 329b (eg, Figure 3-5329a). In the process of removing the respective ends of the etched SiGe nanostructure 325b, the etched SiGe nanostructure 327b, and the etched SiGe nanostructure 329b, the alternating nanostructure column 324a is covered by the shielding mask 334. In this way, the SiGe nanostructures 335b, 337b, and 339b can be formed to extend a distance less than D2 (e.g., D3) along the X direction, while the SiGe nanostructures 325a, 327a, and 329a of the alternating nanostructure column 324a still extend D1 along the X direction. The SiGe nanostructures 325a, 327a, 329a, 335b, 337b, and 339b can then be replaced by several gate stacks. Thus, SiGe nanostructures 325a, 327a, and 329a may be referred to below as SiGe sacrificial nanostructures 325a, 327a, and 329a corresponding to alternating nanostructure columns 324, and SiGe nanostructures 335b, 337b, and 339b may be referred to below as SiGe sacrificial nanostructures 335b, 337b, and 339b corresponding to alternating nanostructure columns 324b.

[0036] In some embodiments of the present invention, the etched SiGe nanostructures 325b, 327b, and 329b can be removed using a first application, which is referred to as a "pull-back process" for pulling the etched SiGe nanostructures 325b, 327b, and 329b back an initial pull-back distance so that the ends of the SiGe sacrificial nanostructures 335b, 337b, and 339b terminate under (e.g., are aligned with) the gate spacer 322b. Figure 3-6 In the illustrated embodiment, the ends of each of the SiGe sacrificial nanostructures 335b, 337b, and 339b are substantially aligned with the sidewalls of the spacer 322b, but it should be understood that the pull-back distance (i.e., the extent to which each of the SiGe sacrificial nanostructures 335b, 337b, and 339b is etched or pulled back) can be increased or decreased arbitrarily. The pull-back process can include anisotropic etching with hydrogen chloride (HCl) gas, which etches SiGe without attacking Si.

[0037] Corresponding to operation 214, Figure 3-73 is a cross-sectional view of a memory device 300 cut by a plane perpendicular to the Y direction at one of the various stages of fabrication, the memory device 300 including internal spacers 340, 342, and 344. During the formation of the internal spacers 340-344, the alternating nanostructure column 324a is covered by the shielding mask 334. As such, only the alternating nanostructure column 324a has the internal spacers 340-344 arranged along the respective sidewalls of the SiGe sacrificial nanostructures 335b, 337b, and 339b. In some embodiments, the internal spacers 340-344 can be conformally formed by chemical vapor deposition (CVD) or chloride monolayer doping (MLD), followed by spacer RIE. In some embodiments, the internal spacers 340-344 can be deposited using, for example, a conformal deposition process and subsequent isotropic or anisotropic etch back to remove excess spacer material on the vertical sidewalls of the alternating nanostructure column 324b and on the surface of the semiconductor substrate 302. Thus, the inner spacers 340-344 may extend a distance D4 along the X direction, and the distance D4 is approximately half of the difference between D2 and D3. The material of the inner spacers 340-344 may be formed of the same or different material (e.g., silicon nitride) as the gate spacers 322a-b. For example, the material of the inner spacers 340-344 may be formed of silicon nitride, silicon boron carbonitride, carbon boron nitride, silicon carbon oxynitride, or any other type of dielectric material (e.g., a dielectric material having a dielectric constant k less than about 5) suitable for functioning as an isolation gate sidewall spacer for forming a FET device.

[0038] Corresponding to operation 216, Figure 3-8 3 is a cross-sectional view of a memory device 300 cut through a plane perpendicular to the Y direction at one of the various stages of manufacturing, and the memory device 300 includes a drain feature 346, a source feature 348, a drain feature 350, and a source feature 352. The drain / source features 346-352 may be formed after removing the shield mask 334 (e.g., Figure 3-7). In some embodiments, drain feature 346 may be formed by using an epitaxial layer growth process on the exposed ends of etched Si nanostructures 326a, 328a, and 330a on the left hand side of alternating nanostructure column 324a. Source 348 may be formed by using an epitaxial layer growth process on the exposed ends of etched Si nanostructures 326a, 328a, and 330a on the left hand side of alternating nanostructure column 324a. Drain 350 may be formed by using an epitaxial layer growth process on the exposed ends of etched Si nanostructures 326b, 328b, and 330b on the left hand side of alternating nanostructure column 324b. Source 352 may be formed by using an epitaxial layer growth process on the exposed ends of etched Si nanostructures 326b, 328b, and 330b on the right hand side of alternating nanostructure column 324b. In some embodiments, source 348 and drain 350 may merge with each other to form a continuous feature or region, such as Figure 3-8 shown.

[0039] According to some embodiments, drain component 346 and source component 348 are electrically coupled to Si nanostructures 326a, 328a, and 330a; and drain component 350 and source component 352 are electrically coupled to Si nanostructures 326b, 328b, and 330b. Si nanostructures 326a, 328a, and 330a may collectively constitute a conduction channel of a first transistor 354a; and Si nanostructures 326b, 328b, and 330b may collectively constitute a conduction channel of a second transistor 354b. In an example where the memory device is an anti-fuse cell, first transistor 354a may function as a programming transistor, and second transistor 354b may function as a read transistor and be electrically connected in series to programming transistor 354a.

