Semiconductor structure and memory unit
By using a fully wound gate field-effect transistor design and optimizing the active region width, the problem of insufficient read and write performance in traditional SRAM devices during miniaturization is solved, achieving a more efficient improvement in SRAM cell performance.
Patent Information
- Application Number
- CN202420365150.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- Priority Date
- 2023-08-03
- Filing Date
- 2024-02-27
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2034-02-27
AI Technical Summary
Traditional SRAM devices cannot meet the requirements of next-generation SRAM devices during miniaturization, especially in terms of read and write performance optimization. Furthermore, FinFET devices may experience cell mismatch issues during miniaturization.
Employing a fully wound gate field-effect transistor (GAA FET) design, the transistor width ratio between the read and write ends is optimized by adjusting the width of the active region and the configuration of the gate structure, and the gate structure is cut using dielectric components to improve gate control and reduce short-channel effects.
It improves the read speed and write performance of SRAM cells, reduces parasitic capacitance and leakage current, solves the cell mismatch problem in the miniaturization process of FinFET devices, and realizes more flexible channel miniaturization.
Smart Images

Figure CN221977605U_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to semiconductor technology, and more particularly to semiconductor structures and methods for manufacturing the same. Background Art
[0002] The semiconductor integrated circuit (IC) industry has experienced rapid growth. Technological advances in IC materials and design have produced successive generations of ICs, each with smaller and more complex circuits than the previous one. During the evolution of ICs, functional density (i.e., the number of interconnect devices per unit chip area) has generally increased while geometry size (i.e., the smallest component (or line) that can be produced using a process) has decreased. This process of miniaturization generally provides benefits by increasing production efficiency and reducing associated costs. Such size reductions also increase the complexity of processing and manufacturing ICs.
[0003] In deep sub-micron integrated circuit technology, static random-access memory (SRAM) devices have become a common storage unit for high-speed communications, image processing, and system-on-chip (SOC) products. The number of embedded SRAMs in microprocessors and SOCs continues to increase to meet the performance requirements of new generation technologies. As silicon technology continues to scale from one generation to the next, traditional SRAM devices and / or their manufacturing may encounter limitations. For example, the read and write ends of an SRAM device may have different considerations and requirements. However, traditional SRAM designs do not fully consider these differences. Therefore, the SRAM read performance and / or write performance has not been fully optimized. As another example, SRAM devices are traditionally manufactured based on planar devices or FinFET devices. As device scaling continues, planar devices and even FinFET devices may not meet the needs or flexibility requirements of new generation SRAM devices.
[0004] Thus, while existing SRAM devices are generally adequate for their intended purposes, they are not completely satisfactory in every aspect. Utility Model Content
[0005] The present disclosure provides a memory cell, comprising: a first active region, the first active region providing a plurality of first nanostructures for a write-port pass-gate transistor, the first active region having a first width; a second active region, the second active region providing a plurality of second nanostructures for a write-port pull-up transistor, the second active region having a second width; and a third active region, the third active region providing a plurality of third nanostructures for a read-port pull-down transistor, the third active region having a third width, wherein the third width is greater than the first width, and the first width is greater than the second width.
[0006] In some embodiments, a ratio of the first width to the second width is between about 1.5:1 and about 5:1.
[0007] In some embodiments, a ratio of the third width to the first width is between about 1.1:1 and about 3:1.
[0008] In some embodiments, the memory cell further includes: a gate structure line extending across the first active region, the second active region, and the third active region; and a gate cutting dielectric component that divides the gate structure line into a first gate structure surrounding each first nanostructure and a second gate structure surrounding each second nanostructure and each third nanostructure.
[0009] In some embodiments, the gate cutting dielectric feature extends vertically deeper in an inner region of the gate structure line than in an outer region of the gate structure line.
[0010] In some embodiments, the memory cell further includes a fourth active region, the fourth active region providing a plurality of fourth nanostructures for a write-end pull-up transistor, the fourth active region having a fourth width, wherein the fourth width is smaller than the first width and different from the second width.
[0011] The present disclosure provides a semiconductor structure, comprising: a first active region and a second active region, the first active region and the second active region extending longitudinally along a first direction; a gate stack extending longitudinally along a second direction perpendicular to the first direction; a dielectric component disposed between the first active region and the second active region, wherein the dielectric component divides the gate stack into a first segment above the first active region and a second segment above the second active region; a first epitaxial component disposed on the first active region; a second epitaxial component disposed on the second active region, wherein the first epitaxial component and the second epitaxial component are disposed on opposite sides of the dielectric component; and a back conductive component electrically coupled to the first epitaxial component and the second epitaxial component and straddling the dielectric component.
[0012] In some embodiments, the back conductive component includes a first foot located on a bottom surface of the first epitaxial component, a second foot located on a bottom surface of the second epitaxial component, and a middle portion physically connecting the first foot and the second foot, wherein the middle portion is located directly below the dielectric component.
[0013] In some embodiments, the semiconductor structure further includes a front contact disposed on the first epitaxial component and the second epitaxial component, wherein the dielectric component is vertically stacked between the front contact and the back conductive component.
[0014] In some embodiments, the first section and the second section of the gate stack are configured to respectively match the first active region and the second active region to form a plurality of first transistors and a plurality of second transistors in a plurality of first read ends of a plurality of first memory cells and a plurality of second read ends of a plurality of second memory cells in a memory array. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] The embodiments of the present disclosure are best understood by the following detailed description in conjunction with the accompanying drawings. It should be noted that, in accordance with standard practice in the industry, the various components are not drawn to scale. In fact, the sizes of the various elements may be arbitrarily enlarged or reduced to clearly represent the components of the embodiments of the present disclosure.
[0016] Figure 1A and Figure 1B A perspective view and a top view of a portion of a memory device according to some embodiments of the present disclosure are respectively illustrated.
[0017] Figure 2 Cross-sectional views of various layers of a memory device according to some embodiments of the present disclosure are illustrated.
[0018] Figure 3 A circuit schematic diagram of a two-terminal static random access memory (SRAM) cell according to some embodiments of the present disclosure is illustrated.
[0019] Figure 4 The following examples illustrate some embodiments according to the present disclosure. Figure 3 Top view of the device layer of the two-terminal SRAM cell in FIG.
[0020] Figure 5 The following examples illustrate some embodiments according to the present disclosure. Figure 3 Top view of the interconnect structure of the two-terminal SRAM cell in FIG.
[0021] Figure 6 , Figure 7 and Figure 8 Examples of some embodiments according to the present disclosure are as follows Figure 4 and Figure 5 A cross-sectional view of a portion of a two-terminal SRAM cell.
[0022] Fig. 9 According to some embodiments of the present disclosure, Figure 4 and Figure 5 Top view of the layout of a tiled 2×2 memory array of two-terminal SRAM cells in FIG.
[0023] Fig.10 According to some embodiments of the present disclosure, Figure 4 and Figure 5 Backside view of the layout of a tiled 2×2 memory array of two-terminal SRAM cells.
[0024] Fig.11 According to some alternative embodiments of the present disclosure, Figure 4 and Figure 5 Backside view of the layout of a tiled 2×2 memory array of two-terminal SRAM cells.
[0025] Fig.12 A flow chart illustrating a method for forming an integrated circuit having a plurality of SRAM cells according to some embodiments of the present disclosure is illustrated.
[0026] Fig.13 , Fig.14 , Fig.15 , Fig.16 , Fig.17 , Fig.18 , Fig.19 , Fig. 20 and Fig.21 According to some embodiments of the present disclosure, Fig.12 A cross-sectional view of an integrated circuit having a SRAM cell during a manufacturing process of the method.
[0027] The reference numerals are described as follows:
[0028] 10: IC device
[0029] 12: Substrate
[0030] 14: Active Zone
[0031] 15: Doping area
[0032] 16: Source / Drain Components
[0033] 18: Isolation Structure
[0034] 20: Gate structure
[0035] 22: Gate electrode
[0036] 24: Gate dielectric
[0037] 25: Gate spacer
[0038] 26: Nanostructures
[0039] 100: SRAM cell
[0040] 100-1: SRAM cell
[0041] 100-2: SRAM cell
[0042] 100-3: SRAM cell
[0043] 100-4: SRAM cell
[0044] 110A: P-type well
[0045] 110B: N-type well
[0046] 100W: writing end
[0047] 100R: Reading end
[0048] 112: Isolation Structure
[0049] 130: Dielectric components
[0050] 130A: Dielectric components
[0051] 130B: Dielectric components
[0052] 130C: Dielectric parts
[0053] 130D: Dielectric components
[0054] 130E: Dielectric components
[0055] 200: SRAM cells
[0056] 210: Active Zone
[0057] 211: Active Zone
[0058] 212: Active Zone
[0059] 213: Active Zone
[0060] 214: Active Zone
[0061] 220: Gate structure
[0062] 221: Gate structure
[0063] 222: Gate structure
[0064] 223: Gate structure
[0065] 224: Gate structure
[0066] 260: Source / Drain Components
[0067] 300: SRAM bit cell array
[0068] 360A: Gate contact.
