Integrated circuit structure and manufacturing method thereof

By adopting shallow trench isolation and backside dielectric layer design in GAA devices, the current leakage and packaging density problems in SRAM devices are solved, and an integrated circuit structure with high efficiency isolation and density is realized, meeting the complex functional requirements.

CN119031693APending Publication Date: 2024-11-26TAIWAN SEMICONDUCTOR MANUFACTURING CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411066289.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2024-01-12
Filing Date
2024-08-05
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

Existing full-ring gate (GAA) devices have current leakage problems in static random access memory (SRAM) devices, especially under high circuit speed requirements, and the device package density is low, making it difficult to meet the complex function requirements.

Method used

Using an integrated circuit design including shallow trench isolation (STI) structure and backside dielectric layer, the formation of multiple vertical gates and active regions on the semiconductor substrate, combining n-type and p-type doping wells can achieve the Collective isolation of SRAM cells reduces current leakage and optimizes device layout to ensure efficient isolation and density.

Benefits of technology

Significantly reduces current leakage, improves device performance and package density, enabling a smaller SRAM circuit layout while maintaining high circuit speeds and low power consumption.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119031693A_ABST
    Figure CN119031693A_ABST
Patent Text Reader

Abstract

The present disclosure provides an integrated circuit (IC) structure and a method of manufacturing the same, the IC structure including: a semiconductor substrate having a static random access memory (SRAM) region, an input / output and peripheral (IOP) region, and an edge region spanning between the SRAM region and the IOP region; a shallow trench isolation (STI) structure formed on the semiconductor substrate and defining an active region; the SRAM unit is formed in the SRAM region; and a backside dielectric layer disposed on a backside of the semiconductor substrate and landing on a bottom surface of the STI structure, in which the active region is longitudinally oriented along a first direction, a plurality of gates formed on the semiconductor substrate and longitudinally oriented along a second direction perpendicular to the first direction, the gates uniformly distributed at a pitch P along the first direction, the SRAM cell spans a first dimension Ds along a first direction, the edge region spans a second dimension De along the first direction, and a ratio De / Ds is equal to or less than 2.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] Embodiments of the present application relate to integrated circuit structures and methods for manufacturing the same. Background Art

[0002] The electronics industry continues to demand ever-increasing amounts of smaller, faster electronic devices that are capable of supporting more and more complex and sophisticated functions simultaneously. To meet these demands, there is a continuing trend in the integrated circuit (IC) industry to manufacture low-cost, high-performance, and low-power ICs. To date, these goals have been achieved in large part by scaling down IC size (e.g., minimum IC component size), thereby increasing production efficiency and reducing associated costs. However, this scaling down has also increased the complexity of the IC manufacturing process. Therefore, achieving continued advancements in IC devices and their performance requires similar advancements in IC manufacturing processes and technologies.

[0003] Recently, multi-gate devices have been introduced to improve gate control. Multi-gate devices have been observed to increase gate-channel coupling, reduce off-state current, and / or reduce short channel effects (SCE). One such multi-gate device is a gate-all-around (GAA) device, whose gate structure can extend partially or completely around the channel region to provide access to the channel region on at least two sides. GAA devices enable aggressive scaling of IC technology, maintain gate control and mitigate SCE, while seamlessly integrating with traditional IC manufacturing processes. As GAA devices continue to expand, challenges have emerged in some areas, such as reduced device packaging density and current leakage issues, especially in static random access memory (SRAM) devices. Therefore, although existing GAA SRAM devices and methods of manufacturing such devices are generally adequate for their intended purposes, they are not completely satisfactory in all aspects. Summary of the invention

[0004] According to one aspect of an embodiment of the present application, an integrated circuit (IC) structure is provided, comprising: a semiconductor substrate having a static random access memory (SRAM) area, an input / output and peripheral (IOP) area, and an edge area spanning between the SRAM area and the IOP area; a shallow trench isolation (STI) structure formed on the semiconductor substrate and defining an active area; an SRAM cell formed in the SRAM area; and a backside dielectric layer disposed on the back side of the semiconductor substrate and landing on a bottom surface of the STI structure, wherein the active area is longitudinally oriented along a first direction, a plurality of gates are formed on the semiconductor substrate and longitudinally oriented along a second direction perpendicular to the first direction, the gates are uniformly distributed along the first direction at a pitch P, the SRAM cell spans a first dimension Ds along the first direction, the edge area spans a second dimension De along the first direction, and the ratio De / Ds is equal to or less than 2.

[0005] According to another aspect of an embodiment of the present application, an integrated circuit structure is provided, comprising: a semiconductor substrate having a static random access memory (SRAM) region, an input / output and peripheral (IOP) region, and an edge region spanning between the SRAM region and the IOP region; an n-type doped well extending continuously along a first direction through the SRAM region and the edge region; a p-type doped well extending continuously along the first direction through the SRAM region and the edge region; a shallow trench isolation (STI) structure formed on the semiconductor substrate and defining a first active region, a second active region, and a third active region longitudinally oriented along the first direction; a plurality of gates longitudinally oriented along a second direction perpendicular to the first direction, wherein the gates are uniformly distributed in the SRAM region and the edge region with a periodic pitch P; and a backside dielectric layer contacting a bottom surface of the semiconductor substrate and a bottom surface of the STI structure, wherein the SRAM region comprises SRAM cells, each of the SRAM cells spanning a first dimension Ds along the first direction, the edge region spanning a second dimension De along the first direction, and a ratio De / Ds is equal to or less than 2.

[0006] According to another aspect of an embodiment of the present application, a method for manufacturing an integrated circuit (IC) structure is provided, comprising: receiving a semiconductor substrate having a front side and a back side, wherein the semiconductor substrate comprises a static random access memory (SRAM) region, an input / output and peripheral (IOP) region, and an edge region spanning between the SRAM region and the IOP region; forming an SRAM structure having SRAM cells in the SRAM region on the front side of the semiconductor substrate; forming an interconnect structure above the SRAM structure; thinning the semiconductor substrate from the back side so that a shallow trench isolation (STI) structure is exposed from the back side; and forming a backside dielectric layer directly on a bottom surface of the STI structure and a bottom surface of the semiconductor substrate. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Various aspects of the present disclosure can be best understood from the following detailed description when read in conjunction with the accompanying drawings. It should be emphasized that, in accordance with standard practice in the industry, the various components are not drawn to scale and are only used for illustrative purposes. In fact, the size of the various components may be arbitrarily increased or reduced for clarity of discussion.

[0008] Figure 1 is a top view of an integrated circuit structure having a static random access memory (SRAM) structure according to various aspects of the present disclosure.

[0009] Figure 2 is a schematic diagram of an SRAM structure according to various aspects of the present disclosure.

[0010] Figure 3 is a top view of a SRAM structure according to various aspects of the present disclosure.

[0011] Figure 4 is a partial top view of an integrated circuit structure according to various aspects of the present disclosure.

[0012] Figure 5 is a partial top view of an integrated circuit structure according to various aspects of the present disclosure.

[0013] Fig. 6A and Figure 6B is a partial top view of an integrated circuit structure according to various embodiments of the present disclosure.

[0014] Fig. 7A and Figure 7B is a partial cross-sectional view of an integrated circuit structure according to various aspects of the present disclosure.

[0015] Fig. 8A is a top view of an integrated circuit structure according to various aspects of the present disclosure.

[0016] Figure 8B , Figure 8C and Fig.8D According to various aspects of the present disclosure Fig. 8A A cross-sectional view of an integrated circuit structure.

[0017] Fig. 9A is a top view of an integrated circuit structure according to various aspects of the present disclosure.

[0018] Fig. 9B According to various aspects of the present disclosure Fig. 9A A cross-sectional view of an integrated circuit structure.

[0019] Fig. 10A , Fig. 10B and Fig. 10C is a flow chart of a method for fabricating an integrated circuit structure according to various aspects of the present disclosure.

[0020] Fig.11A , Fig. 12A , Fig.13A , Fig.14A , Fig.15A and Fig.16A is a partial perspective view of a portion or the entirety of an integrated circuit structure at various stages of fabrication in accordance with various aspects of the present disclosure.

[0021] Fig. 11B , Fig. 12B , Fig. 13B , Fig. 14B , Fig. 15B and Fig. 16B are partial cross-sectional views of a portion or all of an integrated circuit structure at various stages of fabrication in accordance with various aspects of the present disclosure.

[0022] Fig.17 , Fig.18 , Fig.19 , Fig. 20 , Fig.21 , Fig. 22 , Fig.23 and Fig.24 is a partial top view of a portion or the entirety of an integrated circuit structure according to various aspects of the present disclosure. DETAILED DESCRIPTION

[0023] The present disclosure relates generally to integrated circuit devices and, more particularly, to multi-gate devices, such as gate-all-around (GAA) devices.

[0024] The following disclosure provides many different embodiments or examples for realizing different features. Reference numbers and / or letters may be repeated in various examples described herein. This repetition is for the purpose of simplicity and clarity, and does not itself specify the relationship between various disclosed embodiments and / or configurations. In addition, specific examples of components and arrangements are described below to simplify the present disclosure. Of course, these are merely examples and are not restrictive. For example, in the following description, forming a first component on or on a second component may include an embodiment in which the first component and the second component are formed in direct contact, and may also include an embodiment in which an additional component may be formed between the first component and the second component so that the first component is not in direct contact with the first component. In addition, in the present disclosure, forming a special component on another component, connecting to another component and / or coupling to another component may include an embodiment in which the component is formed in a direct contact manner, and may also include an embodiment in which an additional component may be formed as an insertion special component so that the component may not be in direct contact.