[0040] In-situ doping (ISD) may be applied to form doped drain / source features 346-352 to produce the necessary junctions corresponding to programming transistor 354a and reading transistor 354b. N-type FETs and p-type FETs are formed by implanting different types of dopants corresponding to selected regions of the device (e.g., drain / source features 346-352) to form one or more necessary junctions. N-type devices can be formed by implanting arsenic (As) or phosphorus (P), and p-type devices can be formed by implanting boron (B).

[0041] Corresponding to operation 218, Figure 3-9is a cross-sectional view of memory device 300 cut through a plane perpendicular to the Y direction at one of the various stages of fabrication, memory device 300 including an interlayer dielectric (ILD) material 356. ILD material 356 can be formed by depositing a bulk oxide material (e.g., silicon dioxide) and polishing the bulk oxide (e.g., using CMP) back to the level of gate spacers 322a-b and hard masks 320a-b.

[0042] Corresponding to operation 220, Figure 3-10 is a cross-sectional view of the memory device 300 cut through a plane perpendicular to the Y direction at one of the various stages of fabrication, wherein the dummy gate stack 316a-b ( Figure 3-9 After forming the protective ILD material 356, the dummy gate stack 316a (including the dummy gate 318a and the hard mask 320a) and the dummy gate stack 316b (including the dummy gate 318b and the hard mask 320b) are removed. Figure 3-9 The dummy gate stacks 316a-b can be removed by a known etching process, such as RIE or chemical oxide removal (COR).

[0043] After the dummy gate stacks 316a-b are removed, the respective top boundaries of the alternating nanostructure columns 324a and 324b can be exposed again. Specifically, the respective top boundaries of the etched Si nanostructures 330a of the alternating nanostructure column 324a and the etched Si nanostructures 330b of the alternating nanostructure column 324b can be exposed. Figure 3-10 , but it should be understood that in addition to the top boundary, the respective sidewalls of the alternating nanostructure columns 324a and 324b may also be exposed along the Y direction.

[0044] Corresponding to operation 222, Figure 3-11 3 is a cross-sectional view of the memory device 300 cut through a plane perpendicular to the Y direction at one of the various stages of fabrication, wherein the SiGe sacrificial nanostructures 325a, 327a, and 329a of the alternating nanostructure column 324a and the SiGe sacrificial nanostructures 335b, 337b, and 339b of the alternating nanostructure column 324b are removed (at Figure 3-10 ). A selective etch (eg, hydrochloric acid (HCl)) can be applied to remove the SiGe sacrificial nanostructures 325a, 327a, 329a, 335b, 337b, and 339b.

[0045] After the removal of the SiGe sacrificial nanostructures 325a, 327a, 329a, 335b, 337b, and 339b, the respective bottom boundaries of the etched Si nanostructures 326a, 328a, and 330a of the alternating nanostructure column 324a and the etched Si nanostructures 326b, 328b, and 330b of the alternating nanostructure column 324b can be exposed. As described above, the etched Si nanostructures 326a, 328a, and 330a of the alternating nanostructure column 324a can be collectively configured as the conductive channel of the programming transistor 354a; and the etched Si nanostructures 326b, 328b, and 330b of the alternating nanostructure column 324b can be collectively configured as the conductive channel of the reading transistor 354b. Thus, etched Si nanostructures 326a, 328a, and 330a may be referred to herein as "conductive channels 360a"; and etched Si nanostructures 326b, 328b, and 330b may be referred to herein as "conductive channels 360b."

[0046] Conductive channels 360a and 360b are configured to conduct current through programming transistor 354a and reading transistor 354b, respectively. In general, such conductive channels have a length and a width. The length can be parallel to the current flow, and the width can be perpendicular to the current flow. Figure 3-11 As shown, conduction channel 360a may be characterized as having a length of approximately D1, and conduction channel 360b may be characterized as having a length of approximately D2. Although three Si nanostructures form the conduction channels of programming transistor 354a and read transistor 354b of memory device 300, it should be understood that a memory device fabricated by the methods disclosed herein can include any number of nanostructures to form one or more conduction channels thereof while remaining within the scope of the present invention.

[0047] Corresponding to operation 224, Figure 3-12364a and 364b. 364a and 364b are cross-sectional views of a memory device 300 cut by a plane perpendicular to the Y direction at one of the various stages of manufacturing, and the memory device 300 includes gate dielectrics 364a and 364b. As shown, the gate dielectric 364a can surround each of the Si nanostructures (Si nanostructures 326a, 328a, and 330a) of the conductive channel 360a; and the gate dielectric 364b can surround each of the Si nanostructures (Si nanostructures 326b, 328b, and 330b) of the conductive channel 360b. The gate dielectrics 364a and 364b include different high-k dielectric materials or the same high-k dielectric materials. The gate dielectrics 364a and 364b can include a stack of multiple high-k dielectric materials. Any suitable method can be used, including, for example, atomic layer deposition (ALD), to deposit the gate dielectrics 364a and 364b. In some embodiments, gate dielectric 364a and gate dielectric 364b may optionally include a substantially thinner oxide (eg, SiO x In some embodiments, gate dielectrics 364a-b may be formed as substantially conformal layers characterized by having a thickness D5 and a thickness D6, respectively.