[0069] 360B: Gate contact
[0070] 360C: Source / Drain Contacts
[0071] 360D: Gate contact
[0072] 360E: Source / Drain Contacts
[0073] 360F: Source / Drain Contacts
[0074] 360G: Source / Drain Contacts
[0075] 360H: Source / Drain Contacts
[0076] 360I: Source / Drain Contacts
[0077] 360J: Source / Drain Contacts
[0078] 360K: Source / Drain contacts
[0079] 360L: Gate contact
[0080] 360M: Dummy contacts
[0081] 360N: Source / Drain Contacts
[0082] 360P: Gate contact
[0083] 360HB: Back side source / drain contacts
[0084] 360GB: Backside source / drain contacts
[0085] 410: Inter-Layer Dielectric (ILD)
[0086] 420: Top dielectric layer
[0087] 441: Gate electrode components
[0088] 443: Gate electrode components
[0089] 451: Gate dielectric components
[0090] 453: Gate dielectric components
[0091] 470: Gate spacer
[0092] 480: Internal spacer
[0093] 500: Methods
[0094] 502: Operation
[0095] 504: Operation
[0096] 506: Operation
[0097] 508: Operation
[0098] 510: Operation
[0099] 512: Operation
[0100] 514: Operation
[0101] 516: Operation
[0102] 600: Installation
[0103] 602: Substrate
[0104] 604: Isolation Structure
[0105] 606: Semiconductor fins
[0106] 608: Source / Drain Components
[0107] 608a: Buffer epitaxial layer
[0108] 608b: Middle layer
[0109] 608c: Doped epitaxial layer
[0110] 610: Contact etch stop layer (CESL)
[0111] 612: First inter-layer dielectric (ILD) layer
[0112] 614: Second inter-layer dielectric (ILD) layer
[0113] 616: Etch stop layer (ESL)
[0114] 618: Cut-metal-gate (CMG) components
[0115] 619: Gate spacer
[0116] 620: Source / drain contacts
[0117] 622: Silicide components
[0118] 624: Interconnect Layer
[0119] 626: Source / drain contact vias
[0120] 628: First Metal Wire
[0121] 630: Second metal wire
[0122] 640: Carrier
[0123] 642: Hard Mask Layer
[0124] 644: Impedance layer
[0125] 646: Bottom
[0126] 648: Middle layer
[0127] 650: Upper
[0128] 652: Groove opening
[0129] 654: Back guide hole
[0130] 656: Dielectric Liner
[0131] 658: Back guide hole
[0132] 660: Backside silicide components
[0133] 670: Backside interconnect layer
[0134] 672: Back metal wire
[0135] X: X-axis
[0136] Y: Y axis
[0137] Z: Z axis
[0138] -Z: -Z axis
[0139] AA: Section Line
[0140] BB: Cutting Line
[0141] CC: Cut Line
[0142] DL: Device Layer
[0143] CO: Contact layer
[0144] V0: Zeroth via layer
[0145] M0: Metal layer 0
[0146] V1: First guide hole layer
[0147] V2: Second guide hole layer
[0148] V3: The third guide hole layer
[0149] M1: First metal layer
[0150] M2: Second metal layer
[0151] M3: The third metal layer
[0152] MD: Source / Drain Contact
[0153] VG: Gate via
[0154] VD: Source / drain contact vias
[0155] PU-1, PU-2: Pull-up transistors
[0156] PD-1, PD-2: Pull-down transistor
[0157] PG-1, PG-2: Pass-gate transistors
[0158] SN: Storage Node
[0159] SNB: Complementary Storage Node
[0160] VDD: power supply voltage
[0161] W_BL: bit line
[0162] W_BLB: complementary bit line
[0163] W_WL: word line
[0164] R-PG: Read-side pass gate transistor
[0165] R-PD: Read-side pull-down transistor
[0166] R_WL: Read end word line
[0167] R_BL: Read end bit line
[0168] D: Width
[0169] H: Unit height
[0170] W: Width
[0171] W1: Width
[0172] W2: Width
[0173] W3: Width
[0174] W4: Width
[0175] W5: Width DETAILED DESCRIPTION
[0176] The following disclosure provides many embodiments or examples for implementing different elements of the subject matter provided. Specific examples of each element and its configuration are described below to briefly illustrate the embodiments of the present disclosure. Of course, the above are merely examples and are not intended to define the embodiments of the present disclosure. For example, if the description mentions that a first element is formed above a second element, it may include an embodiment in which the first and second elements are in direct contact, and it may also include an embodiment in which an additional element is formed between the first and second elements so that they are not in direct contact. In addition, the embodiments of the present disclosure may repeat reference numbers and / or letters in various examples. Such repetition is for the purpose of simplicity and clarity, rather than for indicating the relationship between the different embodiments and / or configurations discussed.
[0177] In addition, the embodiments of the present disclosure may repeat reference values and / or letters in various examples. Such repetition is for the purpose of simplicity and clarity, rather than to represent the relationship between the different embodiments and / or configurations discussed. In addition, in the subsequent disclosure, if the description mentions that the first element is formed on the second element, connected to and / or coupled to the second element, it may include an embodiment in which the first and second elements are directly in contact, and it may also include an embodiment in which an additional element is formed between the first and second elements so that they are not in direct contact. Furthermore, spatially relative words such as "under...", "below", "lower", "above", "higher" and the like may be used to facilitate the description of the relationship between one (or some) parts or components and another (or some) parts or components in the figure. Spatially relative words are used to include different orientations of devices in use or operation. In addition, when using words such as "approximately" and "approximately" to describe a number or a range of numbers, the meaning of the word is to include numbers that do not differ from the described number by more than ±10%, unless otherwise specified. For example, the word "about 5 nanometers" includes a size range from 4.5 nanometers to 5.5 nanometers.
[0178] The present disclosure relates generally to static random-access memories (SRAM) structures, and more particularly to multi-port SRAM cells. Two-port (2P) SRAM cells and corresponding layouts are provided according to various exemplary embodiments. Some variations of some embodiments are discussed. Some exemplary embodiments relate to, but are not limited to, multi-gate devices.
[0179] Multi-gate devices have been introduced to improve gate control, reduce OFF-state current, and reduce short-channel effects (SCEs) by increasing gate-channel coupling. One such multi-gate device that has been introduced is fin field-effect transistors (FinFETs). FinFETs are named after the fin-like structures that extend from a substrate in which the fins are formed and are used to form a field effect transistor (FET) channel. Another multi-gate device transistor that has been introduced in part to address the performance challenges associated with FinFETs is the gate-all-around (GAA) transistor. GAA transistors are named after the gate structure that can extend around a channel region (e.g., a nanosheet stack) to provide access to the channel on four sides. GAA transistors are compatible with conventional complementary metal-oxide-semiconductor (CMOS) processes, and the structure of the GAA transistor enables it to be scaled significantly while maintaining gate control and mitigating short-channel effects. The following disclosure will continue to illustrate various embodiments of the present disclosure with one or more GAA examples.
[0180] The details of the device structure of the present disclosure will be described in the accompanying drawings. The accompanying drawings summarize the features of several embodiments so that those skilled in the art can better understand the detailed description that follows. Those skilled in the art should understand that they can easily use the present disclosure as a basis for designing or modifying other processes and structures to achieve the same purposes and / or achieve the same advantages as the embodiments described herein. Those skilled in the art should also recognize that such equivalent constructions do not depart from the spirit and scope of the present disclosure, and that they can make various changes, substitutions and modifications without departing from the spirit and scope of the present disclosure.
[0181] Figure 1A and Figure 1B A perspective view and a top view of a portion of an integrated circuit (IC) device 10 (eg, an SRAM device) implemented using GAA transistors are illustrated, respectively. Figure 1A, the IC device 10 includes a substrate 12. The substrate 12 may include an elemental (single element) semiconductor, such as silicon, germanium and / or other suitable materials; a compound semiconductor, such as silicon carbide, gallium arsenide, gallium phosphide, indium phosphide, indium arsenide, indium antimonide and / or other suitable materials; an alloy semiconductor, such as SiGe, GaAsP, AlInAs, AlGaAs, GaInAs, GaInP, GaInAsP and / or other suitable materials. The base 12 may be a single layer of material having a uniform composition. Alternatively, the substrate 12 may include multiple material layers having similar or different compositions suitable for IC device manufacturing. In one example, the substrate 12 may be a silicon-on-insulator (SOI) substrate having a semiconductor silicon layer formed on a silicon oxide layer. In another example, the substrate 12 may include a conductive layer, a semiconductor layer, a dielectric layer, other layers, or a combination thereof. Various doped regions such as source / drain (S / D) regions may be formed in or on the substrate 12. According to design requirements, the doped region may be doped with an N-type dopant such as phosphorus or arsenic and / or a P-type dopant such as boron. The doped region may be formed directly above the substrate 12 in the form of a P-type well structure, in the form of an N-type well structure, in the form of a double well structure, or in the form of a raised structure. The doped region may be formed by implantation of dopant atoms, in-situ doped epitaxial growth, and / or other suitable techniques.
[0182] A three-dimensional active region 14 is formed on the substrate 12. The active region of a transistor refers to the region formed by the source region, the drain region, and the channel region below the gate structure of the transistor. The active region is also referred to as an "oxide-definition (OD) region" in the above text. Each active region 14 includes an elongated nanostructure 26 (e.g., Figure 2 As shown in FIG. 1 , the elongated nanostructures 26 are vertically stacked in the channel region defined in the active region and above the fin base. The fin base protrudes upwardly out of the substrate 12. The source / drain features 16 are formed in the source / drain regions defined in the active region and above the fin base. The source / drain features 16 abut two opposite ends of the nanostructures 26. The source / drain features 16 may include an epitaxial layer epitaxially grown on the fin base.
[0183] The IC device 10 also includes an isolation structure (or isolation component) 18 formed above the substrate 12. The isolation structure 18 electrically isolates the various components of the IC device 10. The isolation structure 18 may include silicon oxide, silicon nitride, silicon oxynitride, fluoride-doped silicate glass (FSG), low-k dielectric materials, and / or other suitable materials. In some embodiments, the isolation structure 18 may include a shallow trench isolation (STI) component. In one embodiment, the isolation structure 18 is formed by etching a trench in the substrate 12 during the formation of the active region 14. The trench may then be filled with the above-mentioned isolation material, followed by a chemical mechanical planarization (CMP) process. Other isolation structures such as field oxide, local oxidation of silicon (LOCOS), and / or other suitable structures may also be implemented as the isolation structure 18. Alternatively, the isolation structure 18 may include a multi-layer structure, for example, having one or more layers of thermal oxide liner.
[0184] The IC device 10 also includes a gate structure (or gate stack) 20 formed on the active region 14 and engaging the active region 14. The gate structures 20 can be dummy gate structures (e.g., including an oxide gate dielectric and a polysilicon gate electrode), or they can be high-k metal gate (HKMG) structures including a high-k gate dielectric and a metal gate electrode, wherein the HKMG structure is formed by replacing the dummy gate structure. Although not illustrated herein, the gate structure 20 can include additional material layers, such as an interface layer, a cap layer, other suitable layers, or combinations thereof.
[0185] refer to Figure 1B , a plurality of active regions 14 are longitudinally oriented along the X direction, and a plurality of gate structures 20 are longitudinally oriented along the Y direction, i.e., substantially perpendicular to the active regions 14. Transistors are formed at the intersections of the active regions 14 and the gate structures 20. In many embodiments, the IC 10 includes additional features, such as gate spacers disposed along the sidewalls of the gate structures 20, as well as many other features.
[0186] Figure 2 It is along Figure 1A A partial schematic cross-sectional view of line AA of FIG. 1 illustrates various layers (levels) that can be manufactured on substrate 12 according to various aspects of the present disclosure. Figure 2In the embodiment, each layer includes a device layer (device layer, DL) and a multilayer interconnect (multilayer interconnect, MLI) disposed above the device layer DL. The device layer DL includes devices (e.g., transistors, resistors, capacitors and / or inductors) and / or device components (e.g., doped wells, gate structures, and / or source / drain components). In some embodiments, the device layer DL includes a substrate 12, a doped region 15 (e.g., an N-type well and / or a P-type well) disposed in the substrate 12, an isolation component 18, and a transistor T. In the depicted embodiment, the transistor T includes a suspended nanostructure (channel layer) 26 disposed between the source / drain components 16 and a gate structure 20, wherein the gate structure 20 surrounds and / or surrounds the suspended nanostructure 26. The nanostructure 26 may include a nanosheet, a nanotube, or a nanowire, or some other type of nanostructure extending horizontally along the X direction. Each gate structure 20 has a metal gate structure formed by a gate electrode 22 disposed above a gate dielectric 24 and a gate spacer 25 disposed along the sidewall of the metal gate structure.