[0025] In addition, for ease of description, spatially relative terms such as "lower", "upper", "horizontal", "vertical", "above", "below", "below", "above", "downward", "top", "bottom", etc. and their derivatives (e.g., "horizontally", "downwardly", "upwardly", etc.) are used herein to describe the relationship of one element or component to another element or component. Spatially relative terms are intended to encompass different orientations depicted with a device (or system or apparatus) that includes the element or component, including orientations associated with use or operation of the device. The device may be otherwise oriented (rotated 90 degrees or at other orientations), and the spatially relative descriptors used herein may be interpreted accordingly.

[0026] In addition, when a number or a range of numbers is described using "about," "approximately," etc., the term is intended to encompass numbers within a reasonable range understood by those skilled in the art to include the described number, such as within + / -10% of the described number, or other values. For example, the term "about 5 nm" may encompass a size range of 4.5 nm to 5.5 nm.

[0027] The disclosed device structure and its manufacturing method relate to field effect transistors (FETs), especially FETs with multiple channels stacked vertically, such as gate-all-around (GAA) FET structures. The disclosed device structure includes various structural features and manufacturing steps to provide collective isolation and prevent current leakage of the device structure. The disclosed structure relates to an integrated circuit structure having static random access memory (SRAM) cells configured in an array. Specifically, the present disclosure provides an SRAM circuit with a GAA structure with enhanced isolation and a corresponding SRAM circuit layout with reduced size.

[0028] In existing SRAM structures, current leakage is a problem, especially when high circuit speed is required. The SRAM cell includes two cross-coupled inverters for data storage and two transmission gates for input / output. In some embodiments, the SRAM cell includes six or more FETs, such as 6 FETs or 8 FETs. In a further implementation of the embodiment, the SRAM cell includes an N-type FET (nFET) and a P-type FET (pFET). In some cases, current leakage is overcome by doping wells and well pickups to bias the doped wells to achieve junction isolation. For example, an SRAM circuit structure includes a P-well and a deep N-well (DNW) located below the P-well, as well as a P-well pickup for biasing the P-well and an N-well pickup for biasing the deep N-well. This structure can reduce current leakage, but increases the device area for additional well pickup areas and reduces circuit packaging density.

[0029] Figure 1 2 is a top view of an integrated circuit (IC) structure 20, which has various circuit areas, including an SRAM area 22, an input / output and peripheral (IOP) area 26, and an edge area (or edge cell) 24 spanning between the SRAM area 22 and the IOP area 26. The SRAM area, the edge area 24, and the IOP area 26 are configured along the X direction and extend in the same size along the Y direction. The SRAM area 22 includes a plurality of SRAM cells 28 configured in an array. The IOP area 26 includes various input, output, and peripheral devices formed therein. The IOP device is designed differently for input / output (I / O) and peripheral devices. For example, due to high power surges, electrostatic discharge (ESD), and other I / O environmental challenges, the I / O field effect transistor includes a gate dielectric layer thicker than other FETs in the SRAM cell. The edge area 24 is a region between the SRAM area 22 and the IOP area 26 to provide a transition area and a buffer zone between the SRAM area 22 and the IOP area 26. In previous structures, the edge region occupied a larger circuit area due to the pick-up region for junction isolation of the substrate. The disclosed structure includes various dielectric isolation components to provide a collective isolation function, and the edge region 24 is significantly reduced.

[0030] In some embodiments of the present disclosure, a method of forming a device structure includes, after forming a FET with a fin structure or a GAA structure, an operation of thinning the substrate from the back side to reach the bottom surface of a shallow trench isolation (STI) structure, so that the semiconductor substrate is separated into a plurality of semiconductor islands isolated from each other by the STI structure, thereby achieving isolation of these semiconductor islands and reducing leakage current. Therefore, those well pickup regions used for junction isolation are eliminated, and the circuit area is reduced by more than 80%. Therefore, the disclosed device structure is also referred to as a tapless device structure. The edge region 24 spans a dimension De along the X direction, and the SRAM cell 28 spans a dimension Ds along the X direction. In some embodiments, the ratio De / Ds is equal to or less than 2.

[0031] According to some embodiments, the SRAM cell is Figure 2 and Figure 3 Further shown in . Figure 2 is a schematic diagram of an SRAM cell 28 constructed in one embodiment according to various aspects of the present disclosure. Figure 3 is a top view of an SRAM cell constructed in one embodiment according to various aspects of the present disclosure.

[0032] refer to Figure 2 , the SRAM cell 28 is formed on the fin active area and includes various FETs, such as fin FETs (FinFETs), GAA FETs, or other suitable FETs. The SRAM cell 28 includes a first inverter and a second inverter that are cross-coupled to form a data storage unit. The first inverter includes a first pull-up device (PU1) formed by a p-type field effect transistor (pFET). The first inverter also includes a first pull-down device (PD1) formed by one or more n-type field effect transistors (nFETs). The second inverter includes a second pull-up device (PU2) formed by a pFET. The second inverter also includes a second pull-down device (PD2) formed by one or more nFETs. The SRAM cell 28 also includes various transmission gate devices (PG1 and PD2) connected to the cross-coupled inverters and configured to form two ports for reading and writing.

[0033] According to some embodiments, Figure 3The SRAM cell 28 is further described in detail in the layout. The included SRAM cell 28 is formed on a semiconductor substrate. The semiconductor substrate includes silicon. Alternatively, the substrate includes germanium, silicon germanium or other suitable semiconductor materials. The semiconductor substrate may include other suitable components and structures. In one embodiment, the semiconductor substrate is isolated using a semiconductor material layer formed on an insulating layer on a support wafer. This technology and structure is called semiconductor on isolation (SOI). The SOI structure can be formed by different techniques, including separation by implantation of oxygen (SIMOX), bonding and etch back (BESOI), and zone melting and recrystallization (ZMR).

[0034] The SRAM cell 28 is formed in a unit cell region 112 of a semiconductor substrate. The unit cell region 112 is defined by a unit cell boundary 114. In one embodiment, the unit cell region 112 is defined as a rectangular shape that spans a first dimension Ds along the X direction and a second dimension Ds2 along the Y direction perpendicular to the X direction. Ds2 is longer than Ds. In one embodiment, the ratio of the second dimension to the first dimension is greater than 3. The SRAM cell 28 includes an N-well 120 disposed in a middle portion of the cell; and a P-well 122 disposed on both sides of the N-well 120. In one embodiment, the N-well 120 and the P-well 122 extend to multiple cells outside the unit cell boundary. For example, the N-well 120 and the P-well 122 extend to 4 or more cells in the second direction.

[0035] Various active areas are defined in the substrate by isolation components and isolated from each other by isolation components. The isolation components are formed in the semiconductor substrate using appropriate technology. In one embodiment, the isolation components are formed by shallow trench isolation (STI) technology. In another embodiment, the formation of the STI components includes etching trenches in the substrate and filling the trenches with one or more insulator materials (e.g., silicon oxide, silicon nitride, or silicon oxynitride). The filled trenches may have a multilayer structure, such as having a thermal oxide liner layer when silicon nitride fills the trenches. Active areas are defined in the semiconductor substrate when the isolation components are formed.

[0036] The SRAM cell 28 forms a planar FET using a planar active area and forms a FinFET using a fin active area; a GAA structure forms a GAAFET or other structures to form a corresponding FET. For example, the fin active area is formed by a suitable technique and can be formed in the process of forming an STI component and a fin active area. In one embodiment, the fin active area is formed by a process including etching a semiconductor substrate to form a trench, partially filling the trench to form a shallow trench isolation (STI) structure around each fin active area. In a further implementation of this embodiment, an epitaxial semiconductor layer is selectively formed on the fin active area. In another embodiment, a fin active area is formed by a process including depositing a dielectric material layer on a semiconductor substrate, etching the dielectric material layer to form an opening thereof, and selectively epitaxially growing a semiconductor material (e.g., silicon) on a semiconductor substrate within the opening to form a fin active area and an isolation component. In yet another embodiment, various FETs may include strain components for enhancing mobility and device performance. For example, a pFET includes silicon germanium epitaxially grown on a silicon substrate. An n-type nFET includes silicon carbide epitaxially grown on a silicon substrate. In the GAAFET, the semiconductor substrate further includes a semiconductor stack having two different semiconductor materials stacked alternately. In some embodiments, the semiconductor stack includes a first semiconductor layer of silicon and a second semiconductor layer of silicon germanium stacked alternately. During the formation of the fin active fin and STI structure, the semiconductor stack is patterned into a well.

[0037] In one embodiment, the SRAM cell 28 includes a first active region 126 and a second active region 128 formed in the P-well 122. The SRAM cell 100 also includes a third active region 130 and a fourth active region 132 formed in the N-well 120. All active regions 126 to 132 are arranged in parallel and oriented along the X direction. The first to fourth active regions or a subset thereof may extend to a plurality of cells, for example, 4 or more cells, in the second direction.