[0048] Corresponding to operation 226, Figure 3-13 366a-b is a cross-sectional view of a memory device 300 cut by a plane perpendicular to the Y direction at one of the various stages of manufacturing, and the memory device 300 includes gate metals 366a and 366b. In some embodiments, the gate metal 366a can surround each of the Si nanostructures of the conductive channel 360a with the gate dielectric 364a arranged therebetween; and the gate metal 366b can surround each of the Si nanostructures of the conductive channel 360b with the gate dielectric 364b arranged therebetween. The gate metals 366a-b can be formed of different metal materials or the same metal material. The gate metals 366a-b can include a stack of multiple metal materials. Any suitable method can be used, including, for example, CVD, to deposit the gate metals 366a-b.

[0049] Although each of gate metals 366a-b is shown as Figure 3-13 , but it should be understood that each of the gate metals 366a-b is formed as a three-dimensional structure. Specifically, each of the gate metals 366a-b can include a plurality of gate metal portions separated from each other along the Z direction. Each of the gate metal portions can extend not only along a horizontal plane (e.g., a plane extending in the X direction and the Y direction), but can also extend along a vertical direction (e.g., the Z direction). In this way, two adjacent gate metal portions can be adjacent to each other so as to surround the corresponding Si nanostructures arranged therebetween with the gate dielectric.

[0050] For example, in Figure 3-13 , the gate metal 366a can include gate metal parts 366a1, 366a2, 366a3 and 366a4. The gate metal parts 366a1 and 366a2 can be adjacent to each other to surround the Si nanostructure 330a with a portion of the gate dielectric 364a arranged therebetween. The gate metal parts 366a2 and 366a3 can be adjacent to each other to surround the Si nanostructure 328a with a portion of the gate dielectric 364a arranged therebetween. The gate metal parts 366a3 and 366a4 can be adjacent to each other to surround the Si nanostructure 326a with a portion of the gate dielectric 364a arranged therebetween. Similarly, the gate metal 366b can include gate metal parts 366b1, 366b2, 366b3 and 366b4. The gate metal parts 366b1 and 366b2 can be adjacent to each other to surround the Si nanostructure 330b with a portion of the gate dielectric 364b arranged therebetween. Gate metal portions 366b2 and 366b3 may be adjacent to each other to surround the Si nanostructure 328b with a portion of the gate dielectric 364b disposed therebetween. Gate metal portions 366b3 and 366b4 may be adjacent to each other to surround the Si nanostructure 326b with a portion of the gate dielectric 364b disposed therebetween. In some embodiments, such gate metal portions that at least partially surround one of the Si nanostructures together with the corresponding portion of the gate dielectric may be referred to as a gate stack. The gate stack may be operably associated with the surrounded Si nanostructure (e.g., to modulate the current conducted into the Si nanostructure). The gate stack may sometimes be referred to as a full surround gate stack.

[0051] In some embodiments, several gate stacks consisting of portions of gate metal 366a and gate dielectric 364a can function as gate components of programming transistor 354a to modulate current conducted from drain component 346 to source component 348 via conduction channel 360a; and several gate stacks consisting of portions of gate metal 366b and gate dielectric 364b can function as gate components of reading transistor 354b to modulate current conducted from drain component 350 to source component 352 via conduction channel 360b.

[0052] exist Figure 3-1 to Figure 3-13In the illustrated embodiment, no internal spacers are formed in the alternating nanostructure columns 324a, and the gate stack of the programming transistor 354a can be in direct contact with the respective drain features 346 and 348. As such, each of the Si nanostructures (e.g., 326a, 328a, 330a) of the conductive channel 360a can extend a distance (or length) along the X direction that is substantially equal to a distance (or length) (e.g., about D1) that each of the gate stacks extends along the X direction. The distance that the conductive channel 360a extends along the X direction can sometimes be referred to as the channel length of the programming transistor 354a. Specifically, each of the gate metal portions 366a2-4 can be electrically coupled to the drain / source features 346 and 348 between which the gate dielectric 364a is disposed. As such, each of the gate metal portions 366a2-4 extends a distance of about D1–2×D5 along the X direction.

[0053] On the other hand, the internal spacers 340-344 are formed in the alternating nanostructure columns 324b, and the gate stack of the read transistor 354b can be electrically isolated from the respective drain features 350 and 352 by the internal spacers 340-344. As such, each of the Si nanostructures (e.g., 326b, 328b, 330b) of the conductive channel 360b can extend a distance (or length) along the X direction, for example, about D1, which is substantially greater than the distance (or length) along the X direction that each of the gate stacks extends (e.g., about D2-2×D4). The distance along the X direction that the conductive channel 360b extends can sometimes be referred to as the channel length of the read transistor 354b. Specifically, each of the gate metal portions 366b2-4 can be electrically isolated from the drain feature 350 and the source feature 352 disposed therebetween by the gate dielectric 364b and a corresponding one of the internal spacers 340-344. . Thus, each of the gate metal portions 366 b 2 - 4 extends along the X direction by a distance of approximately D2 − 2×D4 − 2×D6 .

[0054] By forming the programming transistor and the reading transistor of the memory cell in such a configuration, the contact area of ​​each of the gate stacks with the source / drain features corresponding to the programming transistor can be increased, which may advantageously reduce the V BD and T BD At the same time, retaining the internal spacer corresponding to the read transistor can effectively suppress the parasitic capacitance from adjusting the switching speed of the read transistor.