[0187] The multi-layer interconnect MLI electrically couples various devices and / or elements of the device layer DL so that the various devices and / or elements can operate in a manner specified by the design requirements of the memory. In the described embodiment, the multi-layer interconnect MLI includes a contact layer (C0 layer), a zeroth via layer (V0 level), a zeroth metal (M0) layer, a first via layer (V1 level), a first metal layer (M1 level), a second via layer (V2 level), a second metal layer (M2 level), a third via layer (V3 level), and a third metal layer (M3 level). The present disclosure contemplates a multi-layer interconnect MLI with more or fewer layers and / or layers, for example, the total number of metal layers (layers) of the multi-layer interconnect MLI is N, where N is an integer ranging from 2 to 10. Each layer of the multi-layer interconnect MLI includes conductive features (e.g., metal lines, metal vias, and / or metal contacts) disposed in one or more dielectric layers (e.g., an inter-layer dielectric (ILD) layer and a contact etch stop layer (CESL)). In some embodiments, the conductive features at the same layer (e.g., M1 layer) of the multilayer interconnect MLI are formed simultaneously. In some embodiments, the conductive features at the same layer of the multilayer interconnect MLI have top surfaces that are substantially flat to each other and / or bottom surfaces that are substantially flat to each other. The CO layer includes a source / drain contact (MD) disposed in the dielectric layer 28; the V0 layer includes a gate via VG, a source / drain contact via VD, and a butted contact (MD) disposed in the dielectric layer 28. contacts); the M0 layer includes an M0 metal line disposed in the dielectric layer 28, wherein the gate via VG connects the gate structure to the M0 metal line, the source / drain via V0 connects the source / drain to the M0 metal line, and the butting contact connects the gate structure and the source / drain together and to the M0 metal line; the V1 layer includes a V1 via disposed in the dielectric layer 28, wherein the V1 via connects the M0 metal line to the M1 metal line; the M1 layer includes an M1 metal line disposed in the dielectric layer 28; the V2 layer includes a V2 via disposed in the dielectric layer 28, wherein the V2 via connects the M1 line to the M2 line; the M2 layer includes an M2 metal line disposed in the dielectric layer 28; the V3 layer includes a V3 via disposed in the dielectric layer 28, wherein the V3 via connects the M2 line to the M3 line. For clarity, Figure 2 It has been simplified to better understand the novel concepts of the present disclosure. Additional components may be added in various layers of the memory, and some of the components described may be replaced, modified or deleted in other embodiments of the memory. Figure 2 This is merely an example and may not reflect an actual cross-sectional view of the IC device 10 and / or the SRAM cell 100 discussed in further detail below.
[0188] Reference now Figure 3, which illustrates an exemplary circuit schematic diagram of a two-terminal SRAM cell 100. The two-terminal SRAM cell 100 includes a write terminal 100W and a read terminal 100R. The write terminal 100W includes pull-up transistors PU-1, PU-2, pull-down transistors PD-1, PD-2, and pass gate transistors PG-1, PG-2. In the illustrated embodiment, transistors PU-1 and PU-2 are P-type transistors, and transistors PG-1, PG-2, PD-1, and PD-2 are N-type transistors.
[0189] The drains of the pull-up transistor PU-1 and the pull-down transistor PD-1 are coupled together, and the drains of the pull-up transistor PU-2 and the pull-down transistor PD-2 are coupled together. The transistors PU-1 and PD-1 are cross-coupled with the transistors PU-2 and PD-2 to form a data latch. The gates of the transistors PU-2 and PD-2 are coupled together and coupled to the common drain of the transistors PU-1 and PD-1 to form a storage node SN, and the gates of the transistors PU-1 and PD-1 and the transistors PU-2 and PD-2 are coupled together and coupled to the common drain of the transistors PU-2 and PD-2 to form a complementary storage node SNB. The sources of the pull-up transistors PU-1 and PU-2 are coupled to a power supply voltage VDD (also referred to as Vcc), and the sources of the pull-down transistors PD-1 and PD-2 are coupled to the power supply voltage VDD (also referred to as Vcc), and in some embodiments, the voltage Vss can be electrically grounded.
[0190] The storage node SN of the data latch is coupled to the bit line W_BL of the write terminal 100W through the pass gate transistor PG-1, and the complementary storage node SNB is coupled to the complementary bit line W_BLB of the write terminal 100W through the pass gate transistor PG-2. The storage node SN and the complementary storage node SNB are complementary nodes that are often at opposite logic levels (logic high or logic low). The gates of the pass gate transistors PG-1 and PG-2 are coupled to the word line W_WL of the write terminal 100W.
[0191] The read end 100R of the SRAM cell 100 includes a read end pass gate transistor (R-PG) and a read end pull-down transistor (R-PD). The gate of the read end pass gate transistor R-PG is coupled to the read end word line R_WL. The gate of the read end pull-down transistor R-PD is coupled to the storage node SN (or the gates of transistors PU-2 and PD-2). The read end transistors R-PG and R-PD are coupled between the read end bit line R_BL and Vss.
[0192] According to various aspects of the present disclosure, transistors R-PD and R-PG at the read end are implemented to have a wider channel width than all transistors at the write end. Even within the write end, transistors PU-1 and PU-2 are implemented to have a narrower channel width than transistors PD-1, PD-2, PG-1, and PG-2. In some embodiments, transistors R-PD, PD-2, and PU-2 can share a continuous gate structure, but their respective gate electrodes are implemented using different work function metals and / or using different gate dielectrics (e.g., doped and undoped gate dielectrics). Such a configuration helps to optimize the performance of the two-terminal SRAM cell 100, for example by increasing the read and / or write speed while reducing parasitic capacitance or leakage.
[0193] Figure 4 A simplified schematic top view of a two-terminal SRAM cell 100 is illustrated, which includes a write terminal 100W and a read terminal 100R. The write terminal 100W includes transistors PG-1, PG-2, PU-1, PU-2, PD-1, and PD-2. The read terminal 100R includes transistors R_PD and R_PG. For the sake of clarity and simplicity, Figure 4 The well regions, active regions, gate structures and gate cut features of these transistors are illustrated in FIG. Figure 4 Interconnect elements such as contacts, vias, and metal lines are omitted.
[0194] In the illustrated embodiment, the transistors in the SRAM cell 100 are implemented as gate-all-around FETs (GAA FETs). Although FinFETs can be used to implement SRAM devices, FinFETs may face problems as the size of SRAM devices continues to shrink. For example, SRAM devices implemented using FinFETs may be more prone to cell mismatch problems, which may adversely affect the stability of the SRAM devices. In order to solve the problems associated with FinFET devices, the present disclosure utilizes multi-channel devices such as GAA transistors to implement SRAM devices. Compared with FinFETs or planar transistors, GAA transistors can achieve more flexible channel scaling and lower standby leakage because they have better drain-induced-barrier-lowering (DIBL) and swing performance.
[0195] like Figure 4 As shown, the two-terminal SRAM cell 100 includes active regions 210, 211, 212, 213, and 214. The active regions 210-214 are each Figure 4In the illustrated embodiment, the active regions 210-214 may each include (or may be implemented as) the above-mentioned Figure 2 Nanostructure 26. Active regions 210-213 are components of the write end 100W, and active region 214 is a component of the read end 100R. In the illustrated embodiment, active regions 211, 212 belong to transistors PU-1 and PU-2 of a PMOS device, respectively. As such, active regions 211, 212 are formed over an N-type well 110B. Meanwhile, active region 210 belongs to transistors PD-1 and PG-1 of an NMOS device. As such, active region 210 is formed over a P-type well 110A (or a P-type substrate). Similarly, active regions 213, 214 belong to transistors PG-2, PD-2, R-PG, and R-PD of an NMOS device. As such, active regions 213, 214 are formed over a P-type well 110A (or a P-type substrate). The P-type well 110A and the N-type well 110B are staggered with each other along the Y direction in the SRAM array.
[0196] like Figure 4 As shown, the two-terminal SRAM cell 100 further includes gate structures 220, 221, 222, 223, and 224. The gate structures 220-224 are each Figure 4 Each gate structure 220-224 has a critical dimension (CD) or gate width (denoted as G), which is the width measured along the Y direction. The gate structures 220-224 may each include (or may be implemented as) the above-discussed Figure 2 . The gate structure 220, 222 and 223 are components of the write end 100W. The gate structure 224 is a component of the read end 100R. The gate structure 221 extends through the write end 100W and the read end 100R. In this way, a portion of the gate structure 221 is a component of the write end 100W, and another portion of the gate structure 221 is a component of the read end 100R. In other words, the gate structure 221 is shared by the transistor R_PD of the read end 100R and the transistors PU-2 and PD-2 of the write end 100W.
[0197] Different active regions in different transistors of the SRAM cell 200 can have different widths (e.g., dimensions measured in the Y direction) in order to optimize device performance. In more detail, the active region 210 of the PD-1 transistor and the PG-1 transistor has a width W1, the active region 211 of the PU-1 transistor has a width W2, the active region 212 of the PU-2 transistor has a width W3, the active region 213 of the PG-2 transistor and the PD-2 transistor has a width W4, and the active region 214 of the R-PG transistor and the R-PD transistor has a width W5. The widths W1-W5 are measured in the portions of the respective active regions below the gate structures 220-224. In other words, these portions of the active regions (the portions from which the widths W1-W5 are measured) are the channel regions of the transistors (e.g., the vertically stacked nanostructures of a GAA device). To optimize SRAM performance, width W5 is configured to be equal to or greater than W1-W4 to increase the read speed of the SRAM cell 100, and widths W2 and W3 are configured to be smaller than widths W1 and W4 to balance the speed between PMOS devices and NMOS devices. In addition, in some embodiments, widths W2 and W3 may be the same, and widths W1 and W4 may be the same. Alternatively, in some embodiments, widths W2 and W3 may be different, and widths W1 and W4 may be different to balance the asymmetric layout caused by the presence of the read end 100R on a single side of the write end 100W. In some alternative embodiments, width W2 may be greater than W3 but less than either of widths W1 and W4, and width W1 may be greater than W4.
[0198] In some embodiments, the ratio of width W5 to gate width G is between about 0.5:1 and about 5:1, the ratio of width W1 (or W4) to gate width G is between about 0.5:1 and about 5:1, and the ratio of width W2 (or W3) to gate width G is in the range of about 0.3:1 and about 3:1. These ranges are not arbitrarily selected, but are specifically configured to fully increase the speed of the read end 100R and balance the speed between the read end 100R and the write end 100W without increasing the footprint of the entire device or increasing the risk of electrical shorts, or causing excessive manufacturing difficulties. In addition, the use of GAA transistors allows for more flexible selection of the active region width; in comparison, the use of FinFET transistors limits the active region width to discrete values, particularly integer multiples of the width of a single fin.