[0038] In some embodiments, each active area includes one or more fin active components configured to form various FinFETs. In some further embodiments, each active area includes multiple channel layers stacked vertically to form one or more GAAFETs. Each of the active areas 126 and 128 in the P-well 122 includes an active area having a first size Wn along the Y direction; and each of the active areas 130 to 132 in the N-well 120 includes a second width Wp along the Y direction. In some embodiments, Wp is different from Wn. In a further implementation of the embodiment, Wn is greater than Wp, or the ratio Wn / Wp is greater than 1, such as greater than 2 or 3. This layout is also called a high current layout because nFETs are designed to have high currents to increase the reading and writing of SRAM cells. In active areas 126 and 128, nFETs are formed, such as two pull-down devices (PD), two transmission gate devices (PG). In active areas 130 and 132, pFETs are formed, such as two pull-up devices (PU).

[0039] The gate structure is formed over various active regions and joins the portions of the corresponding active regions as channel configurations to form nFETs and pFETs. The gate structure includes a gate dielectric layer and a gate electrode disposed over the gate dielectric layer and a gate spacer disposed on the sidewalls of the gate electrode. The gate dielectric layer includes silicon oxide, a high-k dielectric material including metal oxides or metal nitrides (e.g., hafnium oxide and hafnium nitride), other suitable dielectric materials, or a combination thereof. The gate electrode includes a metal (such as aluminum, copper, and tungsten), an alloy, other suitable conductive materials, or a combination thereof. In some embodiments, the gate electrode also includes a metal (or metal alloy) layer having an appropriate work function to tune the threshold voltage of the corresponding FET. The metal layers for nFETs and pFETs are referred to as n-type work function metal layers and p-type work function metal layers, respectively. The compositions of the n-type and p-type work function metal layers of nFETs and pFETs are different. In some embodiments, the n-type work function metal layer includes tantalum (Ta), titanium aluminum (TiAl), titanium aluminum carbide (TiAlC), titanium aluminum oxide (TiAlO), titanium aluminum nitride (TiAlN), or a combination thereof. In some embodiments, the p-type work function metal layer includes titanium nitride (TiN), tantalum nitride (TaN), tungsten nitride (WN), or a combination thereof. The gate dielectric layer and the gate electrode are collectively referred to as a gate stack. The gate structures are arranged in parallel and are longitudinally oriented along the Y direction. The SRAM cell 28 includes gate structures 136, 140, 142, and 144.

[0040] According to some embodiments, gate structures 136, 140, 142, and 144 are configured with active regions 126, 128, 130, and 132 to form nFETs and pFETs for PD devices, PU devices, and PG devices integrated into SRAM cells, such as Figure 3As shown. Specifically, the gate structure 136 extends from the P-well 122 to the N-well 120; the gate structure 142 extends from the P-well 122 to the N-well 120; the gate structure 140 is disposed in the P-well 122, and the gate structure 144 is disposed in the P-well 122. The active area 128 and the gate structure 144 form PG1; the active area 128 and the gate structure 142 form PD1; the active area 130 and the gate structure 142 form PU1; the active area 130 and the gate structure 136 form PU2; the active area 126 and the gate structure 136 form PD2; and the active area 126 and the gate structure 140 form PG2. The SRAM cell 28 also includes other components, such as source / drain components and interconnect structures having contacts, vias, and metal lines distributed in multiple metal layers. For simplicity, Figure 3 It is not shown in the figure, but will be described with other figures.

[0041] refer to Figure 4 The edge region 24 is further described. Figure 4 The edge region 24 (specifically Figure 1 The structure and size of the edge region 24 are designed to eliminate or reduce interference between the SRAM region 22 and the IOP region 26 to provide effective and efficient isolation, while the edge region 24 occupies a minimum circuit area.

[0042] The edge region 24 includes an N-well 42 and a P-well 44 that share a common longitudinal side. The N-well 42 and the P-well 44 are oriented longitudinally along the X direction. In some embodiments, the N-well 42 and the P-well 44 extend continuously from the SRAM region 22 to the edge region 24, which will be further described below. Figure 5 shown.

[0043] The edge region 24 also includes a first active region 36 formed in the N-well 42 and a second active region 38 formed in the P-well 44. The first active region 36 and the second active region 38 are longitudinally oriented along the X direction. In some embodiments, at least one of the first active region 36 and the second active region 38 extends continuously from the SRAM region 22 to the edge region 24.

[0044] The active area 36 in the N well 42 spans a first width Wp along the Y direction; and the active area 38 in the P well 44 spans a second width Wn along the Y direction. In some embodiments, Wp is different from Wn. In further implementations of the embodiments, Wn is greater than Wp, or the ratio Wn / Wp is greater than 1, such as greater than 2 or 3.

[0045] The edge region 24 also includes a gate structure 40 disposed in parallel and oriented longitudinally along the Y direction. In the disclosed embodiment, the edge region 24 includes a gate structure 40 disposed above the active region or above the edge of the active region, such as Figure 4 The gate structure 40 is similar to the gate structure 40 in terms of formation and composition. Figure 3 The gate structures 40 are similar to those in the embodiment of the present invention. For simplicity, similar descriptions are not repeated. The gate structures 40 are uniformly distributed along the X direction and have a periodic size Pg or CPP. Pg is defined as the distance measured from center to center or from one side to the same side from one gate structure to an adjacent gate structure. In some embodiments, Pg ranges between 40nm and 50nm.

[0046] Figure 1 The portion 32 of the IC structure 20 is Figure 5 Further shown in . Figure 5 2 is a top view of a portion of an IC structure 20 constructed according to various aspects of the present disclosure. Specifically, a portion 32 of the IC structure 20 includes an SRAM cell 28 and an edge region 24, which will be referred to as Figure 5 Describe together.

[0047] The IC structure 20 includes an N-well 42 and a P-well 44 oriented longitudinally along the X direction. The P-well 44 includes two portions, wherein the N-well 42 is interposed along the Y direction and spans between the two portions of the P-well 44. In the disclosed embodiment, the N-well 42 and the P-well 44 extend continuously along the X direction from the edge region 24 to the SRAM region 22 or, for example, extend through the SRAM region 22 of a plurality of SRAM cells in an array and across the SRAM cell 28.

[0048] The IC structure 20 also includes a first active region 36 disposed in the N-well 42 to form various pFETs for the PU device of the SRAM cell 28, and a second active region 38 disposed in the P-well 44 to form various nFETs for the PD device and the PG device of the SRAM cell 28. The first active region 36 and the second active region 38 are longitudinally oriented along the X direction. In the disclosed embodiment, the second active region 38 extends continuously from the edge region 24 to the SRAM region 22, and further extends through the SRAM cell 28 along the X direction. At least one of the first active regions 36 extends continuously from the edge region 24 to the SRAM region 22. Specifically, one first active region 36 disposed in the SRAM cell 28 and another first active region 36 disposed in the edge region 24 are spaced apart from each other and aligned along the X direction. Another first active region 36 extends continuously from the edge region 24 to the SRAM cell 28.

[0049] In addition, the first active region 36 in the N-well 42 spans a first width Wp along the Y direction; and the second active region 38 in the P-well 44 spans a second width Wn along the Y direction. In the disclosed embodiment, Wp is different from Wn. In a further implementation of the embodiment, Wn is greater than Wp, or the ratio Wn / Wp is greater than 1, such as greater than 2 or 3, thereby increasing the current to the PG device and the PD device, increasing the speed of the SRAM cell 28 and enhancing its performance.

[0050] IC structure 20 also includes gate structures 40 arranged in parallel and oriented longitudinally along the Y direction. Gate structures 40 are disposed above the active area and extend to STI structures 58. Note that the active area is surrounded by the STI structures, so that in a top view of IC structure 20, the STI structures are defined in any other area not occupied by the active area.

[0051] The gate structures 40 are uniformly distributed with a periodic size Pg along the X direction. Pg is defined as the size measured from the same position (e.g., from center to center or from side to the same side) from one gate structure to an adjacent gate structure. As described above, the edge region 24 spans a size De along the X direction, and the SRAM cell 28 spans a size Ds. In the present embodiment, De=4Pg and Ds=2Pg, or the ratio De / Ds=2. In some other embodiments described below, the ratio De / Ds is less than 2, such as 1.5 or 1. In some embodiments, Pg ranges between 40nm and 50nm; in the case of De=4Pg, De ranges between 160nm and 200nm. In some embodiments, Pg ranges between 40nm and 50nm; in the case of De=2Pg, De ranges between 80nm and 100nm. According to some embodiments, the spacing Da1 between the end of the active region and the boundary line of the edge region 24 is equal to 1.5*Pg. According to some embodiments, a spacing Da2 between an end of the active region and a boundary line of the SRAM region 22 is equal to 0.5*Pg. According to some embodiments, a minimum length of the active region is equal to 2*Pg.