[0055] After forming gate metals 366a-b, one or more interconnect structures may be formed to connect each of gate metal 366a, gate metal 366b, and source feature 352 to connect memory device 300 to other elements or devices. For example, one or more interconnect structures (e.g., via structures, generally referred to as VG) may be formed over gate metal 366a to connect it to one or more upper metal layers, which may include programming word lines (WLP); one or more interconnect structures (e.g., VG) may be formed over gate metal 366b to connect it to one or more upper metal layers, which may include reading word lines (WLR); and one or more interconnect structures (e.g., metal structures, generally referred to as MD, via structures, generally referred to as VD) may be formed in or over ILD 356 and over source feature 352 to connect it to one or more upper metal layers, which may include bit lines (BL). In this way, memory device 300, as an exemplary anti-fuse memory cell, can be connected to one or more other memory cells similar to memory device 300. For example, several such memory devices 300 may be arranged (eg, coupled) to each other through respective WLPs, read WLs, and BLs to form a memory array.

[0056] Figure 4 4 shows a cross-sectional view of another exemplary memory device 400 according to some embodiments. The memory device 400 may be substantially similar to the memory device 300 ( Figure 3-1 to Figure 3-13 ), except for the case where both the programming transistor and the reading transistor of the memory device 400 include an internal spacer. Therefore, the following discussion will focus on the differences between the memory devices 300 and 400.

[0057] As shown, the memory device 400 includes a programming transistor 404a and a reading transistor 404b formed on a substrate 402. Similar to the programming transistor 354a, the programming transistor 404a also includes a gate metal 406a, a gate dielectric 408a, a gate spacer 409a, several Si nanostructures that collectively function as a conductive channel 410a, a drain feature 412, and a source feature 414. Similar to the reading transistor 354b, the reading transistor 404b also includes a gate metal 406b, a gate dielectric 408b, a gate spacer 409b, several Si nanostructures that collectively function as a conductive channel 410b, a drain feature 416, and a source feature 418. At least a portion of each of the programming transistor 404a and the reading transistor 404b is embedded in the ILD 420.

[0058] Unlike the memory device 300, both the programming transistor 404a and the reading transistor 404b include internal spacers. Specifically, the gate metal 406a of the programming transistor 404a includes gate metal portions 406a1, 406a2, 406a3, and 406a4. The gate metal portion 406a1 and the gate dielectric 408a may constitute a first of several gate stacks corresponding to the programming transistor 404a; the gate metal portion 406a2 and the gate dielectric 408a may constitute a second of several gate stacks corresponding to the programming transistor 404a; the gate metal portion 406a3 and the gate dielectric 408a may constitute a third of several gate stacks corresponding to the programming transistor 404a; and the gate metal portion 406a4 and the gate dielectric 408a may constitute a fourth of several gate stacks corresponding to the programming transistor 404a. Each of the gate stacks may at least partially surround a corresponding Si nanostructure of the conductive channel 410a.

[0059] Likewise, gate metal 406b of read transistor 404b includes gate metal portions 406b1, 406b2, 406b3, and 406b4. Portions of gate metal portion 406b1 and gate dielectric 408b may constitute a first of several gate stacks corresponding to read transistor 404b; portions of gate metal portion 406b2 and gate dielectric 408b may constitute a second of several gate stacks corresponding to read transistor 404b; portions of gate metal portion 406b3 and gate dielectric 408b may constitute a third of several gate stacks corresponding to read transistor 404b; and portions of gate metal portion 406b4 and gate dielectric 408b may constitute a fourth of several gate stacks corresponding to read transistor 404b. Each of the gate stacks may at least partially surround a corresponding Si nanostructure of conduction channel 410b.

[0060] Some of the gate stacks of the programming transistor 404a are isolated from respective drain / source features 412 and 414 by internal spacers 424 (e.g., a gate stack including gate metal portion 406a2, a gate stack including gate metal portion 406a3, and a gate stack including gate metal portion 406a4). Some of the gate stacks of the reading transistor 404b are isolated from respective drain / source features 416 and drain / source features 418 by internal spacers 426 (e.g., a gate stack including gate metal portion 406b2, a gate stack including gate metal portion 406b3, and a gate stack including gate metal portion 406b4). In some embodiments, the internal spacers 424-426 may be formed of a dielectric material selected from silicon oxide, silicon nitride, silicon oxynitride, SiBCN, SiOCN, SiOC, or combinations thereof.

[0061] reaches V which reduces the programming transistor 404a. BD / T BD While suppressing the degree of parasitic capacitance of the read transistor 404b, the internal spacer 424 and the internal spacer 426 have different effective capacitances. In some embodiments, the thickness of the internal spacer 424 is substantially equal to the thickness of the internal spacer 426, but the dielectric constant of the internal spacer 424 is different from the dielectric constant of the internal spacer 426. For example, the internal spacer 424 can be formed of a dielectric material characterized by having a higher dielectric constant than the dielectric constant of the dielectric material of the internal spacer 426. In another example, the internal spacers 424 and 426 can be formed to have different geometric dimensions. Each of the internal spacers 424 can extend a distance (sometimes referred to as the width of the internal spacer) D7 along the X direction and a distance (sometimes referred to as the height of the internal spacer) D9 along the Z direction; and each of the internal spacers 426 can extend a distance (width) D8 along the X direction and a distance (height) D9 along the Z direction. 10 In some embodiments, the sum of the D7's of the internal spacer 424 can be selected to be less than the sum of the D8's of the internal spacer 426; and / or the sum of the D9s of the internal spacer 424 can be selected to be less than the sum of the D 10 In at least one embodiment, the internal spacer 424 and the internal spacer 426 have different geometric dimensions but have the same dielectric constant.