[0199] Still refer to Figure 4, the two-terminal SRAM cell 100 further includes a plurality of gate cutting components extending longitudinally along the X direction, including dielectric components 130A, 130B, 130C, 130D, 130E (collectively referred to as dielectric components 130). In the illustrated embodiment, the dielectric component 130A is disposed between the active regions 210, 211 and is adjacent to the gate structure 220 and the gate structure 221. The dielectric component 130A divides the originally continuous gate structure line into two isolated segments corresponding to the gate structure. Similarly, the dielectric component 130B is disposed between the active regions 212, 213 and is adjacent to the gate structure 222 and the gate structure 223, and the dielectric component 130C is disposed between the active regions 223, 224 and is adjacent to the gate structure 223 and the gate structure 224. The dielectric components 130B and 130C divide the originally continuous gate structure line into three isolated sections corresponding to the gate structure 222, the gate structure 223 and the gate structure 224. The dielectric components 130D and 130E are arranged on the edge of the SRAM cell 100 to separate the gate structure from the adjacent SRAM cell. Each dielectric component 130 is formed by filling a corresponding cut-metal-gate (CMG) groove at the position of the dielectric component. The dielectric component 130 is also called a CMG component. In the illustrated embodiment, the dielectric components 130A and 130B are arranged on the interface between the P-type well 100A and the N-type well 110B. The dielectric components 130C, 130D, and 130E are completely arranged in the P-type well 110A.
[0200] The CMG process refers to a manufacturing process in which a metal gate (e.g., a high-k metal gate or HKMG) is cut (e.g., by an etching process) to divide the metal gate into two or more gate segments after a dummy gate structure (e.g., a polysilicon gate) is replaced by a metal gate (e.g., a high-k metal gate or HKMG). Each gate segment serves as a metal gate for a single transistor. Isolation material is then filled into the trenches between adjacent portions of the metal gate. In the present disclosure, these trenches are referred to as cut metal gate trenches or CMG trenches. The dielectric material that fills the CMG trenches for isolation is referred to as a CMG component. In order to ensure that the metal gate is completely cut, the CMG component is typically extended further to adjacent areas, such as a dielectric layer that fills the space between the metal gates. The CMG component typically has an elongated shape in a top view.
[0201] Still refer to Figure 4 ,exist Figure 4The boundary 100B of the two-terminal SRAM cell 100 is illustrated using a dashed line. Note that some active regions and gate structures may extend beyond the illustrated boundary 100B, as these active regions and gate structures may also form elements of other adjacently positioned SRAM cells. The boundary 100B is longer in the Y direction than in the X direction, for example, about 3.5 times to about 6 times longer. In other words, the boundary 100B may be rectangular. The first dimension of the boundary 100B along the X direction is represented as the cell width W, and the second dimension of the boundary 100B along the Y direction is represented as the cell height H. When the two-terminal SRAM cell 100 is repeated in a memory array, the cell width W may represent and be referred to as the memory cell pitch along the X direction in the memory array, and the cell height H may represent and be referred to as the memory cell pitch along the Y direction in the memory.
[0202] Figure 5 The conductive features of the two-terminal SRAM cell 100 in the contact layer and the V0 layer are illustrated. For clarity, Figure 4 The active regions 210-214, gate structures 220-224 and cell boundary 100B shown in FIG. Figure 5 , and for simplicity, from Figure 5 Many other parts are omitted.
[0203] The gate contact 360A electrically connects the gate of the channel gate transistor PG-1 (formed by the gate structure 220) to the write-end word line W_WL, and the gate contact 360L electrically connects the gate of the channel gate transistor PG-2 (formed by the gate structure 223) to the write-end word line W_WL. The source / drain contact 360K electrically connects the drain region of the pull-down transistor PD-1 (formed on the active region 210 (which may include an N-type epitaxial source / drain component)) and the drain region of the pull-up transistor PU-1 (formed on the active region 211 (which may include a P-type epitaxial source / drain component)), so that the common drain of the pull-down transistor PD-1 and the pull-up transistor PU-1 forms a storage node SN. The gate contact 360B electrically connects the gate of the pull-up transistor PU-2 (formed by the gate structure 221) and the gate of the pull-down transistor PD-2 (also formed by the gate structure 221) to the storage node SN. Gate contact 360B may be a butt contact adjacent to source / drain contact 360K. Source / drain contact 360C electrically connects the drain region of pull-down transistor PD-2 (formed on active region 213 (which may include N-type epitaxial source / drain components)) and the drain region of pull-up transistor PU-2 (formed on active region 212 (which may include P-type epitaxial source / drain components)) such that the common drain of pull-down transistor PD-2 and pull-up transistor PU-2 form a complementary storage node SNB. Gate contact 360D electrically connects the gate of pull-up transistor PU-1 (formed by gate structure 222) and the gate of pull-down transistor PD-1 (also formed by gate structure 222) to complementary storage node SNB. Gate contact 360D may be a butt contact adjacent to source / drain contact 360C. Gate contact 360P electrically connects the gate of read-side pass-gate transistor R-PG (formed by gate structure 224) to the read-side word line R_WL.
[0204] The source / drain contact 360E and the source / drain contact via 370E landing thereon electrically connect the source region of the pull-up transistor PU-1 (formed on the active region 211 (which may include a P-type epitaxial source / drain component)) to the power supply voltage VDD. The source / drain contact 360F and the source / drain contact via 370F landing thereon electrically connect the source region of the pull-up transistor PU-2 (formed on the active region 212 (which may include a P-type epitaxial source / drain component)) to the power supply voltage VDD. The source / drain contact 360G and the source / drain contact via 370G landing thereon electrically connect the source region of the pull-down transistor PD-1 (formed on the active region 210 (which may include an N-type epitaxial source / drain component)) to the electrical ground VSS. The source / drain contact 360H and the source / drain contact via 370H thereon electrically connect the source region of the pull-down transistor PD-2 (formed on the active region 213 (which may include the N-type epitaxial source / drain component)) and the source region of the read-side pull-down transistor R-PD (formed on the active region 214 (which may include the N-type epitaxial source / drain component)) to the electrical ground VSS. The source / drain contact 360G and the source / drain contact 360H may be device-level contacts shared by adjacent SRAM cells 100 (e.g., four SRAM bit cells 100 adjacent at the same corner may share one source / drain contact 360H). The source / drain contact 360I and the source / drain contact via 370I thereon electrically connect the source region of the channel gate transistor PG-1 (formed on the active region 210 (which may include an N-type epitaxial source / drain component)) to the write-end bit line W_BL. The source / drain contact 360J and the source / drain contact via 370J thereon electrically connect the source region of the channel gate transistor PG-2 (formed on the active region 213 (which may include an N-type epitaxial source / drain component)) to the write-end complementary bit line W_BLB. The dummy contact 360M is formed above the drain region of the read-end pull-down transistor R-PD (formed on the active region 214 (which may include an N-type epitaxial source / drain component)). The dummy contact 360M is electrically floating. The source / drain contact 360N and the source / drain contact via 370N thereon electrically connect the source region of the read-side pass-gate transistor R-PG to the read-side bit line R_BL.
[0205] To further illustrate the various aspects of the present disclosure discussed above, Figure 6 , Figure 7 and Figure 8 Various cross-sectional views of a two-terminal SRAM cell 100 are illustrated in FIG. Figure 4 and Figure 5The cross-sectional views are taken at different cross sections of the top view of the two-terminal SRAM cell 100. These cross-sectional views are taken at different cross sections of the two-terminal SRAM cell 100. In more detail, Figure 6 Corresponding to the Figure 4 and Figure 5 Cross-sectional view taken along line AA. Figure 7 Corresponding to the Figure 4 and Figure 5 Cross-sectional view taken along line BB. Figure 8 Corresponding to the Figure 4 and Figure 5 Cross-sectional view taken along line CC.
[0206] Reference now Figure 6 , the two-terminal SRAM cell 100 includes a substrate, a P-type well 110A and an N-type well 110B formed above the substrate 110 (e.g., by doping different portions of the substrate 110), and active regions 210-214 formed above the P-type well 110A and the N-type well 110B. The lower sections (e.g., fin structures) of the active regions 210-214 are separated from each other in the Y direction by an isolation structure 112 (e.g., shallow trench isolation (STI)). The upper portions of the active regions 210-214 are respectively implemented as nanostructures 210A-210C, 211A-211C, 212A-212C, 213A-213C, and 214A-214C, such as nanosheets, nanotubes, nanowires, etc. These nanostructures act as vertically stacked conductive channels for their respective transistors. For example, nanostructures 210A-210C serve as the channel of transistor PD-1, nanostructures 211A-211C serve as the channel of transistor PU-1, nanostructures 212A-212C serve as the channel of transistor PU-2, nanostructures 213A-213C serve as the channel of transistor PG-2, and nanostructures 214A-214C serve as the channel of transistor R_PG. Note that Figure 6 Each vertical stack of gate structures is shown to have three nanostructures (e.g., 210A disposed above 210B, 210B disposed above 210C), but this is only for simplicity. In other embodiments, each vertical stack of nanostructures can include a different number of nanostructures, such as two, four, or five.
[0207] The nanostructures 210A-210C of transistor PD-1 each have a channel width W1, the nanostructures 211A-211C of transistor PU-1 each have a channel width W2, the nanostructures 212A-212C of transistor PU-2 each have a channel width W3, the nanostructures 213A-213C of transistor PG-2 each have a channel width W4, and the nanostructures 214A-214C of transistor R_PG each have a channel width W5. These channel widths W1-W5 are each measured in the Y direction. As described above with reference to Figure 4 As discussed, to optimize the performance of the two-terminal SRAM cell 100, the channel widths W1 and W4 are greater than the channel widths W2 and W3, for example, by about 1.5 times and about 5 times. In contrast, the channel widths W1 and W4 are approximately equal to each other, and the channel widths W2 and W3 are approximately equal to each other. In addition, the channel width W5 can be greater than the channel widths W1 and W4, for example, by about 1.1 times and about 3 times.
[0208] Nanostructures 210A-210C are circumferentially surrounded by gate structure 222A (as a first portion of gate structure 222 of transistor PD-1), and nanostructures 211A-211C and 212A-212C are circumferentially surrounded by gate structure 222B (as a second portion of gate structure 222 of transistor PU-1). At the same time, nanostructures 213A-213C are circumferentially surrounded by gate structure 223 of transistor PG-2, and nanostructures 214A-214C are circumferentially surrounded by gate structure 224 of transistor R_PG. Gate structures 222, 223, and 224 are also isolated from each other by gate cut features 130. In the illustrated embodiment, in the channel region, gate cut features 130B, 130C, 130D can each extend downward through STI features 112 to ensure adequate isolation.
[0209] In some embodiments, both gate structure 222A and gate structure 223 have a first work function (WF) metal, gate structure 224 has a second work function metal different from the first work function metal, and gate structure 222B has a third work function metal different from the first work function metal and different from the second work function metal. In some embodiments, the first work function metal and the second work function metal each include TiN, TaN, WCN, TiAl, W or a combination thereof. However, the first work function metal has thicker TiN and / or thicker WCN than the second work function metal. Implementing different work function metals for different gate structures allows different threshold voltages (Vt) to be adjusted for these different gate structures. This is conducive to further optimization of SRAM performance because it allows the write speed of the write end 100W to be increased while reducing (or at least not increasing) the leakage of the read end 100R.