[0052] refer to Fig. 6A and Figure 6B The IC structure 20 is further described. Fig. 6A and Figure 6B is a top view of a portion of an IC structure 20 constructed according to various embodiments of the present disclosure. Figure 5 Likewise, only a portion 32 of IC structure 20 is shown, portion 32 including SRAM cell 28 and edge region 24 . Fig. 6A Similar to Figure 5, but also includes a through-hole component (or simply through-hole) 48 formed on the substrate. The through-hole component 48 is electrically connected to various components, such as the source / drain (S / D) component of the FET. For example, the through-hole component 48 is formed on the front side of the substrate and is electrically connected to the S / D component through the contact component. In some embodiments, the through-hole component 48 is formed from the back side of the substrate and is electrically connected to the bottom surface of the corresponding S / D component. The through-hole component 48 formed on the back side of the substrate is also called a back-side via (VB).

[0053] Figure 6B and Fig. 6A Similar, but the widths of active regions 36 and 38 are the same or the ratio Wn / Wp=1.

[0054] refer to Fig. 7A and Figure 7B The IC structure 20 is further described. Fig. 7A and Figure 7B is a cross-sectional view of a portion of an IC structure 20 constructed according to various embodiments of the present disclosure, taken along Fig. 6A or Figure 6B The dotted lines AA' and BB' are intercepted.

[0055] refer to FIG. 7A to FIG. 7B, the IC structure 20 includes a semiconductor substrate 56, such as a silicon substrate, a gallium arsenide substrate or other suitable semiconductor substrate. An isolation structure 58 is formed in the semiconductor substrate 56 and separates the portions of the semiconductor substrate 56 along the Y direction, thereby defining active regions 36 and 38 on the semiconductor substrate 56. Those active regions 36 and 38 are surrounded by the isolation structure 58 and separated from each other by the isolation structure 58. In the present disclosure, the isolation structure 58 is a shallow trench isolation (STI) structure formed by a suitable procedure including patterning, deposition and chemical mechanical polishing (CMP). These active regions 36 and 38 can be squeezed above the STI structure 58 and are accordingly referred to as fin active regions. The IC structure 20 includes various field effect transistors (FETs) formed on the active regions. The field effect transistor includes a gate structure 40, a source / drain (S / D) component 70 inserted by the gate structure 40, and a channel region 78 below the gate structure 40. The gate structure 40 includes a gate dielectric layer and a gate electrode disposed on the gate dielectric layer, and a gate spacer disposed on the sidewall of the gate electrode. In various embodiments, the FET may be a planar FET, a fin FET, a multi-gate FET such as a full-all-around gate (GAA) FET, or other suitable FET structures. The FET is formed on the front side of a semiconductor substrate 56. The method of forming an IC structure includes thinning the semiconductor substrate 56 from the back side, thereby exposing the STI structure 58. The bottom surface of the semiconductor substrate 56 and the bottom surface of the STI structure 58 are coplanar after thinning. Other structures (such as an interconnect structure 65) are formed on the front side of the semiconductor substrate 56. The backside interconnect structure 63 is formed on the back side of the semiconductor substrate 56 after thinning.

[0056] The above disclosed structure provides isolation for various FETs distributed along the Y direction (along the longitudinal direction of the gate). However, various FETs distributed along the X direction (along the longitudinal direction of an active region such as a fin active region) on an active region are not properly isolated from each other by the STI structure 58. The IC structure 20 in the present disclosure also includes multiple components to collectively achieve enhanced isolation of those FETs (particularly along the X direction), as described below.

[0057] Fig. 8A shows a top view of IC structure 20, and Figure 8B (and Fig.8D )and Figure 8C 2 and 3 show cross-sectional views of an IC structure 20 constructed according to some embodiments along dashed lines BB' and CC', respectively. Fig. 8A, the device structure includes an active region 62 oriented along the X direction and a gate structure 40 oriented along the Y direction. The active region 62 may include some active regions (such as 36) formed in the N well and some active regions (such as 38) formed in the P well. Backside vias (or "VB") 64 are formed from the back side of the substrate 56. The backside vias 64 are conductive features and are part of a backside interconnect structure 63 for electrical routing. The backside vias 64 are electrically connected to the FETs, such as from the back side to the S / D features 70. In some embodiments, a subset of the backside vias 64 are replaced by backside dielectric vias for isolation functions, such as Fig. 9A and Fig. 9B Those shown. The backside dielectric via is a dielectric feature and is different from the backside conductive via. The formation of the backside conductive via and the backside dielectric feature will be further described later.

[0058] exist Figure 8B In the present invention, various devices including FETs such as GAAFETs are formed on the front side of the substrate 56. The FET includes a plurality of vertically stacked channel layers 78, source / drain (S / D) features 70, and a gate structure 40 interposed between the S / D features 70 and covering the channel layers 78. The gate structure 40 further extends to wrap around each of the vertically stacked channel layers 78. The gate structure 40 includes a gate dielectric layer, a gate electrode disposed on the gate dielectric layer, and a gate spacer disposed on the sidewall of the gate electrode.

[0059] In addition, the S / D features 70 are formed with an embedded dielectric feature (or dielectric layer) 68, thereby obtaining the corresponding S / D features 70 from the semiconductor substrate 56. The dielectric feature 68 may include any suitable dielectric material, such as silicon nitride, silicon oxide, silicon oxynitride, other suitable dielectrics or combinations thereof. The formation of the S / D features 70 (which are embedded in the dielectric layer) may include etching to recess the S / D region; epitaxially growing a semiconductor material with a lower doping concentration (such as phosphorus doping for N-type FETs or boron doping for P-type FETs); forming the dielectric feature 68; and epitaxially growing a semiconductor material with a higher doping concentration. The semiconductor material may include silicon, silicon germanium, or other suitable semiconductor materials. The formation of the dielectric feature 68 may include depositing a dielectric material and anisotropic etching such as plasma etching to remove portions deposited on the sidewalls of the recess.

[0060] In some alternative embodiments, dielectric feature 68 is formed on the bottom surface of epitaxial S / D feature 70, such as Fig.8D shown. Fig.8D is a cross-sectional view of an IC structure 20 constructed in accordance with some embodiments. Fig.8D and Figure 8B6 is similar, but dielectric feature 68 is formed on the bottom surface of S / D feature 70. For example, the formation of S / D feature 70 (which is embedded in the dielectric layer) may include etching to recess the S / D region; forming dielectric feature 68; and epitaxially growing a semiconductor material with a lower doping concentration (such as phosphorus doping for N-type FETs or boron doping for P-type FETs). The epitaxial growth process may include epitaxially growing a semiconductor material with a lower doping concentration and epitaxially growing a semiconductor material with a higher doping concentration. The formation of dielectric feature 68 may include depositing a dielectric material and anisotropic etching such as plasma etching to remove portions deposited on the sidewalls of the recess.

[0061] A front-side interconnect structure is further formed above the FET. The front-side interconnect structure includes contacts, vias, and metal lines distributed in multiple metal layers. Some components of the front-side interconnect structure (such as contacts 72) are as shown in FIG. Figure 8B As shown. For example, an interlayer dielectric (ILD) layer 74 is formed over the FET by a suitable procedure, such as a procedure including deposition and CMP. The ILD layer 74 may include an etch stop layer and a low-k dielectric material disposed on the etch stop layer. The ILD layer 74 is patterned to form contact holes; one or more metals or other conductive materials are deposited in the contact holes; and a CMP process is applied to remove excess metal and planarize the top surface, thereby forming a contact 72 aligned with, landed on, and electrically connected to the corresponding S / D feature 70.

[0062] The IC structure 20 also includes backside vias 64 and other conductive features of the backside interconnect structure 63, such as metal lines 80. In some embodiments, after forming the FETs (and other devices) and the frontside interconnect structures on the FETs, a carrier substrate can be bonded to the front side. Thereafter, the semiconductor substrate 56 is thinned from the back side so that the STI structures 58 are exposed from the back side. Other processes may additionally be applied to planarize the backside surface, such as etching, deposition, and chemical mechanical polishing (CMP). Thus, the bottom surface of the substrate 56 is coplanar with the bottom surface of the STI structure 58. A backside dielectric layer 82 is deposited on the back side and directly contacts the coplanar bottom surfaces of the semiconductor substrate 56 and the STI structure 58, such as Figure 8B and Figure 8C The backside dielectric layer 82 includes one or more suitable dielectric materials, such as silicon nitride, silicon oxide, or a combination thereof.

[0063] Backside vias 64 are formed in semiconductor substrate 56 and electrically connected to S / D components 70, such as Figure 8BAs shown. Each backside via 64 includes a metal plug (also indicated by the number 64) and a dielectric barrier layer 66 surrounding the sidewalls of the backside via to provide isolation between adjacent semiconductor islands and the metal plug. In some embodiments, the dielectric barrier layer 66 includes silicon nitride, other suitable dielectric materials, or a combination thereof. The metal plug 64 includes one or more metals, such as copper, tungsten, other suitable metals, or a combination thereof. The formation of the backside via 64 includes patterning the backside dielectric layer 82 and the semiconductor substrate 56 to form an opening in which the corresponding S / D component 70 is exposed; depositing the dielectric barrier material; performing a plasma etching process to remove a portion of the dielectric barrier material deposited on the sidewalls of the opening; depositing a metal to fill the opening; and performing a CMP process for planarization. Specifically, the backside dielectric layer 82 and the dielectric barrier layer 66 surround the semiconductor islands of the semiconductor substrate 56, thereby providing enhanced isolation and reducing leakage problems.