[0062] Given the non-zero thickness of the internal spacers 424 and 426, each of the Si nanostructures of the conductive channels 410a-b can be characterized by having a channel length that is greater than a distance that the corresponding gate stack extends along the X direction. Figure 4 In the embodiment, the Si nanostructure of the conductive channel 410a may have a distance (eg, D ) greater than that of the corresponding gate stack extending along the X direction. 11 –2×D7) channel length D 11 In another embodiment, the Si nanostructure of the conductive channel 410b may have a distance (eg, D ) greater than that of the corresponding gate stack extending along the X direction. 12 –2×D8) channel length D 12 .

[0063] To manufacture the memory device 400, a substantially similar Figure 2The method of method 200 may be performed in a manner similar to that of method 200. For example, operation 210 may be skipped. In this way, the first alternating nanostructure column is also exposed, while the ends of the first nanostructures of the first alternating nanostructure column are removed (operation 212). In addition, operation 212 may be changed to remove the ends of the first nanostructures of both the first alternating nanostructure column and the second alternating nanostructure column.

[0064] Figure 5 , Figure 6 and Figure 7 Various exemplary nanostructure transistors including partial internal spacers are shown in accordance with some embodiments. The term "partial internal spacer" refers to an internal spacer of a nanostructure transistor that does not completely isolate a corresponding gate stack from a respective source component or drain component. In contrast, the term "full internal spacer" refers to an internal spacer of a nanostructure transistor that completely isolates a corresponding gate stack from a respective source component or drain component. In some embodiments, nanostructure transistors having partial internal spacers are not limited to programming transistors or read transistors configured as anti-fuse cells. However, to reduce the V of the programming transistor BD / T BD After simultaneously suppressing the degree of parasitic capacitance of the read transistor, the programming transistor and the read transistor, for example, can be configured as a nanostructure transistor with a partial internal spacer and a nanostructure transistor with a complete internal spacer, respectively. In another example, the programming transistor and the read transistor can be configured as a nanostructure transistor with a partial internal spacer and a nanostructure transistor also with a partial internal spacer, respectively, but the sum of the widths (or heights) of the partial internal spacers in the programming transistor is less than the sum of the widths (or heights) of the partial internal spacers in the read transistor.

[0065] Reference Figure 5 , a nanostructure transistor 500 including a portion of an internal spacer is shown. The nanostructure transistor 500 is formed on a substrate 502. The nanostructure transistor 500 includes a gate metal 506, a gate dielectric 508, a gate spacer 510, several Si nanostructures that collectively function as a conduction channel 512, a drain feature 514, and a source feature 516. At least a portion of the nanostructure transistor 500 is embedded in an ILD 518.

[0066] The gate metal 506 includes gate metal portions 506a, 506b, 506c, and 506d. The gate metal portion 506a and a portion of the gate dielectric 508 may constitute a first of several gate stacks corresponding to the nanostructure transistor 500; the gate metal portion 506b and a portion of the gate dielectric 508 may constitute a second of several gate stacks corresponding to the nanostructure transistor 500; the gate metal portion 506c and a portion of the gate dielectric 508 may constitute a third of several gate stacks corresponding to the nanostructure transistor 500; and the gate metal portion 506d and a portion of the gate dielectric 508 may constitute a fourth of several gate stacks corresponding to the nanostructure transistor 500. Each of the gate stacks may be capable of at least partially surrounding a corresponding Si nanostructure of the conductive channel 512. Figure 5 In the illustrated embodiment, the nanostructure transistor 500 includes a partial internal spacer 520 that isolates the corresponding gate stack only from the source feature 516 , but not from both the drain feature 514 and the source feature 516 .

[0067] Reference Figure 6 , another nanostructure transistor 600 including a portion of an internal spacer is shown. The nanostructure transistor 600 is formed on a substrate 602. The nanostructure transistor 600 includes a gate metal 606, a gate dielectric 608, a gate spacer 610, several Si nanostructures that collectively function as a conduction channel 612, a drain feature 614, and a source feature 616. At least a portion of the nanostructure transistor 600 is embedded in an ILD 618.

[0068] The gate metal 606 includes gate metal portions 606a, 606b, 606c, and 606d. The gate metal portion 606a and a portion of the gate dielectric 608 may constitute a first of several gate stacks corresponding to the nanostructure transistor 600; the gate metal portion 606b and a portion of the gate dielectric 608 may constitute a second of several gate stacks corresponding to the nanostructure transistor 600; the gate metal portion 606c and a portion of the gate dielectric 608 may constitute a third of several gate stacks corresponding to the nanostructure transistor 600; and the gate metal portion 606d and a portion of the gate dielectric 608 may constitute a fourth of several gate stacks corresponding to the nanostructure transistor 600. Each of the gate stacks may be capable of at least partially surrounding a corresponding Si nanostructure of the conductive channel 612. In Figure 6 In the illustrated embodiment, the nanostructure transistor 600 includes a partial internal spacer 620 that isolates the corresponding gate stack only from the drain feature 614 , but not from both the drain feature 614 and the source feature 616 .