[0210] The gate contacts 360L and the gate contacts 360P are formed over the gate structures 223 and 224, respectively. The butt contact 360D is formed over the gate structure 222B. The gate contacts 360L, 360P, and the butt contact 360D help provide electrical connectivity between the gate structures 222-224 and other components of the two-terminal SRAM cell 100 and / or with external devices. The gate contacts 360L, 360P, and the butt contact 360D are electrically and physically isolated from each other by an interlayer dielectric (ILD) 410 formed over the gate structures 222-224. Note that in some embodiments, such as the embodiments shown herein, a gate top dielectric layer 420 may be formed over the gate structures 222-224. In this case, the ILD 410 may be formed over the gate top dielectric layer 420.
[0211] Now refer to Figure 7 The cross-sectional view shown in this article (corresponding to the CC line) is the same as the above Figure 6 The cross-sections of have many similarities. Therefore, Figure 6 and Figure 7 Similar components appearing in the drawings will be marked as the same, and their detailed discussion will not be repeated hereinafter. Figure 6 different, Figure 7The CC line is taken from outside the gate structure and does not intersect the nanostructure (i.e., the channel of the GAA transistor). Instead, the CC line intersects with multiple source / drain components 210E-214E epitaxially grown on the active regions 210-214, respectively. The source / drain components 210E, 213E, and 215E are N-type source / drain components, while the source / drain components 211E and 212E are P-type source / drain components. The source / drain components 210E-215E are also electrically and physically isolated from each other by the ILD 410. The source / drain contact 360K is formed above the source / drain components 210E and 211E and electrically couples the source / drain components 210E and 211E together. Source / drain contact 360C is formed above source / drain components 212E and 213E, and electrically couples source / drain components 212E and 213E together. Dummy source / drain contact 360M is formed above source / drain component 214E. In the illustrated embodiment, outside the gate structure, gate cut components 130 (e.g., 130A, 130B, 130C, 130D) can each extend downward to below the top surface of STI component 112, but do not extend through STI component 112. This is mainly due to the etching rate of the source / drain region being less than the etching rate of the channel region during the CMG trench formation process. Therefore, each gate cut component 130 has a larger depth in the channel region and a shallower depth outside the gate structure.
[0212] Reference now Figure 8 , the gate structure 221 (of transistor PD-2) and the gate structure 223 (of transistor PG-2) are illustrated as being formed over the P-type well 110A. The gate structure 221 includes a plurality of gate electrode features 441 and a plurality of gate dielectric features 451. The gate structure 223 includes a plurality of gate electrode features 443 and a plurality of gate dielectric features 453. The gate dielectric features 451 surround or wrap around the respective nanostructures 213A-213C of the active region 213. The gate dielectric features 453 surround or wrap around the respective nanostructures 213A-213C of the active region 213. The gate electrode features 441 and 443 each have a first work function metal. The source / drain features 213E are disposed on opposite sides of the gate structure 221 and on opposite sides of the gate structure 223. The gate spacers 470 are also formed on the sidewalls of the portions of the gate structures 221 and 223 located over the upper surface of the source / drain features 213. The inner spacer 480 is formed on the portions of the gate structures 221 and 223 below the upper surface of the source / drain feature 213E. The gate top dielectric layer 420 is formed over the gate structures 221 and 223. The ILD 410 is formed over the gate top dielectric layer 420.
[0213] Source / drain contacts 360H, 360C, and 360J are formed on the source / drain feature 213E. Source / drain contact vias 370J electrically connect the source / drain contact 369J to the write-end complementary bit line (W_BLB). Source / drain contacts 360H, 360C, 360J and source / drain contact vias 370J are formed on the front side of the semiconductor device, which are also referred to as front source / drain contacts and front vias. Figure 8 Also depicted are backside vias or backside contacts 360HB. The backside contacts 360HB electrically connect the backside of the respective source / drain features 213E to electrical ground (VSS).
[0214] Fig. 9 2 illustrates a top view (front view) of an SRAM bit cell array 300 according to the present disclosure. Fig. 9 , a plurality of SRAM bit cells 100-1, 100-2, 100-3 and 100-4 are arranged along the X direction and the Y direction to form a 2×2 array of SRAM bit cells. Each SRAM bit cell in the array can be used as Figure 4 and Figure 5 In some embodiments, two adjacent SRAM bit cells in the X direction are arranged relative to a common boundary (defined by Fig. 9 The SRAM cells 100-2 and 100-3 are symmetrical about the X-axis (indicated by the dotted line in the figure), and two adjacent SRAM cells in the Y direction are symmetrical about the common boundary line therebetween. That is, the SRAM cell 100-2 is a duplicate cell of the SRAM cell 100-1, but flipped on the X-axis; the SRAM cell 100-3 is a duplicate cell of the SRAM cell 100-1, but flipped on the Y-axis; and the SRAM cell 100-4 is a duplicate cell of the SRAM cell 100-2, but flipped on the Y-axis.
[0215] For ease of reference, in this example, the row (column) direction is referred to as being in the X direction. Thus, SRAM bit cells 100-1, 100-3 are in the first row, and SRAM bit cells 100-2, 100-4 are in the second row. As described above, adjacent cells in the array are mirror images along a common boundary between adjacent cells. Some active regions in the SRAM cells may extend through multiple SRAM cells in a row. For example, the active regions of transistors PG-1 and PD-1 in SRAM cell 100-1 extend through the active regions of transistors PD-1 and PG-1 in SRAM cell 100-3. The active region of transistor PU-2 in SRAM cell 100-1 extends to the active region of transistor PU-2 in SRAM cell 100-3. The active regions of transistors PG-2 and PD-2 in SRAM cell 100-1 extend to the active regions of transistors PG-2 and PD-2 in SRAM cell 100-3. The active regions of transistors R-PG and R-PD in SRAM cell 100-1 extend to the active regions of transistors R-PG and R-PD in SRAM cell 100-3. The active regions in SRAM cells 100-2 and 100-4 are configured similarly to the active regions in SRAM cells 100-1 and 100-3.
[0216] Similarly, some gate structures can extend longitudinally across multiple SRAM cells without being interrupted by gate cutting features, while some gate structures are divided into multiple sections by gate cutting features (e.g., CMG features 130). Gate cutting features set at the boundaries of SRAM cells can also be shared by adjacent SRAM bit cells. In the illustrated embodiment, CMG features 130E extend longitudinally in the X direction and enter the corner regions of four SRAM cells 100-1, 100-2, 100-3, 100-4, and are shared by these four SRAM cells. CMG features 130A, 130B, 130C extend longitudinally in the X direction across the boundaries between adjacent SRAM cells in the same row and are shared by these two SRAM cells.
[0217] Contacts disposed at the boundaries of SRAM cells may also be shared by adjacent SRAM cells. In the illustrated embodiment, the source / drain contact 360H extends into the corner regions of the four SRAM cells 100-1, 100-2, 100-3, 100-4 and is shared by these four SRAM bit cells, thereby connecting the VSS nodes of the four SRAM cells together. Along the Y direction, the source / drain contact 360H spans four active regions, namely the active region of the transistor PD-2 in the SRAM cells 100-1, 100-3, the active region of the transistor R-PD in the SRAM cells 100-1, 100-3, the active region of the transistor R-PD in the SRAM cells 100-2, 100-4, and the active region of the transistor PD-2 in the SRAM cells 100-2, 100-4. Similarly, the source / drain contact 360G is also dedicated to the VSS node and is also in Fig. 9 The source / drain contact 360G may also extend across the boundary and be shared with four adjacent SRAM cells. The source / drain contact 360G may extend along the Y direction and span two active regions, both of which are used for transistor PD-1 in adjacent SRAM cells.
[0218] Fig.10 The bottom view (back view) of the SRAM bit cell array 300 according to the present disclosure is illustrated. The power supply voltage lines (VDD and VSS) and the signal lines do not have to be all formed on the front side of the integrated circuit structure, but can be distributed on the front side and the back side of the integrated circuit structure. For example, the integrated circuit structure may include a front interconnect structure and a back interconnect structure respectively arranged on the front side and the back side of the integrated circuit structure, and configured to connect various components of the pull-up device, the pull-down device and the channel gate device to form the SRAM cell 100. The design of the above configuration takes into account various factors and parameters, including the size of various conductive components, packaging density, resistance of conductive components, parasitic capacitance between adjacent conductive components, overlay shifting, and processing margins. For example, if the conductive components are too close, overlay shifting may cause short circuit and leakage problems; the size of the conductive components is reduced, resulting in increased resistance; parasitic capacitance is also increased; the processing margin is reduced; and so on. If the size of the conductive components is increased, the resistance of the conductive components is reduced, but the spacing between adjacent conductive components is reduced, resulting in increased parasitic capacitance and reduced processing margins. If shielding conductive members are placed between adjacent conductive members, parasitic capacitance is reduced. However, packaging density is reduced and / or the resistance of the conductive members is increased. In the depicted embodiment, ground lines VSS are formed on the front and back sides, and power lines VDD are formed on the front side.
[0219] Still refer to Fig.10 , the back source / drain contacts (or back vias) 360HB and 360GB, which are the counterparts of the front source / drain contacts 360H and 360G, are marked with numbers. Similar to the front source / drain contacts 360H and 360G, the back source / drain contacts 360HB and 360GB are electrically coupled to the electrical ground VSS. Unlike the front counterparts, each of the back source / drain contacts 360HB and 360GB has a width substantially the same as the width of the corresponding active region along the Y direction. This is due to an exemplary back manufacturing process in which the back source / drain contacts are formed by etching the fin structure in the active region from the back to form a back trench and filling the back trench with a conductive material. Therefore, the back source / drain contact inherits the width of the active region. In other words, the front source / drain contact 360G has a width greater than the width of the corresponding active region, while the back source / drain contact 360GB has a width substantially the same as the width of the corresponding active region. Similarly, the front source / drain contact 360H spans across four active regions, and there are four back source / drain contacts 360HB as their corresponding portions, each of which has a width substantially the same as the corresponding active region.
[0220] Fig.11 An alternative embodiment of a bottom view (back view) of an SRAM bit cell array 300 according to the present disclosure is illustrated. Fig.10 and Fig.11 One difference between the embodiments in FIG. 1 and FIG. 2 is that the two back source / drain contacts 360HB in the middle are merged together as Fig.11 A wider back source / drain contact 360HB is provided in the middle, while the other two back source / drain contacts 360HB on the sides remain unchanged. The middle back source / drain contact 360HB spans the two active regions of transistor R-PD and straddles CMG feature 130E. The left edge of the middle back source / drain contact 360HB is aligned with the left edge of the left one of the two active regions, and the right edge of the middle back source / drain contact 360HB is aligned with the right edge of the right one of the two active regions. Therefore, the width of the middle back source / drain contact 360HB is 2*W5+D, where W5 is the width of the active region of transistor R-PD and D is the distance between the two active regions. In comparison, the width of the other two back source / drain contacts 360HB is W4, which is the width of the active region of transistor PD-2, and the width of the back source / drain contact 360GB is W1, which is the width of the active region of transistor PD-1.