[0064] A backside interlayer dielectric (ILD) layer 84 is formed on the backside dielectric layer 82. The backside ILD layer 84 includes one or more dielectric materials such as an etch stop layer and a low-k dielectric material formed by suitable techniques such as chemical vapor deposition (CVD), spin coating, other suitable techniques, or combinations thereof.

[0065] Other conductive features, such as metal lines (or backside metal lines) 80, are formed in the backside ILD layer 84 and are electrically connected to the backside vias 64, such as Figure 8B As shown. The formation of metal line 80 may include any appropriate process, such as a dual damascene process. For example, the backside ILD layer 84 is patterned by photolithography and etching processes to form a trench; one or more metals, such as a barrier layer (e.g., titanium and titanium nitride) and a filling metal are sequentially deposited in the trench; and a CMP process is performed to remove excess deposited metal and planarize the surface.

[0066] exist Figure 8B In the embodiment of the present invention, a first subset of S / D components 70 is associated with backside vias 64, while a second subset of S / D components 70 does not have backside vias. The second subset of S / D components 70 includes dielectric components 68 embedded therein for isolation, while the first subset of S / D components 70 does not have dielectric components 68. The channel length Dc is less than Pg. In some embodiments, the ratio Pg / Dc ranges between 2 and 4.

[0067] As described above, the backside vias 64 may have Fig. 9A and Fig. 9B Some alternative structures described in . Fig. 9A shows a top view of IC structure 20, and Fig. 9B A cross-sectional view along dashed line BB′ of IC structure 20 constructed in accordance with some embodiments is shown. Fig. 9A and Fig. 9BThe IC structure 20 shown is similar to Figure 8A-8B The integrated circuit structure 20 shown in FIG. However, some of the backside vias 64 are replaced by backside dielectric vias 88. The backside dielectric vias 88 are dielectric components and are configured to have isolation with enhanced isolation effectiveness. Therefore, the semiconductor islands of the semiconductor substrate 56 are separated and isolated from each other by the backside dielectric vias 88. For clarity, the backside vias 64 are also referred to as backside conductive vias 64.

[0068] exist Fig. 9A In the embodiment of the present invention, the device structure includes an active region oriented along the X direction and a gate oriented along the Y direction. In addition, the backside dielectric layer 82, the dielectric component 68, the dielectric barrier layer 66 and the backside dielectric via 88 are configured to collectively isolate one semiconductor island from an adjacent semiconductor island. In some embodiments, the device structure includes a dielectric gate, a dielectric gate cut component, or both to provide additional isolation effects to the semiconductor island and the FET formed thereon.

[0069] Other components and methods can be used to provide further isolation. For example, Fig. 9B As shown, the metal line 80 on the back side extends along the X direction. Figure 8B In FIG. 8 , the metal lines 80 on the back side are segmented, and dielectric features (eg, backside ILD layer 84 ) are inserted to provide additional isolation.

[0070] Fig. 10A , Fig. 10B and Fig. 10C A flow chart of a method 150 of fabricating an IC structure (or workpiece) 200 constructed in accordance with some embodiments is shown. Figures 11A-16B 2 is a perspective view or a cross-sectional view of an IC structure 200 constructed according to some embodiments at various manufacturing stages. Specifically, Fig.11A , Fig. 12A , Fig.13A , Fig.14A , Fig.15A and Fig.16A is a perspective view of an IC structure 200; Fig. 11B , Fig. 12B and Fig. 13B is a cross-sectional view of the IC structure 200 along the dashed line AA′; Fig. 14B , Fig. 15B and Fig. 16B is a cross-sectional view of the IC structure 200 along the dashed line BB'. Figures 10A-16B Method 150 is further described. In some examples, IC structure (or workpiece) 200 is a portion or all of IC structure 20 .

[0071] refer to Fig. 10A , Fig.11A and Fig. 11BThe method 150 begins at box 152 by providing or receiving a workpiece 200, which includes various devices, including SRAM devices configured in an array. The device includes a FET of any suitable structure formed on the front side of the semiconductor substrate 56, such as a planar FET, a fin FET, a GAAFET, a complementary FET (CFET), other suitable devices or combinations thereof, and an interconnect structure 204 including contacts, vias and metal lines is formed on the device. The interconnect structure 204 is also referred to as a front-side interconnect structure 204. In the disclosed embodiment, the device is a GAAFET having a gate structure surrounding each of a plurality of vertically stacked channel layers 78.

[0072] Note that artifact 200 is Fig.11A and Fig. 11B 56 is inverted in the semiconductor substrate 56, so the front side of the substrate 56 is shown at the bottom and the back side of the substrate 56 is shown at the top. Specifically, an etch stop layer 210 is formed in the semiconductor substrate 56. The etch stop layer 210 is used as an etch stop layer during the back side process, as will be described in a later operation of the method 150. The etch stop layer 210 is embedded in the semiconductor substrate 56 and has a material different from the semiconductor substrate 56 for etching selectivity. The etch stop layer 21 includes any suitable material to achieve etching selectivity, such as silicon oxide, silicon nitride, other dielectric materials, other suitable materials or combinations thereof. In some embodiments, the semiconductor substrate 56 is a silicon substrate, and the etch stop layer 210 is silicon germanium or silicon oxide. The etch stop layer 210 can be formed in the semiconductor substrate 56 by any suitable method, such as implanting oxygen (SIMOX), or implanting to introduce other components. In some embodiments, the etch stop layer 210 is formed by epitaxial growth, such as epitaxially growing a silicon germanium layer on the semiconductor substrate 56 as an etch stop layer; and epitaxially growing a silicon layer on the silicon germanium layer so that the silicon germanium layer is embedded in the semiconductor substrate 56.

[0073] The formation of the front side structure includes forming the device and the front side interconnect structure 204 as described above, and also includes forming other components and assemblies, such as a gate cutting component 206 and a dielectric gate 208. The gate cutting component 206 is a dielectric component and is formed to cut the long gate structure into a segmented gate structure. The gate cutting component 206 can be formed before, during or after forming the gate structure 60, and is longitudinally oriented along the X direction, while the gate structure 60 is longitudinally oriented along the Y direction. The dielectric gate 208 is also a dielectric component, but is longitudinally oriented along the Y direction and parallel to the gate structure. In some embodiments, a dummy gate structure is formed, and then the dummy gate structure is replaced by the gate structure and the dielectric gate 208, respectively. The gate cutting component 206 is formed to cut the gate structure into a segmented gate structure 60.

[0074] Still reference Fig. 10A , Fig.11Aand Fig. 11B , the method 150 proceeds to operation 154, bonding the carrier substrate 202 to the front side of the workpiece 200; and proceeds to operation 156, thinning the semiconductor substrate 56 from the back side. At operation 156, after bonding, the semiconductor substrate 56 is thinned from the back side by a suitable technique, such as grinding, chemical mechanical polishing, or a combination thereof. In the disclosed embodiment, the thinning process reduces the thickness of the substrate 56 so that the STI structure 58 is exposed from the back side. According to some embodiments, the carrier substrate 202 is a semiconductor substrate (such as a silicon substrate), a dielectric substrate, or other suitable substrate.

[0075] exist Fig. 10C Detailed operations for forming devices and interconnect structures are further described in the flowchart of FIG.

[0076] refer to Fig. 10A , Fig. 12A and Fig. 12B , the method 150 proceeds to operation 158 by performing a wet etching process with an etching solution to selectively etch the semiconductor material, such as silicon, of the substrate 56. Due to the selective etching, the wet etching process stops on the etch stop layer 210, thereby exposing the etch stop layer from the back side after the wet etching. In some embodiments, the wet etching process uses a potassium hydroxide (KOH) solution, or an etching solution including nitric acid (HNO3), hydrofluoric acid (HF), and water (H2O).

[0077] refer to Fig. 10A , Fig.13A and Fig. 13B , the method 150 proceeds to operation 160, where the etch stop layer 210 is selectively removed by a suitable method, such as another wet etching process using an etchant, to selectively remove the etch stop layer 210. Thereafter, the semiconductor substrate 56 with the associated active regions 62 is exposed from the back side. In some embodiments, if the etch stop layer 210 is silicon oxide, the etching solution includes dilute hydrofluoric acid.

[0078] refer to Fig. 10A , Fig.14A and Fig. 14B The method 150 proceeds to operation 162 by depositing a dielectric material layer 212 on the back side using a suitable method such as chemical vapor deposition (CVD), flowable CVD (FCVD), other suitable methods, or combinations thereof. The dielectric material layer 212 may include silicon oxide, silicon oxynitride, other suitable dielectric materials, or combinations thereof.

[0079] refer to Fig. 10A , Fig.15A and Fig. 15BThe method 150 proceeds to operation 164 by performing a CMP process on the backside to recess and planarize the backside until both the STI structure 58 and the semiconductor substrate 56 are exposed from the backside. Therefore, the bottom surfaces of the STI structure 58 and the semiconductor substrate 56 are coplanar.

[0080] refer to Fig. 10A , Fig.16A and Fig. 16B , the method 150 proceeds to operation 166, and a backside dielectric layer 82 is formed on the backside using a suitable method such as chemical vapor deposition (CVD), flowable CVD (FCVD), other suitable methods, or combinations thereof. According to some embodiments, the backside dielectric layer 82 may include silicon oxide, silicon oxynitride, silicon nitride, other suitable dielectric materials, or combinations thereof. The backside dielectric layer 82 may be used as a hard mask, an etch stop layer, dielectric isolation, other functions, or combinations thereof.