[0069] Reference Figure 7 , another nanostructure transistor 700 including a portion of an internal spacer is shown. The nanostructure transistor 700 is formed on a substrate 702. The nanostructure transistor 700 includes a gate metal 706, a gate dielectric 708, a gate spacer 710, several Si nanostructures that collectively function as a conduction channel 712, a drain feature 714, and a source feature 716. At least a portion of the nanostructure transistor 700 is embedded in an ILD 718.

[0070] The gate metal 706 includes gate metal portions 706a, 706b, 706c, and 706d. The gate metal portion 706a and a portion of the gate dielectric 708 may constitute a first of several gate stacks corresponding to the nanostructure transistor 700; the gate metal portion 706b and a portion of the gate dielectric 708 may constitute a second of several gate stacks corresponding to the nanostructure transistor 700; the gate metal portion 706c and a portion of the gate dielectric 708 may constitute a third of several gate stacks corresponding to the nanostructure transistor 700; and the gate metal portion 706d and a portion of the gate dielectric 708 may constitute a fourth of several gate stacks corresponding to the nanostructure transistor 700. Each of the gate stacks may be capable of at least partially surrounding a corresponding Si nanostructure of the conductive channel 712. Figure 7 In the illustrated embodiment, nanostructure transistor 700 includes a partial internal spacer 720, an internal spacer 724, and an internal spacer 726, wherein the partial internal spacer 720 only isolates the corresponding gate stack from the source feature 716 (i.e., leaving the other side of the corresponding gate stack in direct contact with the drain feature 714), the internal spacer 724 only isolates the corresponding gate stack from the drain feature 714 (i.e., leaving the other side of the corresponding gate stack in direct contact with the source feature 716), and the internal spacer 726 only isolates the corresponding gate stack from the source feature 716 (i.e., leaving the other side of the corresponding gate stack in direct contact with the drain feature 714).

[0071] In one aspect of the present invention, a memory device is disclosed. The memory device includes a plurality of first nanostructures stacked on each other; a plurality of first gate stacks, wherein two adjacent ones of the first gate stacks surround a corresponding one of the plurality of first nanostructures; a plurality of second nanostructures stacked on each other; a plurality of second gate stacks, wherein two adjacent ones of the second gate stacks surround a corresponding one of the plurality of second nanostructures; a first drain / source component electrically coupled to a first end of the first nanostructure; a second drain / source component electrically coupled to both a second end of the first nanostructure and a first end of the second nanostructure; and a third drain / source component electrically coupled to a second end of the second nanostructure. At least one of the plurality of first gate stacks directly contacts at least one of the first drain / source component and the second drain / source component.

[0072] In the above memory device, each of the plurality of second gate stacks is electrically isolated from the second drain / source feature by the first dielectric spacer and is electrically isolated from the third drain / source feature by the second dielectric spacer.

[0073] In the above memory device, each of the plurality of first gate stacks includes a first gate metal and a first gate dielectric, and each of the plurality of second gate stacks includes a second gate metal and a second gate dielectric.

[0074] In the above memory device, each of the plurality of first gate stacks directly contacts both of the first drain / source feature and the second drain / source feature.

[0075] In the above memory device, each of the plurality of first gate stacks directly contacts only the first drain / source feature.

[0076] In the above memory device, each of the plurality of first gate stacks directly contacts only the second drain / source feature.

[0077] In the above memory device, a first one of the plurality of first gate stacks directly contacts the first drain / source feature and is electrically isolated from the second drain / source feature, and a second one of the plurality of first gate stacks is electrically isolated from the first drain / source feature and directly contacts the second drain / source feature.

[0078] In the above-mentioned memory device, multiple first gate stacks, first drain / source components and second drain / source components are configured as programming transistors of an anti-fuse memory cell, and multiple first nanostructures are collectively configured as a conduction channel of the programming transistor, and, wherein multiple second gate stacks, second drain / source components and third drain / source components are configured as read transistors of the anti-fuse memory cell, and multiple second nanostructures are collectively configured as a conduction channel of the read transistor.

[0079] In another aspect of the present invention, a memory cell is disclosed. The memory cell includes a first transistor, a second transistor electrically coupled in series to the first transistor. The first transistor includes a plurality of first nanosheets spaced apart from each other along a vertical direction, wherein the plurality of first nanosheets have a first length along a horizontal direction; and a plurality of first all-around gate stacks operably associated with the plurality of first nanosheets, wherein the plurality of first all-around gate stacks have a second length along the horizontal direction, and the second length is equal to or less than the first length. The second transistor includes a plurality of second nanosheets spaced apart from each other vertically, wherein the plurality of second nanosheets have a third length along the horizontal direction; and a plurality of second all-around gate stacks operably associated with the plurality of second nanosheets, wherein the plurality of second all-around gate stacks have a fourth length along the horizontal direction, and the fourth length is equal to or less than the third length.

[0080] In the above memory cell, each of the plurality of first all-around gate stacks includes a first gate metal and a first gate dielectric, and each of the plurality of second all-around gate stacks includes a second gate metal and a second gate dielectric.