[0221] As the geometry continues to shrink, the width of the active region (Cd) becomes very small, resulting in vias with small openings and large aspect ratios. The small openings and large aspect ratios reduce the process windows for filling the conductive material, which may result in incomplete via formation and overlay errors. In addition, expanding the size of the backside via along the X-direction and / or the Y-direction may not be feasible because etching of the expanded via may also etch through the dielectric layer between the via and the gate structure and result in metal gate protrusion and device failure. On the other hand, regarding Fig.10 The back-side vias of the two middle back-side source / drain contacts 360HB in the embodiment of the present invention are formed, and the CMG component 130E in between protects the adjacent gate structure and allows the back-side vias to expand along the Y direction without causing further risk of metal gate protrusion. Therefore, the other two separated back-side vias can form a continuous back-side via that straddles the CMG component 130E. The resulting back-side via (middle back-side source / drain contact 360HB) has two leg portions that fill the two vias on both sides of the CMG component 130E and a center portion that expands above the CMG component 130E and connects the two leg portions. It can also be considered as a conductive structure including a first back-side via formed on the back side of the left active region, a second back-side via formed on the back side of the right active region, and a conductive component that connects the first and second back-side vias and straddles the CMG component 130E. The expanded opening and reduced via aspect ratio expand the process window, reduce the difficulty of filling conductive materials in high aspect ratio vias, and alleviate the inaccuracy of overlay.
[0222] Fig.12 1 illustrates a flow chart of a method 500 for manufacturing a semiconductor device (or apparatus) 600 according to various embodiments. The semiconductor device 600 is substantially similar to the method described in Figure 3-Figure 11 The integrated circuit structure including the SRAM cell 100 or as a part of the integrated circuit structure shown in various embodiments of the present disclosure. The present disclosure contemplates additional processes. Additional operations may be provided before, during, and after the method 500, and for additional embodiments of the method 500, some of the described operations may be moved, replaced, or deleted.
[0223] Combine the following Figures 13 to 21 To describe method 500, Figures 13 to 21 A cross-sectional view of a semiconductor device 600 is illustrated at various fabrication steps according to method 500 , in accordance with some embodiments. Figures 13 to 21 The cross-sectional view of the device 600 is along Fig.11 The AA line in FIG. 1 illustrates the formation of back-side vias and other components across the CMG components in the source / drain region, which is essentially similar to Fig.11 The middle back source / drain contact (or back via) 360HB in FIG. has been simplified for clarity. Figures 13 to 21 To better understand the novel concepts of the present disclosure. Additional components may be added to the device 600, and some components described below may be replaced, modified, or deleted in other embodiments of the device 600.
[0224] At operation 502, method 500 ( Fig.12 ) provides a device 600 having a substrate 602 on its back side and having various elements including transistors built on its front side. These elements include an isolation structure 604 above the substrate 602, two semiconductor fins (or multiple fins) 606 protruding upward from the substrate 602 (e.g., corresponding to Figure 7 The device 600 further includes a fin structure below the nanostructure in the corresponding active region of the fin 606 and adjacent to the isolation structure 604 and the source / drain feature 608 above the fin 606. The device 600 also includes a contact etch stop layer (CESL) 610 on the sidewalls of the source / drain feature 608 and above the isolation structure 604, a first inter-layer dielectric (ILD) layer 612 located above the CESL 610, and a second inter-layer dielectric layer 614 located above the first inter-layer dielectric layer 612. The first and second ILD layers can be separated by an etch stop layer (ESL) 616. Fig.13 600 is not illustrated, but in the direction of entering and exiting the paper, there are one or more nanostructures (or channel layers), the nanostructures are suspended above the fins 606 in the channel region of the transistor and connect the source / drain components 608 at both ends of a pair of channel layers, the gate structure (or gate stack, the gate structure surrounds the gate structure (or gate stack) of each nanostructure, and the inner spacer, the inner spacer is between the source / drain component 608 and the gate stack. The device 600 also includes a CMG component 618 (e.g., corresponding to the gate stack) that divides the gate stack into a plurality of segments. Fig.11 a CMG component 130E in the gate stack, a gate spacer 619 arranged on the sidewall of the gate stack and extending into the source / drain region to the sidewall of the buffer epitaxial layer 608a, a source / drain contact 620 arranged on the two source / drain components 608 and the CMG component 618, and a silicide component 622 inserted between the source / drain component 608 and the source / drain contact 620.
[0225] The device 600 also includes one or more interconnect layers, wherein the wires and vias are embedded in the dielectric layer. The one or more interconnect layers connect the gates, sources, and drains of various transistors and other circuits in the device 600 to partially or completely form an integrated circuit. The device 600 may also include a passivation layer, an adhesion layer, and / or other layers constructed on the front side of the device 600. These layers and the one or more interconnect layers are collectively indicated by the symbol 624. The interconnect layer 624 includes a source / drain contact via 626 located on the source / drain contact 620, a first metal line 628 in the M0 interconnect layer disposed on the source / drain contact via 626, and a second metal line 630 in the M1 interconnect layer disposed on the first metal line 628. The various elements of the device 600 are further described below.
[0226] In some embodiments, substrate 602 is a bulk silicon substrate (i.e., including bulk single crystal silicon). In various embodiments, substrate 602 may include other semiconductor materials, such as germanium, silicon carbide, gallium arsenide, gallium phosphide, indium phosphide, indium arsenide, indium antimonide, SiGe, GaAsP, AlInAs, AlGaAs, GaInAs, GaInP, GaInAsP, or combinations thereof. In alternative embodiments, substrate 602 is a semiconductor-on-insulator substrate, such as a silicon-on-insulator (SOI) substrate, a silicon germanium-on-insulator (SGOI) substrate, or a germanium-on-insulator (GOI) substrate.
[0227] In some embodiments, the fin 606 may include silicon, silicon germanium, germanium, or other suitable semiconductors, and may be doped with N-type or P-type dopants. The fin 606 may be patterned by any suitable method. For example, one or more optical lithography processes (including double patterning or multiple patterning processes) may be used to pattern the fin 606. In general, the double patterning or multiple patterning process combines the optical lithography process with a self-aligned process, thereby allowing the creation of a pattern having a pitch smaller than that obtainable using a single, direct optical lithography process, for example. For example, in one embodiment, a sacrificial layer is formed over a substrate and patterned using an optical lithography process. Spacers are formed along the patterned sacrificial layer using a self-aligned process. The sacrificial layer is then removed, and the remaining spacers or mandrels may then be used as mask elements for patterning the fin 606. For example, the mask element may be used to etch a groove in a semiconductor layer on or in the substrate 602, thereby leaving the fin 606 on the substrate 602. The etching process may include dry etching, wet etching, reactive ion etching (RIE), and / or other suitable processes. For example, the dry etching process may be performed with an oxygen-containing gas, a fluorine-containing gas (e.g., CF4, SF6, CH2F2, CHF3, and / or C2F6), a chlorine-containing gas (e.g., Cl2, CHCl3, CCl4, and / or BCl3), a bromine-containing gas (e.g., HBr and / or CHBr3), an iodine-containing gas, other suitable gases and / or plasmas, and / or combinations thereof. For example, the wet etching process may include etching in dilute hydrofluoric acid (dHF); a potassium hydroxide (KOH) solution; ammonia; a solution containing hydrofluoric acid (HF), nitric acid (HNO3), and / or acetic acid (CH3COOH); or other suitable wet etchants. Many other embodiments of the method of forming the fin 606 may be suitable.
[0228] The isolation structure 604 may include silicon oxide, silicon nitride, silicon oxynitride, other suitable isolation materials (e.g., including silicon, oxygen, nitrogen, carbon or other suitable isolation components), or a combination thereof. The isolation structure 604 may include different structures, such as shallow trench isolation (STI) components and / or deep trench isolation (DTI) components. In an embodiment, the trenches between the fins 606 may be filled with an insulator material (e.g., by using a chemical vapor deposition (CVD) process or a spin-on glass process), performing a chemical mechanical polishing (CMP) process to remove excess insulator material and / or planarize the top surface of the insulator material layer, and etching back the insulator material layer to form the isolation structure 604. In some embodiments, the isolation structure 604 includes multiple dielectric layers, such as a silicon nitride layer disposed on a thermal oxide liner layer.
[0229] The source / drain component 608 includes one or more layers of epitaxially grown semiconductor material, such as epitaxially grown silicon, germanium, or silicon germanium. In the depicted embodiment, the source / drain component 608 includes a buffer epitaxial layer 608a, an intermediate layer 608b, and a doped epitaxial layer 608c. For example, the epitaxial growth of the buffer epitaxial layer 608a can be performed by vapor-phase epitaxy (VPE), ultra-high vacuum chemical vapor deposition (UHV-CVD), molecular beam epitaxy (MBE), and / or other suitable processes. In some embodiments, the buffer epitaxial layer 608a includes the same material as the substrate 602, such as silicon (Si). In some alternative embodiments, the buffer epitaxial layer 608a includes a different semiconductor material than the substrate 602, such as silicon germanium (SiGe). In some embodiments, the buffer epitaxial layer 608a is undoped, where, for example, no intentional doping is performed during the epitaxial growth process. Alternatively, the buffer epitaxial layer 608a may be lightly doped with a dopant such as Ge or Sn. In comparison, in one case, the substrate 602 is lightly doped and has a higher doping concentration than the buffer epitaxial layer 608a. The buffer epitaxial layer 608a provides a high resistance path from the upper portion of the source / drain component to the semiconductor substrate, so that leakage current through the semiconductor substrate is suppressed.
[0230] The intermediate layer 608b can be conformally deposited on the device 600. In various examples, the intermediate layer 608b is a blanket deposited nitride layer by a chemical vapor deposition (CVD) process, a sub-atmospheric chemical vapor deposition (SACVD) process, a flowable chemical vapor deposition (FCVD) process, an atomic layer deposition (ALD) process, a physical vapor deposition (PVD) process, and / or other suitable processes. The intermediate layer 608b may also be referred to as a blanket nitride layer. In some embodiments, the blanket nitride layer 608b includes silicon nitride (SiN). Subsequently, a doped epitaxial layer 608c is formed on the blanket nitride layer 608b. After forming the doped epitaxial layer 608c, an etching process is performed to remove the portion of the blanket nitride layer 608b that is not stacked between the epitaxial layers 608a and 608c. The blanket nitride layer 608b acts as an etch stop layer in a subsequent backside etching process. For example, the doped epitaxial layer 608c can be formed by any epitaxial process, including VPE, UHV-CVD, MBE, and / or other suitable processes. The doped epitaxial layer 608c can be doped with N-type dopants and / or P-type dopants. In some embodiments, for an N-type transistor, the doped epitaxial layer 608c includes silicon and can be doped with carbon, phosphorus, arsenic, other N-type dopants, or combinations thereof (e.g., to form Si:C epitaxial source / drain components, Si:P epitaxial source / drain components, or Si:C:P source / drain components). In some embodiments, for a P-type transistor, the doped epitaxial layer 608c includes silicon germanium or germanium, and may be doped with boron, other P-type dopants, or a combination thereof (e.g., to form a Si:Ge:B epitaxial source / drain component). In one embodiment, the doped epitaxial layer 608c and the buffer epitaxial layer 608a are both formed of silicon (Si). In another embodiment, the doped epitaxial layer 608c is formed of silicon (Si) and the buffer epitaxial layer 608a is formed of silicon germanium (SiGe). The doped epitaxial layer 608c may also include multiple epitaxial semiconductor layers having different degrees of dopant density. In some embodiments, an annealing process (e.g., rapid thermal annealing (RTA) and / or laser annealing) is performed to activate the dopants in the doped epitaxial layer 608c.