[0081] The method 150 proceeds to operation 168 by forming a backside interconnect structure including backside vias, backside dielectric vias, and backside metal lines distributed in one or more metal layers.

[0082] The method 150 may include other processes before, during, or after the above operations.

[0083] The backside interconnect structure 63 formed at operation 168 is similar in formation and composition to the frontside interconnect structure 204. For example, the backside interconnect structure 63 includes backside vias 64, metal lines 80, and vias distributed in one or more metal layers, and can be formed by suitable techniques, such as a damascene process, a dual damascene process, a procedure including deposition and patterning, other suitable methods, or a combination thereof. In some embodiments, the backside interconnect structure 63 includes a backside via 64, a metal line 80, and a via distributed in one or more metal layers. Figure 8A-8D and Figure 9A-9B The backside vias 64 and backside metal lines 80 are formed by the method described above.

[0084] For example, Fig. 10B As shown, and further reference Figure 8A-8D and Figure 9A-9B , method 168 of forming backside interconnect structure 63 includes operation 170, in which backside dielectric layer 82 and semiconductor substrate 56 are patterned to form contact holes. Backside dielectric layer 82 can be used as a hard mask during the patterning process. Method 168 proceeds to operation 172, using a suitable technique (e.g., a process including depositing one or more dielectric materials (e.g., silicon oxide, silicon nitride, silicon oxynitride, or other suitable dielectric materials or combinations thereof) and applying plasma etching to the dielectric materials) to form dielectric barrier layer 66 on the sidewalls of the contact hole. Method 168 proceeds to operation 174, forming metal plug 64 in the contact hole, such as Figure 8BAs shown. Operation 174 may include deposition and application of a CMP process. Method 168 proceeds to operation 176 to form a backside interlayer dielectric layer 84 by a suitable method such as deposition and application of a CMP process. Method 168 proceeds to operation 178 to pattern the backside interlayer dielectric layer 84 to form a trench. According to some embodiments, method 168 proceeds to operation 180 to form a metal line in the trench using a procedure including deposition and a CMP process.

[0085] Operation 152 of forming a front-side device (such as a GAAFET or other multi-gate device) and a front-side interconnect structure 204 includes various operations, such as Fig. 10C shown.

[0086] In some embodiments, method 152 manufactures a multi-gate device including a p-type GAA transistor and an n-type GAA crystal. In some embodiments, method 102 manufactures a multi-gate device including a first GAA transistor and a second GAA transistor having different characteristics, such as a first GAA transistor in a critical path and a second GAA transistor in a non-critical path. In this embodiment, a path is defined as a path for distributing a signal in a circuit. The critical path is the location that mainly controls the circuit speed (or signal distribution speed), which depends on different circuit applications. On the other hand, if the circuit speed varies significantly with the performance of the transistor, the signal path will be referred to as a critical path. In some aspects, the critical path and the non-critical GAA path may have different power consumption during field operation. In an integrated circuit, the current (and electrical power) in the circuit may be unevenly distributed. The average current density of some local areas is greater than the current density of other local areas. Those areas with larger average current density are called critical paths, which lead to various problems, such as reducing power efficiency, reducing circuit performance, reducing circuit speed, increasing unit size, and causing reliability problems. In existing methods, the device size is increased, such as the channel width of the transistor in the critical path, to adjust or reduce the corresponding average current density. However, existing methods will add other problems. For example, the circuit area is increased and the packaging density is reduced. In other examples, the adjustment of the size of the device in the critical path introduces a step (jog) in the active area, which further increases the complexity of the circuit layout and poses challenges to the circuit design due to the smaller circuit cell height and gate pitch in advanced technology nodes.

[0087] The disclosed multi-gate device and its manufacturing method solve these problems. Specifically, in order to improve performance, the present disclosure selects high-drive devices (or devices with more sheet count devices) at critical paths; and selects low-power devices (or devices with fewer sheet count devices) at non-critical paths.

[0088] At frame 182, a semiconductor layer stack is formed over the substrate. The semiconductor layer stack includes a first semiconductor layer and a second semiconductor layer vertically stacked in an alternating configuration. At frame 184, a gate structure is formed over a first region of the semiconductor layer stack. The gate structure includes a dummy gate stack and a gate spacer. At frame 186, a portion of the semiconductor layer stack in the second region is removed to form a source / drain recess. At frame 188, an internal spacer is formed along the sidewall of the first semiconductor layer in the semiconductor layer stack. At frame 190, an epitaxial source / drain component is formed in the source / drain recess. At frame 192, an interlayer dielectric (ILD) layer is formed over the epitaxial source / drain component. At frame 194, the dummy gate stack is removed to form a gate trench that exposes the semiconductor layer stack in the gate region. At frame 196, the first semiconductor layer is removed from the semiconductor layer stack exposed by the gate trench to form a gap between the second semiconductor layers. At block 198, a gate stack is formed in the gate trench around the second semiconductor layer in the gate region. At block 199, other manufacturing processes including forming an interconnect structure are performed from the front side of the workpiece. The present disclosure contemplates additional processing. Additional steps may be provided before, during, and after method 152, and some of the steps described may be moved, replaced, or eliminated for additional embodiments of method 152.

[0089] According to various embodiments of the present disclosure, the IC structure 20 may include other layouts and configurations, such as Figures 17 to 24 Layout and configuration shown.

[0090] Fig.17 FIG. 2 is a top view of an IC structure 20 according to some embodiments. The SRAM region 22 and the edge region 24 of the IC structure 20 are shown in FIG. Fig.17 As shown. The SRAM region 22 includes a plurality of SRAM cells arranged in an array. The N-well 42 and the P-well 44 extend from the edge region 24 to the SRAM region 22 and extend through the plurality of SRAM cells 28 along the X direction. The active regions 36 and 38 are formed on the N-well 42 and the P-well 44, respectively, and are longitudinally oriented along the X direction. Specifically, the active regions 36 and 38 span different widths along the Y direction, as described above, for example Fig. 6A 4. The gate structure 40 is formed above the active area and is oriented longitudinally along the Y direction. The contact 90 and the through hole 48 of the front side interconnect structure are shown. The contact 90 has an elongated shape and is oriented longitudinally along the Y direction, and lands on the S / D component 70. The through hole 48 is configured to land on the contact 90 and electrically connect the contact to the metal line 92 in the first metal layer. The metal line 92 is oriented longitudinally along the X direction. Various circuit components are designed and configured to have a robust edge region and a reduced size edge region 24. In the present embodiment, the ratio De / Ds is equal to 2.

[0091] Fig.18 FIG. 2 is a top view of an IC structure 20 according to some embodiments. The SRAM region 22 and the edge region 24 of the IC structure 20 are shown in FIG. Fig.18 As shown. The SRAM region 22 includes a plurality of SRAM cells arranged in an array. The N-well 42 and the P-well 44 extend from the edge region 24 to the SRAM region 22 and extend through the plurality of SRAM cells 28 along the X direction. The active regions 36 and 38 are formed on the N-well 42 and the P-well 44, respectively, and are longitudinally oriented along the X direction. Specifically, the active regions 36 and 38 span different widths along the Y direction, as described above, for example Fig. 6A 4. The gate structure 40 is formed above the active area and is oriented longitudinally along the Y direction. The contact 90 and the through hole 48 of the front side interconnect structure are shown. The contact 90 has an elongated shape and is oriented longitudinally along the Y direction, and lands on the S / D component 70. The through hole 48 is configured to land on the contact 90 and electrically connect the contact to the metal line 92 in the first metal layer. The metal line 92 is oriented longitudinally along the X direction. Various circuit components are designed and configured to have a robust edge region and a reduced size edge region 24. In the present embodiment, the ratio De / Ds is equal to 2.

[0092] Fig.19 FIG. 2 is a top view of an IC structure 20 according to some embodiments. The SRAM region 22 and the edge region 24 of the IC structure 20 are shown in FIG. Fig.19 As shown. The SRAM region 22 includes a plurality of SRAM cells arranged in an array. The N-well 42 and the P-well 44 extend from the edge region 24 to the SRAM region 22 and extend through the plurality of SRAM cells 28 along the X direction. The active regions 36 and 38 are formed on the N-well 42 and the P-well 44, respectively, and are longitudinally oriented along the X direction. Specifically, the active regions 36 and 38 span the same width along the Y direction, as described above, for example Figure 6B 4. The gate structure 40 is formed above the active area and is oriented longitudinally along the Y direction. The contact 90 and the through hole 48 of the front side interconnect structure are shown. The contact 90 has an elongated shape and is oriented longitudinally along the Y direction, and lands on the S / D component 70. The through hole 48 is configured to land on the contact 90 and electrically connect the contact to the metal line 92 in the first metal layer. The metal line 92 is oriented longitudinally along the X direction. Various circuit components are designed and configured to have a robust edge region and a reduced size edge unit 24. In the present embodiment, the ratio De / Ds is equal to 2.