[0081] In the above memory cell, it also includes: a first drain / source component arranged on a first side of the plurality of first fully-surround gate stacks; and a second drain / source component arranged on a second side of the plurality of first fully-surround gate stacks, wherein each of the plurality of first fully-surround gate stacks directly contacts both of the first drain / source component and the second drain / source component, so that the second length is substantially equal to the first length.

[0082] In the above-mentioned memory cell, it also includes: a first drain / source component arranged on a first side of a plurality of first fully surrounding gate stacks; and a second drain / source component arranged on a second side of the plurality of first fully surrounding gate stacks, wherein each of the plurality of first fully surrounding gate stacks directly contacts the first drain / source component or the second drain / source component, so that the second length is less than the first length.

[0083] In the above memory cell, it also includes: a first drain / source component arranged between a plurality of first all-around gate stacks and a plurality of second all-around gate stacks; a second drain / source component arranged to be opposite to the plurality of first all-around gate stacks from the first drain / source component; and a third drain / source component arranged to be opposite to the plurality of second all-around gate stacks from the first drain / source component. Each of the plurality of first all-around gate stacks is electrically isolated from both the first drain / source component and the second drain / source component by a plurality of first spacers, and each of the plurality of second all-around gate stacks is electrically isolated from both the second drain / source component and the third drain / source component by a plurality of second spacers.

[0084] In the above memory cell, a dielectric constant of the plurality of first spacers is greater than a dielectric constant of the plurality of second spacers.

[0085] In the above memory cell, a dielectric constant of the plurality of first spacers is smaller than a dielectric constant of the plurality of second spacers.

[0086] In the above memory cell, each of the plurality of first spacers has a first width along the horizontal direction, and each of the plurality of second spacers has a second width along the horizontal direction, and the sum of the first widths of the first spacers is smaller than the sum of the second widths of the second spacers.

[0087] In the above memory cell, each of the plurality of first spacers has a first height along the vertical direction, and each of the plurality of second spacers has a second height along the vertical direction, and the sum of the first heights of the first spacers is smaller than the sum of the second heights of the second spacers.

[0088] In the above-mentioned memory cell, each of the plurality of first spacers includes a first dielectric material selected from at least one of silicon oxide, silicon nitride, silicon oxynitride, SiBCN, SiOCN and SiOC, and each of the plurality of second spacers includes a second dielectric material selected from at least one of silicon oxide, silicon nitride, silicon oxynitride, SiBCN, SiOCN and SiOC.

[0089] In another aspect of the present invention, a method for manufacturing a memory device is disclosed. The method includes forming a first stack above a substrate. The first stack includes a first nanosheet, a second nanosheet above the first nanosheet, and a third nanosheet above the second nanosheet. The method includes forming a second stack above the substrate. The second stack includes a fourth nanosheet, a fifth nanosheet above the fourth nanosheet, and a sixth nanosheet above the fifth nanosheet. The method includes removing respective ends of the fourth nanosheet and the sixth nanosheet while covering the first stack. The method includes forming a plurality of spacers at respective etched ends of the fourth nanosheet and the sixth nanosheet while still covering the first stack.

[0090] In the above method, it also includes: replacing the first nanosheet and the third nanosheet with a first gate stack to surround the second nanosheet, wherein the first gate stack directly contacts the respective source / drain components; and replacing the fourth nanosheet and the sixth nanosheet with a second gate stack to surround the fifth nanosheet, wherein the second gate stack is electrically isolated from the respective source / drain components by a plurality of spacers.

[0091] The features of several embodiments are summarized above so that those skilled in the art can better understand aspects of the present invention. Those skilled in the art should understand that they can easily use the present invention as a basis to design or modify other processes and structures for implementing the same purpose and / or achieving the same advantages as the embodiments introduced herein. Those of ordinary skill in the art should also be aware that such equivalent constructions do not deviate from the spirit and scope of the present invention, and that various changes, substitutions and modifications may be made without departing from the spirit and scope of the present invention.

Claims

1. A storage device, include: A plurality of first nanostructures are stacked on top of each other; A plurality of first gate stacks, wherein two adjacent ones of the first gate stacks surround a corresponding one of the plurality of first nanostructures; A plurality of second nanostructures are stacked on top of each other; a plurality of second gate stacks, wherein two adjacent ones of the second gate stacks surround a corresponding one of the plurality of second nanostructures; a first drain / source feature electrically coupled to a first end of the first nanostructure; a second drain / source feature electrically coupled to both the second end of the first nanostructure and the first end of the second nanostructure; as well as a third drain / source component electrically coupled to the second end of the second nanostructure, wherein at least one of the plurality of first gate stacks directly contacts at least one of the first drain / source feature and the second drain / source feature, wherein a first one of the plurality of first gate stacks directly contacts the first drain / source feature and is electrically isolated from the second drain / source feature, and a second one of the plurality of first gate stacks is electrically isolated from the first drain / source feature and directly contacts the second drain / source feature.

2. The memory device according to claim 1, in, Each of the plurality of second gate stacks is electrically isolated from the second drain / source feature by a first dielectric spacer and is electrically isolated from the third drain / source feature by a second dielectric spacer.

3. The memory device according to claim 1, in, Each of the plurality of first gate stacks includes a first gate metal and a first gate dielectric, and each of the plurality of second gate stacks includes a second gate metal and a second gate dielectric.