[0231] In some embodiments, the channel layer (nanostructure, Fig.13606 ) includes a semiconductor material suitable for a transistor channel, such as silicon, silicon germanium, or other semiconductor materials. In various embodiments, the channel layer can be in the shape of rods, bars, sheets, or other shapes. In an embodiment, the channel layer is initially part of a semiconductor layer stack, which includes channel layers and other sacrificial semiconductor layers stacked alternately layer by layer. The sacrificial semiconductor layer and the channel layer include different material compositions (e.g., different semiconductor materials, different atomic composition percentages, and / or different weight composition percentages) to achieve etching selectivity. During the gate replacement process to form the gate stack, the sacrificial semiconductor layer is selectively removed, leaving the channel layer suspended above the fin 606 in the channel region.
[0232] In some embodiments, the gate stack includes a gate dielectric layer and a gate electrode layer. The gate dielectric layer may include a high dielectric constant (high-k) dielectric material, such as HfO2, or other suitable high dielectric constant (high-k) dielectric materials. High dielectric constant (high-k) dielectric materials generally refer to dielectric materials with high dielectric constants, such as dielectric materials greater than silicon oxide (k≈3.9). The gate dielectric layer can be formed by chemical oxidation, thermal oxidation, ALD, CVD and / or other suitable methods. In some embodiments, the gate stack also includes an interface layer between the gate dielectric layer and the channel layer. The interface layer may include silicon dioxide, silicon oxynitride or other suitable materials. In some embodiments, the gate electrode layer includes an N-type or P-type work function layer and a metal fill layer. For example, the N-type work function layer may include a metal with a sufficiently low effective work function, such as titanium, aluminum, tantalum carbide, tantalum carbonitride, tantalum silicon nitride or a combination thereof. For example, the P-type work function layer may include a metal having a sufficiently large effective work function, such as titanium nitride, tantalum nitride, ruthenium, molybdenum, tungsten, platinum or a combination thereof. For example, the metal filling layer may include aluminum, tungsten, cobalt, copper and / or other suitable materials. The gate electrode layer may be formed by CVD, PVD, electroplating and / or other suitable processes. Since the gate stack includes a high dielectric constant (high-k) dielectric layer and a metal layer, it is also referred to as a high dielectric constant (high-k) metal gate.
[0233] In some embodiments, the gate spacer 619 includes a dielectric material, such as a dielectric material including silicon, oxygen, carbon, nitrogen, other suitable materials or combinations thereof (e.g., silicon oxide, silicon nitride, silicon oxynitride (SiON), silicon carbide, silicon carbon nitride (SiCN), silicon oxycarbide (SiOC), silicon carbon nitride (SiOCN)). In some embodiments, the gate spacer 619 may include La2O3, Al2O3, SiOCN, SiOC, SiCN, SiO2, SiC, ZnO, ZrN, Zr2Al3O9, TiO2, TaO2, ZrO2, HfO2, Si3N4, Y2O3, AlON, TaCN, ZrSi or other suitable materials. For example, a dielectric layer including silicon and nitrogen, such as a silicon nitride layer, may be deposited on a dummy gate stack (which is subsequently replaced by a high dielectric constant (high-k) metal gate) and subsequently etched (e.g., anisotropically etched) to form the gate spacer 619. The gate spacer 619 is deposited on the sidewalls of the fin 606 not covered by the dummy gate stack, such as in the source / drain region. After the fin 606 is recessed in the source / drain region and the source / drain component 608 is epitaxially grown on the remaining portion of the fin 606, the gate spacer 619 remains on the sidewalls of the bottom portion of the source / drain component 608, such as on the sidewalls of the buffer epitaxial layer 608a. In some embodiments, the gate spacer 619 includes a multi-layer structure, such as a first dielectric layer including silicon nitride and a second dielectric layer including silicon oxide.
[0234] In some embodiments, the CESL 610 may include La2O3, Al2O3, SiOCN, SiOC, SiCN, SiO2, SiC, ZnO, ZrN, Zr2Al3O9, TiO2, TaO2, ZrO2, HfO2, Si3N4, Y2O3, AlON, TaCN, ZrSi or other suitable materials; and may be formed by CVD, PVD, ALD or other suitable methods. The ILD layer 612 and / or the ILD layer 614 may include tetraethylorthosilicate (TEOS) oxide, undoped silicate glass or doped silicon oxide such as borophosphosilicate glass (BPSG), fused silica glass (FSG), fused silica glass (PSG), borondoped silicon glass (BSG), titanium nitride (TiN), low-k dielectric materials, other suitable dielectric materials or combinations thereof. The ILD layer 612 and / or the ILD layer 614 may be formed by PE-CVD (plasma-enhanced chemical vapor deposition, PECVD)), F-CVD (flowable chemical vapor deposition, FCVD) or other suitable methods.
[0235] In some embodiments, the silicide feature 622 may include titanium silicide (TiSi), nickel silicide (NiSi), tungsten silicide (WSi), nickel platinum silicide (NiPtSi), nickel platinum germanium silicide (NiPtGeSi), nickel germanium silicide (NiGeSi), ytterbium silicide (YbSi), platinum silicide (PtSi), iridium silicide (IrSi), erbium silicide (ErSi), cobalt silicide (CoSi), or other suitable compounds.
[0236] In an embodiment, the source / drain contact 620 may include a conductive barrier layer and a metal filling layer above the conductive barrier layer. The conductive barrier layer may include titanium (Ti), tantalum (Ta), tungsten (W), cobalt (Co), ruthenium (Ru) or a conductive nitride, such as titanium nitride (TiN), titanium aluminum nitride (TiAlN), tungsten nitride (WN), tantalum nitride (TaN), or a combination thereof, and may be formed by CVD, PVD, ALD and / or other suitable processes. The metal filling layer may include tungsten (W), cobalt (Co), molybdenum (Mo), ruthenium (Ru), nickel (Ni), copper (Cu) or other metals, and may be formed by CVD, PVD, ALD, electroplating or other suitable processes. In some embodiments, the conductive barrier layer is omitted in the source / drain contact 620.
[0237] In some embodiments, the CMG feature 618 may include one or more dielectric materials, such as silicon nitride, silicon oxide, silicon oxynitride, fluoride-doped silicate glass (FSG), low-k dielectric materials and / or other suitable insulating materials; and may be formed by CVD, PVD, ALD or other suitable methods. The CMG feature 618 fills the CMG trench and divides the gate structure into two segments in the channel region. The CMG feature 618 also extends into the source / drain region and is disposed between two adjacent source / drain features 608. During the formation of the source / drain contacts, the CMG feature 618 is recessed in the source / drain region to form an S / D recess. In this way, the top surface of the CMG feature 618 in the channel region is above the source / drain feature 608, but the top surface of the CMG feature 618 in the source / drain region is below the source / drain feature 608, for example, below the bottom surface of the doped epitaxial layer 608c, as shown in FIG. Fig.13 shown.
[0238] In some embodiments, the source / drain contact via 626 may include a conductive barrier layer and a metal filling layer above the conductive barrier layer. The conductive barrier layer may include titanium (Ti), tantalum (Ta), tungsten (W), cobalt (Co), ruthenium (Ru) or a conductive nitride, such as titanium nitride (TiN), titanium aluminum nitride (TiAlN), tungsten nitride (WN), tantalum nitride (TaN), or a combination thereof, and may be formed by CVD, PVD, ALD and / or other suitable processes. The metal filling layer may include tungsten (W), cobalt (Co), molybdenum (Mo), ruthenium (Ru), nickel (Ni), copper (Cu) or other metals, and may be formed by CVD, PVD, ALD, electroplating or other suitable processes. In some embodiments, the conductive barrier layer is omitted in the source / drain contact via 626.
[0239] At operation 504, method 500 ( Fig.12) Invert the device 600 and attach the front side of the device 600 to the carrier 640, as shown in FIG. Fig.14 As shown. This allows access to the device 600 from the back side of the device 600 for a one-step process. Operation 504 may use any suitable attachment process, such as direct bonding, hybrid bonding, using an adhesive, or other bonding methods. Operation 504 may also include alignment, annealing, and / or other processes. In some embodiments, the carrier 640 may be a silicon wafer. Figure 13-Figure 21 In the figure, the “Z” direction points from the back of the device 600 to the front of the device 600 , while the “−Z” direction points from the front of the device 600 to the back of the device 600 .
[0240] At operation 506, method 500 ( Fig.12 ) Thinning down the device 600 from the back side of the device 600 until the fin 606 and the isolation structure 604 are exposed from the back side of the device 600, such as Fig.15 As shown. The thinning process may include a mechanical grinding process and / or a chemical thinning process. During the mechanical grinding process, a large amount of substrate material may first be removed from the substrate 602. Then, the chemical thinning process may apply an etching chemical to the back side of the substrate 602 to further thin the substrate 602.
[0241] At operation 508, method 500 ( Fig.12 ) A hard mask layer 642 is formed on the back side of the device 600, and a tri-layer resist layer 644 is formed on the hard mask layer 642. Fig.16 The hard mask layer 642 may include oxide (eg, SiO2), nitride (eg, SiN), oxynitride (eg, SiO X N Y ) etc. In other embodiments, the hard mask layer 642 is SiO2. In still other embodiments, the hard mask layer 642 is a high-temperature oxide (HTO) (e.g., SiO2 formed by a high temperature deposition / growth process). In some embodiments, the process for forming the hard mask layer 642 includes depositing a dielectric material on the back side of the device 600 by, for example, CVD, PVD, ALD, sputtering, some other deposition process, or a combination thereof.