[0093] Fig. 20 is a partial top view of IC structure 20 according to some embodiments. Fig. 20 The IC structure 20 is similar to Fig.19The IC structure in FIG. 1 is shown, but the IOP region 26 and the second metal line 94 are additionally shown. For better viewing, other components such as the active region and the gate structure are not shown. Fig. 20 , a portion of the front-side interconnect structure is shown, such as a first metal line 92 in a first metal layer and a second metal line 94 in a second metal layer. The metal lines are laid out and other components (such as Fig.19 The active regions and gate structures in the SRAM region 22 are provided to provide proper electrical routing, enhance the transition between the SRAM cells 28 in the SRAM region 22 and the IOP circuits in the IOP region 26, and reduce the packaging area of ​​the edge region 24.

[0094] Fig.21 FIG. 2 is a top view of an IC structure 20 according to some embodiments. The SRAM region 22 and the edge region 24 of the IC structure 20 are shown in FIG. Fig.21 As shown. The SRAM region 22 includes a plurality of SRAM cells arranged in an array. The N-well 42 and the P-well 44 extend from the edge region 24 to the SRAM region 22 and extend through the plurality of SRAM cells 28 along the X direction. The active regions 36 and 38 are formed on the N-well 42 and the P-well 44, respectively, and are longitudinally oriented along the X direction. Specifically, the active regions 36 and 38 span different widths along the Y direction, as described above, for example Fig. 6A As shown. The gate structure 40 is formed above the active area and is oriented longitudinally along the Y direction. Contacts 90 and through-holes 48 of the front-side interconnect structure are shown. The contact 90 has an elongated shape and is oriented longitudinally along the Y direction, and lands on the S / D component 70. The through-hole 48 is configured to land on the contact 90 and electrically connect the contact to a metal line 92 in the first metal layer. The metal line 92 is oriented longitudinally along the X direction. Various circuit components are designed and configured to have a robust edge region and a reduced size edge region 24. In the present embodiment, the edge region 24 is further reduced in size. Da1=1.5*Pg, Da2=0.5*Pg and Dc=Pg. In addition, Ds=2*Pg, De=3*Pg and the ratio De / Ds is equal to 1.5, which is less than 2.

[0095] Fig. 22 is a partial top view of IC structure 20 according to some embodiments. Fig. 22 The IC structure 20 is similar to Fig.21 The IC structure in FIG. 1 is shown, but a second metal line 94 is additionally shown. For better viewing, other components such as active regions and gate structures are not shown. Fig. 22, portions of the front-side interconnect structure are shown, such as a first metal line 92 in a first metal layer, a second metal line 94 in a second metal layer, and a through-hole 96 vertically connecting the corresponding first metal line 92 and the second metal line 94. Various circuit components are designed and configured to have robust edge regions and reduced-size edge regions 24. For example, the IC structure 20 also includes a through-hole 98 away from any intersection of the first metal line 92 and the second metal line 94. In this case, the through-hole 98 extends vertically from the contact 90 to the second metal line 94. This connection is called a jumper, which provides flexibility for the metal lines and saves circuit area when the IOP area 26 and the SRAM area 22 are configured differently, such as the first metal line in the IOP area 26 can have a different pitch (or spacing) than the first metal line in the SRAM area 22. The metal lines are laid out and configured with other components (such as Fig.21 The active regions and gate structures in the SRAM region 22 are provided to provide proper electrical routing, enhance the transition between the SRAM cells 28 in the SRAM region 22 and the IOP circuits in the IOP region 26, and reduce the packaging area of ​​the edge region 24.

[0096] Fig.23 FIG. 2 is a top view of an IC structure 20 according to some embodiments. The SRAM region 22 and the edge region 24 of the IC structure 20 are shown in FIG. Fig.23 As shown. The SRAM region 22 includes a plurality of SRAM cells arranged in an array. The N-well 42 and the P-well 44 extend from the edge region 24 to the SRAM region 22 and extend through the plurality of SRAM cells 28 along the X direction. The active regions 36 and 38 are formed on the N-well 42 and the P-well 44, respectively, and are longitudinally oriented along the X direction. Specifically, the active regions 36 and 38 span different widths along the Y direction, as described above, for example Fig. 6A As shown. The gate structure 40 is formed above the active area and is oriented longitudinally along the Y direction. Contacts 90 and through holes 48 of the front-side interconnect structure are shown. The contact 90 has an elongated shape and is oriented longitudinally along the Y direction, and lands on the S / D component 70. The through hole 48 is configured to land on the contact 90 and electrically connect the contact to a metal line 92 in the first metal layer. The metal line 92 is oriented longitudinally along the X direction. Various circuit components are designed and configured to have a robust edge region and a reduced size edge region 24. For example, the edge region 24 does not have a pseudo active region, but only a portion of the active region in the SRAM region 22 extends to the edge region 24 with a length of 0.5*Pg or less. In the present embodiment, the edge region 24 is further reduced in size. Da1=1.5*Pg, Da2=0.5*Pg and Dc=Pg. Da2 is defined as the distance from the end of the active region to the boundary line of the SRAM region 22. Furthermore, Ds=2*Pg, De=2*Pg, and the ratio De / Ds is equal to 1, which is less than 2.

[0097] Fig.24 is a partial top view of IC structure 20 according to some embodiments. Fig.24 The IC structure 20 is similar to Fig.23 The IC structure in FIG. 1 is shown, but a second metal line 94 is additionally shown. For better viewing, other components such as active regions and gate structures are not shown. Fig.24 , a portion of the front-side interconnect structure is shown, such as a first metal line 92 in a first metal layer, a second metal line 94 in a second metal layer, and a via 96 vertically connecting the corresponding first metal line 92 and second metal line 94. The metal lines are laid out and other components (such as Fig.21 The active regions and gate structures in the SRAM region 22 are provided to provide proper electrical routing, enhance the transition between the SRAM cells 28 in the SRAM region 22 and the IOP circuits in the IOP region 26, and reduce the packaging area of ​​the edge region 24.

[0098] The present disclosure provides many different embodiments. The disclosed device structure and its manufacturing method relate to an SRAM structure including a field effect transistor (FET), in particular a GAAFET structure of various layouts and configurations. The disclosed device structure includes various structural features and manufacturing steps to provide collective isolation and prevent current leakage of the device structure. The size of the edge region between the SRAM region and the IOP region is significantly reduced. De / Ds is equal to 2 or less than 2, for example, 1.5 or 1. Take the disclosed structure with an SRAM device structure and its manufacturing method as an example. In addition, the disclosed structure and method are also compatible with other manufacturing technologies without enhancing circuit packaging density and power efficiency.

[0099] The present disclosure provides an integrated circuit structure and a method for manufacturing such a structure. In one exemplary aspect, the exemplary integrated circuit structure includes: a semiconductor substrate having a static random access memory (SRAM) region, an input / output and peripheral (IOP) region, and an edge region spanning between the SRAM region and the IOP region; a shallow trench isolation (STI) structure formed on the semiconductor substrate and defining an active region; an SRAM cell formed in the SRAM region; and a backside dielectric layer disposed on the backside of the semiconductor substrate and landing on a bottom surface of the STI structure, wherein the active region is longitudinally oriented along a first direction, a plurality of gates are formed on the semiconductor substrate and longitudinally oriented along a second direction perpendicular to the first direction, the gates are uniformly distributed along the first direction at a pitch P, the SRAM cell spans a first dimension Ds along the first direction, the edge region spans a second dimension De along the first direction, and the ratio De / Ds is equal to or less than 2.

[0100] In some embodiments, Ds is equal to 2P and De is equal to 4P.

[0101] In some embodiments, Ds is equal to 2P and De is equal to 2P.

[0102] In some embodiments, a bottom surface of the STI structure is coplanar with a bottom surface of the semiconductor substrate.

[0103] In some embodiments, the SRAM cell includes two inverters that are cross-coupled and includes a field effect transistor (FET); each active region includes a plurality of channel layers that are vertically stacked and spaced apart from each other; and the FET includes a source, a drain, and a gate between the source and the drain, wherein the gate further extends to wrap around each of the plurality of channel layers.

[0104] In some embodiments, one of the source and the drain further includes a dielectric material layer embedded in the epitaxial semiconductor component.

[0105] In some embodiments, the IC structure further includes a backside via formed on the back side of the semiconductor substrate and landing on one of the source and the drain, wherein the backside via is partially embedded in the semiconductor substrate; and the backside via includes a conductive plug, the dielectric layer surrounding the sidewalls of the conductive plug and separating the conductive plug from the semiconductor substrate.

[0106] In some embodiments, a bottom surface of the backside via is coplanar with a bottom surface of the backside dielectric layer.

[0107] In some embodiments, the IC structure further includes a backside dielectric via formed on the backside of the semiconductor substrate, wherein the backside dielectric via is surrounded by the semiconductor substrate and laterally contacts the backside dielectric layer.

[0108] In another example aspect, the present disclosure provides an IC structure comprising: a semiconductor substrate having a static random access memory (SRAM) region, an input / output and peripheral (IOP) region, and an edge region spanning between the SRAM region and the IOP region; an n-type doped well extending continuously along a first direction through the SRAM region and the edge region; a p-type doped well extending continuously along the first direction through the SRAM region and the edge region; a shallow trench isolation (STI) structure formed on the semiconductor substrate and defining a first active region, a second active region, and a third active region oriented longitudinally along the first direction; a plurality of gates oriented longitudinally along a second direction perpendicular to the first direction, wherein the gates are uniformly distributed in the SRAM region and the edge region with a periodic pitch P; and a backside dielectric layer contacting a bottom surface of the semiconductor substrate and a bottom surface of the STI structure, wherein the SRAM region comprises SRAM cells, each of the SRAM cells spanning a first dimension Ds along the first direction, the edge region spanning a second dimension De along the first direction, and a ratio De / Ds is equal to or less than 2.