4. The memory device according to claim 1, in, An entire sidewall of a first one of the plurality of first gate stacks directly contacts the first drain / source feature.

5. The memory device according to claim 2, in, The first dielectric spacer and the second dielectric spacer include silicon nitride.

6. The memory device according to claim 2, in, The first dielectric spacer is in direct contact with the second drain / source feature.

7. The memory device according to claim 2, in, The second dielectric spacer is in direct contact with the third drain / source feature.

8. The memory device according to claim 1, in, The plurality of first gate stacks, the first drain / source component, and the second drain / source component are configured as a programming transistor of an anti-fuse memory cell, and the plurality of first nanostructures are collectively configured as a conduction channel of the programming transistor, and wherein the plurality of second gate stacks, the second drain / source component, and the third drain / source component are configured as a read transistor of the anti-fuse memory cell, and the plurality of second nanostructures are collectively configured as a conduction channel of the read transistor.

9. A memory cell, include: a first transistor; as well as a second transistor electrically coupled in series to the first transistor, Wherein, the first transistor comprises: a plurality of first nanosheets spaced apart from each other along a vertical direction, the plurality of first nanosheets having a first length along a horizontal direction; and a plurality of first all-around gate stacks operably associated with the plurality of first nanosheets, the plurality of first all-around gate stacks having a second length along the horizontal direction, And wherein, the second transistor comprises: a plurality of second nanosheets vertically spaced apart from each other, the plurality of second nanosheets having a third length along the horizontal direction; a plurality of second all-around gate stacks operably associated with the plurality of second nanosheets, the plurality of second all-around gate stacks having a fourth length along the horizontal direction, the fourth length being less than the third length; a first drain / source feature disposed on a first side of the plurality of first all-around gate stacks; and a second drain / source feature disposed on a second side of the plurality of first all-around gate stacks, wherein a first one of the plurality of first fully-surrounding gate stacks is in direct contact with the first drain / source component and is electrically isolated from the second drain / source component, and a second one of the plurality of first fully-surrounding gate stacks is electrically isolated from the first drain / source component and is in direct contact with the second drain / source component, such that the second length is less than the first length.

10. The memory cell according to claim 9, in, Each of the plurality of first all-around gate stacks includes a first gate metal and a first gate dielectric, and each of the plurality of second all-around gate stacks includes a second gate metal and a second gate dielectric.

11. The memory cell according to claim 9, in, An entire sidewall of a first one of the plurality of first all-around gate stacks directly contacts the first drain / source feature.

12. The memory cell according to claim 9, in, An entire sidewall of a second one of the plurality of first all-around gate stacks directly contacts the second drain / source feature.

13. The memory cell according to claim 9, further comprising: include: a third drain / source feature disposed opposite the first drain / source feature relative to the plurality of second all-around gate stacks, wherein a first one of the plurality of first all-around gate stacks is electrically isolated from the second drain / source feature by a first spacer, and each of the plurality of second all-around gate stacks is electrically isolated from both the second drain / source feature and the third drain / source feature by a plurality of second spacers.

14. The memory cell according to claim 13, in, A dielectric constant of the first spacer is greater than a dielectric constant of the plurality of second spacers.

15. The memory cell according to claim 13, in, A dielectric constant of the first spacer is smaller than a dielectric constant of the plurality of second spacers.

16. The memory cell according to claim 9, in, Each of the first transistor and the second transistor includes an n-type metal oxide semiconductor field effect transistor.

17. The memory cell according to claim 9, in, Each of the first transistor and the second transistor includes a p-type metal oxide semiconductor field effect transistor.

18. The memory cell according to claim 13, in, The first spacer includes a first dielectric material selected from at least one of silicon oxide, silicon nitride, silicon oxynitride, SiBCN, SiOCN and SiOC, and each of the plurality of second spacers includes a second dielectric material selected from at least one of silicon oxide, silicon nitride, silicon oxynitride, SiBCN, SiOCN and SiOC.

19. A method of manufacturing a memory device, include: forming a first stack over a substrate, the first stack comprising a first nanosheet, a second nanosheet over the first nanosheet, and a third nanosheet over the second nanosheet; forming a second stack over the substrate, the second stack comprising a fourth nanosheet, a fifth nanosheet over the fourth nanosheet, and a sixth nanosheet over the fifth nanosheet; while covering the first stack, removing respective ends of the fourth nanosheet and the sixth nanosheet; as well as forming a plurality of spacers at respective etched ends of the fourth nanosheet and the sixth nanosheet while still covering the first stack, The method further includes replacing the first nanosheet and the third nanosheet with a plurality of first gate stacks to surround the second nanosheet, wherein a first source / drain component and a second source / drain component are respectively provided at opposite ends of the plurality of first gate stacks, and wherein a first one of the plurality of first gate stacks directly contacts the source / drain component and is electrically isolated from the second source / drain component, and a second one of the plurality of first gate stacks is electrically isolated from the first source / drain component and directly contacts the second source / drain component.

20. The method according to claim 19, further comprising: include: The fourth nanosheet and the sixth nanosheet are replaced with a second gate stack to surround the fifth nanosheet, wherein the second gate stack is electrically isolated from respective source / drain features by the plurality of spacers.