[0109] In some embodiments, the first active region is formed in a p-type doped well and extends continuously from the SRAM region to the edge region; the second active region is formed in an n-type doped well and disposed within the SRAM region; and the third active region is formed in an n-type doped well and disposed within the edge region.

[0110] In some embodiments, the first active region spans a first width W1 along the second direction; the second active region and the third active region span a second width W2 along the second direction, W1 being greater than W2; and the second active region and the third active region are aligned along the first direction.

[0111] In some embodiments, the IC structure further includes: a first source / drain (S / D) component formed on the first active region; a second S / D component formed on the second active region; a first backside via contacting a bottom surface of the first S / D component; and a second backside via contacting a bottom surface of the second S / D component.

[0112] In some embodiments, each of the first backside via and the second backside via is partially embedded in the semiconductor substrate and includes a conductive plug, and the dielectric layer surrounds a sidewall of the conductive plug and separates the conductive plug from the semiconductor substrate.

[0113] In some embodiments, the IC structure further includes a backside dielectric via formed on the backside of the semiconductor substrate, wherein the backside dielectric via is surrounded by the semiconductor substrate and laterally contacts the backside dielectric layer.

[0114] In some embodiments, Ds is equal to 2P, De is equal to 4P; and a bottom surface of the STI structure is coplanar with a bottom surface of the semiconductor substrate.

[0115] In another exemplary aspect, the present disclosure provides a method for manufacturing an integrated circuit (IC) structure. The method includes: receiving a semiconductor substrate having a front side and a back side, wherein the semiconductor substrate includes a static random access memory (SRAM) region, an input / output and peripheral (IOP) region, and an edge region spanning between the SRAM region and the IOP region; forming an SRAM structure having SRAM cells in the SRAM region on the front side of the semiconductor substrate; forming an interconnect structure above the SRAM structure; thinning the semiconductor substrate from the back side so that a shallow trench isolation (STI) structure is exposed from the back side; and forming a backside dielectric layer directly on a bottom surface of the STI structure and a bottom surface of the semiconductor substrate.

[0116] In some embodiments, forming an SRAM structure having an SRAM cell further includes: forming an n-type doped well extending continuously along a first direction through the SRAM region and the edge region; forming a p-type doped well extending continuously along the first direction through the SRAM region and the edge region; forming a first active region, a second active region, and a third active region oriented longitudinally along the first direction and surrounded by an STI structure; and forming a plurality of gates, the plurality of gates being longitudinally oriented along a second direction perpendicular to the first direction, wherein the gates are uniformly distributed in the SRAM region and the edge region with a periodic pitch P, wherein each SRAM cell spans a first dimension Ds along the first direction, the edge region spans a second dimension De along the first direction, and the ratio De / Ds is equal to 2.

[0117] In some embodiments, the first active region is formed in a p-type doped well and extends continuously from the SRAM region to the edge region; the second active region is formed in an n-type doped well and disposed within the SRAM region; and the third active region is formed in an n-type doped well and disposed within the edge region.

[0118] In some embodiments, the method further includes: forming a first source / drain (S / D) component on the first active region, the first dielectric component isolating the first S / D component from the semiconductor substrate; forming a second S / D component on the second active region, the second dielectric component isolating the second S / D component from the semiconductor substrate; forming a first backside via in contact with a bottom surface of the first S / D component; and forming a second backside via in contact with a bottom surface of the second S / D component, wherein each of the first backside via and the second backside via is partially embedded in the semiconductor substrate and includes a conductive plug, and the dielectric layer surrounds sidewalls of the conductive plug and separates the conductive plug from the semiconductor substrate.

[0119] The features of several embodiments are summarized above so that those skilled in the art can better understand the various aspects of the present disclosure. Those skilled in the art will appreciate that they can easily use the present disclosure as a basis for designing or modifying other processes and structures for achieving the same purpose of the embodiments introduced herein and / or achieving the same advantages thereof. Those skilled in the art will also appreciate that such equivalent structures do not deviate from the spirit and scope of the present disclosure, and that they can make various changes, substitutions, and changes in the present disclosure without departing from the spirit and scope of the present disclosure.

Claims

1. An integrated circuit structure, comprising: A semiconductor substrate having a static random access memory area, an input / output and peripheral area, and an edge area spanning between the static random access memory area and the input / output and peripheral area; A shallow trench isolation structure is formed on the semiconductor substrate and defines an active area; a static random access memory unit formed in the static random access memory area; as well as A backside dielectric layer is disposed on the backside of the semiconductor substrate and lands on the bottom surface of the shallow trench isolation structure, wherein: The active region is longitudinally oriented along a first direction, A plurality of gates are formed on the semiconductor substrate and are longitudinally oriented along a second direction perpendicular to the first direction, The gates are uniformly distributed along the first direction with a pitch P, The SRAM cell spans a first size Ds along the first direction, The edge region spans a second dimension De along the first direction, and The ratio De / Ds is equal to 2 or less.

2. The integrated circuit structure according to claim 1, wherein: A bottom surface of the shallow trench isolation structure is coplanar with a bottom surface of the semiconductor substrate.

3. The integrated circuit structure according to claim 1, wherein The static random access memory cell includes two inverters that are cross-coupled and includes a field effect transistor; Each of the active regions includes a plurality of channel layers vertically stacked and spaced apart from each other; and The field effect transistor comprises a source, a drain, and a gate between the source and the drain, wherein: The gate further extends to wrap around each of the plurality of channel layers.

4. The integrated circuit structure of claim 3, further comprising a backside via formed on the backside of the semiconductor substrate and landing on one of the source and the drain, wherein The backside via is partially embedded in the semiconductor substrate; and The backside via includes a conductive plug, and a dielectric layer surrounds a sidewall of the conductive plug and separates the conductive plug from the semiconductor substrate.

5. The integrated circuit structure of claim 4, further comprising a backside dielectric via formed on the backside of the semiconductor substrate, wherein The backside dielectric via is surrounded by the semiconductor substrate and laterally contacts the backside dielectric layer.

6. An integrated circuit structure comprising: A semiconductor substrate having a static random access memory area, an input / output and peripheral area, and an edge area spanning between the static random access memory area and the input / output and peripheral area; an n-type doped well extending continuously along a first direction through the SRAM region and the edge region; a p-type doped well extending continuously along the first direction through the SRAM region and the edge region; a shallow trench isolation structure formed on the semiconductor substrate and defining a first active region, a second active region, and a third active region longitudinally oriented along the first direction; a plurality of gates, longitudinally oriented along a second direction perpendicular to the first direction, wherein the gates are uniformly distributed in the SRAM region and the edge region at a periodic pitch P; as well as A back dielectric layer contacts the bottom surface of the semiconductor substrate and the bottom surface of the shallow trench isolation structure, wherein The SRAM area includes SRAM cells, each of which spans a first size Ds along the first direction, The edge region spans a second dimension De along the first direction, and The ratio De / Ds is equal to 2 or less.

7. The integrated circuit structure of claim 6, wherein The first active region is formed in the p-type doped well and extends continuously from the SRAM region to the edge region; The second active region is formed in the n-type doped well and is disposed within the static random access memory region; and The third active region is formed in the n-type doped well and disposed within the edge region.

8. The integrated circuit structure of claim 7, wherein The first active region spans a first width W1 along the second direction; The second active region and the third active region span a second width W2 along the second direction, W1 being greater than W2; and The second active region and the third active region are aligned along the first direction.

9. A method of manufacturing an integrated circuit structure, comprising: Receiving a semiconductor substrate having a front side and a back side, wherein the semiconductor substrate includes a static random access memory area, an input / output and peripheral area, and an edge area spanning between the static random access memory area and the input / output and peripheral area; forming an SRAM structure having SRAM cells in an SRAM region on the front side of the semiconductor substrate; forming an interconnect structure over the static random access memory structure; Thinning the semiconductor substrate from the back side so that a shallow trench isolation structure is exposed from the back side; and A backside dielectric layer is formed directly on the bottom surface of the shallow trench isolation structure and the bottom surface of the semiconductor substrate.

10. The method according to claim 9, wherein: Forming a static random access memory structure having a static random access memory cell further comprises: forming an n-type doped well extending continuously along a first direction through the SRAM region and the edge region; forming a p-type doped well extending continuously along the first direction through the SRAM region and the edge region; forming a first active region, a second active region, and a third active region oriented longitudinally along the first direction and surrounded by the shallow trench isolation structure; and forming a plurality of gates, the plurality of gates being longitudinally oriented along a second direction perpendicular to the first direction, wherein the gates are uniformly distributed in the static random access memory area and the edge area at a periodic pitch P, wherein Each of the SRAM cells spans a first size Ds along the first direction, The edge region spans a second dimension De along the first direction, and a ratio De / Ds is equal to two.