[0242] The three-layer impedance layer 644 includes a bottom layer 646 above the hard mask layer 642, an intermediate layer 648 above the bottom layer 646, and an upper layer 650 above the intermediate layer 648. The bottom layer 646 may be a bottom anti-reflective coating (BARC). The bottom layer 646 may include an organic material. The intermediate layer 648 may be formed of or include an inorganic material, and the inorganic material may be a nitride (e.g., silicon nitride), an oxynitride (e.g., silicon oxynitride), an oxide (e.g., silicon oxide), etc. The upper layer 650 is a photosensitive material. The intermediate layer 648 has a higher etching selectivity relative to the upper layer 650 and the bottom layer 646. Therefore, the upper layer 650 is used as a patterned etching mask for the intermediate layer 648, and the intermediate layer 648 is used as a patterned etching mask for the bottom layer 646. In some embodiments, the impedance layer formed above the hard mask layer 642 may be another type of photoresist, such as a single-layer photoresist, a double-layer photoresist, or the like. The upper layer 650 is patterned using any suitable optical lithography technique to form a groove opening 652 therein. As an example of patterning the upper layer 650, the upper layer 650 can be exposed to a radiation beam including UV (ultra-violet) or an excimer laser, such as a 248 nm beam from a krypton fluoride (KrF) excimer laser, a 193 nm beam from an argon fluoride (ArF) excimer laser, or a 157 nm beam from an F2 excimer laser. Exposure of the photosensitive material can be performed using an immersion optical lithography system to improve resolution and reduce the minimum achievable pitch. A baking or curing operation can be performed to harden the upper layer 650, and a developer can be used to remove the exposed or unexposed portions of the upper layer 650 depending on whether a positive photoresist or a negative photoresist is used.
[0243] At operation 510, method 500 ( Fig.12 ) extending the trench opening 652 to the hard mask layer 642, and selectively etching the fin 606 by etching through the trench opening 652 to form a backside via 654, such as Fig.17. The pattern of the upper layer 650 is transferred to the middle layer 648 using a suitable etching process. In some embodiments, the middle layer 648 may be trimmed to increase the side of the trench opening 652 in the middle layer 648. After the trimming process, a suitable etching process is performed to transfer the pattern of the middle layer 648 to the bottom layer 646 so that the trench opening 652 extends through the bottom layer 646. In addition, the pattern of the bottom layer 646 is transferred to the hard mask layer 642 using a suitable etching process. In an embodiment, the etching process used to etch the bottom layer 646 continues to etch the hard mask layer 642. During the etching process, the upper layer 650, the middle layer 648, and the bottom layer 646 may be consumed. In some embodiments, an ashing process may be performed to remove the remaining residue of the bottom layer 646. After the pattern of the hard mask layer 642 exposes the back side of the device 600, operation 510 applies an etching process that is tuned to be selective to the material of the semiconductor material (e.g., Si) in the fin 606. In the present embodiment, the etching process also etches the buffer epitaxial layer 608a of the source / drain component 608. The blanket nitride layer 608b of the source / drain component 608 serves as an etch stop layer to protect the doped epitaxial layer 608c from being etched. The etching steps of the fin 606 and the buffer epitaxial layer 608a may include a first etch that selectively etches the fin 606 and a second etch that selectively etches the buffer epitaxial layer 608a. The etching process may be a dry etch, a wet etch, a reactive ion etch, or other suitable etching method. The etching process also consumes the horizontal portion of the isolation structure 604 covering the CMG component 618, so that the bottom surface of the CMG component 618 is exposed in the trench opening 652. Alternatively, the horizontal portions of the isolation structure 604 may be etched in a separate etch process that is selective to the material of the isolation structure 604, with no (or minimal) etch loss to the CMG features 618. The separate etch process may be an anisotropic etch process, such that the vertical portions of the isolation structure 604 disposed on the sidewalls of the CMG features 618 remain intact. The etch selectivity of the CMG features 618 protects the gate stack from being accidentally exposed in the trench opening 652 if over-etching occurs, because the edge of the trench opening 652 may be too close to the gate stack in the X direction during the etch process.
[0244] At operation 512, method 500 ( Fig.12 ) depositing a dielectric liner 656 on the sidewalls of the trench opening 652 (including the backside via 654), such as Fig.18As shown. When a conductive component is subsequently formed in the trench opening 652, the dielectric liner 656 further protects the gate stack from diffusion of metal elements. In the illustrated embodiment, the dielectric liner 656 is conformally deposited to have a substantially uniform thickness along the various surfaces of the back side of the device 600. In various embodiments, the dielectric liner 656 may include La2O3, Al2O3, SiOCN, SiOC, SiCN, SiO2, SiC, ZnO, ZrN, Zr2Al3O9, TiO2, TaO2, ZrO2, HfO2, Si3N4, Y2O3, AlON, TaCN, ZrSi, combinations thereof, or other suitable materials. The dielectric liner 656 may be deposited using ALD, CVD, or other suitable methods. Subsequently, an etching process is performed to penetrate and remove most of the horizontal portions of the dielectric liner 656. The etching process is also referred to as a breakthrough (BT) etching process. In some embodiments, the BT etching process may include an anisotropic dry etching process, etc. In some embodiments where the dielectric liner 656 is formed of oxide, the BT etching process is a reactive ion etching (RIE) process, wherein the etching process gas includes CHF3, Ar, CF4, N2, O2, CH2F2, SF3, etc. or a combination thereof. The reactive ion etching (RIE) can be performed at an etching time between about 2 seconds and about 20 seconds, a pressure between about 2 mTorr and about 30 mTorr, a temperature between about 10° C. and about 100° C., a radio frequency (RF) power between about 100 W and about 1500 W, and a voltage bias between about 10 V and about 800 V. In the illustrated embodiment, as a result of operation 512, a portion of the dielectric liner 656 remains on the sidewalls of the isolation structure 604. In the depicted embodiment, the dielectric liner 656 can also contact the gate spacer 619 and the blanket nitride layer 608b. Due to the loading effect, the vertical length of the dielectric liner 656 on the sidewall near the CMG feature 618 is less than the vertical length of the dielectric liner 656 on the other sidewall near the ILD layer 612. After the BT etching process, the blanket nitride layer 608b as an etch stop layer is exposed in the back-side via 654. Subsequently, an etching process is applied to remove the exposed portion of the blanket nitride layer 608b. The etching process can be dry etching, wet etching, reactive ion etching, or other etching methods. At the end of operation 512, the back-side via 654 exposes the bottom surface of the doped epitaxial layer 608c of the source / drain feature 260 from the back side. A portion of the blanket nitride layer 608b remains between the doped epitaxial layer 608c and the dielectric liner 656.
[0245] The trench opening 652 provides a much larger opening than the individual backside vias 654. As described above, the trench opening 652 has an opening width 2W5+D along the Y direction, which can be two to ten times the opening width W5 of the backside vias 654. In addition, the aspect ratio of the backside vias 654 is reduced to H2 / W1 instead of H1 / W1, which is about 30% to about 60% smaller. The enlarged opening and reduced aspect ratio of the backside vias expands the process window, reduces the difficulty of filling conductive materials in high aspect ratio vias, and alleviates overlay inaccuracy in the backside photolithography process.
[0246] At operation 514, method 500 ( Fig.12 ) A back-side via 658 is formed in the trench opening 652, such as Fig.19 and Fig. 20 In some embodiments, the backside via 658 is formed by filling the trench opening 652 (including the backside via 654) with one or more conductive materials, such as Fig.19 As shown, and then the excess conductive material is removed from the back side of the device 600 in a planarization process, as shown Fig. 20As shown. The leg portion of the back via 658 filling the back via 654 is seated on the two source / drain features 608, and the middle portion of the back via 658 contacts the CMG feature 618. The leg portion and the middle portion of the back via 658 form a continuous conductive structure that connects the two source / drain features 608 to the same power line (e.g., Vss). The back via 658 may include tungsten (W), cobalt (Co), molybdenum (Mo), ruthenium (Ru), copper (Cu), nickel (Ni), titanium (Ti), tantalum (Ta), titanium nitride (TiN), tantalum nitride (TaN), or other metals, and may be formed by CVD, PVD, ALD, electroplating, or other suitable processes. In the illustrated embodiment, operation 514 forms a back silicide feature 660 on the bottom surface of the doped epitaxial layer 608c of the source / drain feature 608. The back silicide feature 660 reduces the contact resistance between the source / drain feature 608 and the back side. In further embodiments, operation 514 first deposits one or more metals into the backside vias 654, performs an annealing process on the device 600 to cause a reaction between the one or more metals and the source / drain features 608 to produce silicide features, and removes unreacted portions of the one or more metals, leaving silicide features in the backside vias 654. The one or more metals may include titanium (Ti), tantalum (Ta), tungsten (W), nickel (Ni), platinum (Pt), ytterbium (Yb), iridium (Ir), erbium (Er), cobalt (Co), or combinations thereof (e.g., alloys of two or more metals) and may be deposited using CVD, PVD, ALD, or other suitable methods. The back silicide feature 660 may include titanium silicide (TiSi), nickel silicide (NiSi), tungsten silicide (WSi), nickel platinum silicide (NiPtSi), nickel platinum germanium silicide (NiPtGeSi), nickel germanium silicide (NiGeSi) ytterbium silicide (YbSi), platinum silicide (PtSi), iridium silicide (IrSi), erbium silicide (ErSi), cobalt silicide (CoSi), combinations thereof, or other suitable compounds. Alternatively, the back vias 658 may directly contact the two source / drain features 608. A planarization operation, such as a chemical mechanical polishing (CMP) process, may also be performed to remove excess conductive material and expose the back side of the hard mask layer 642, such as Fig. 20 shown.
[0247] At operation 516, method 500 ( Fig.12 ) forms one or more backside interconnect layers 670, wherein the backside power rails are embedded in a dielectric layer on the backside of the device 600, such as Fig.21As shown. In the illustrated embodiment, the back power rail includes a back metal line 672 (e.g., in the back M0 interconnect layer), which is located on the back via 658 and electrically couples the two source / drain components 608 to the ground line VSS. The back metal line 672 may include a conductive barrier layer and a metal fill layer above the conductive barrier layer. The conductive barrier layer may include titanium (Ti), tantalum (Ta), tungsten (W), cobalt (Co), ruthenium (Ru) or a conductive nitride, such as titanium nitride (TiN), titanium aluminum nitride (TiAlN), tungsten nitride (WN), tantalum nitride (TaN) or a combination thereof, and may be formed by CVD, PVD, ALD and / or other suitable processes. The metal fill layer may include tungsten (W), cobalt (Co), molybdenum (Mo), ruthenium (Ru), nickel (Ni), copper (Cu) or other metals, and may be formed by CVD, PVD, ALD, electroplating or other suitable processes. In some embodiments, the conductive barrier layer is omitted. The back power rail may also include other back metal lines that electrically couple other back vias to power lines VDD and / or signal lines. In an embodiment, the back power rail may be formed using a damascene process, a dual damascene process, a metal patterning process, or other suitable processes. Fig.21 6. Although not illustrated in FIG. 6, the back power rail may include multiple back interconnect layers including contacts, vias, wires, and / or other conductive features. Having a back power rail beneficially increases the number of metal tracks available in device 600 for direct connection to source / drain contacts and vias. The back power rail may have a wider dimension than the first metal layer (M0) track on the front side of device 600, which beneficially reduces the back power rail resistance.
[0248] In summary, the present disclosure relates to implementing a two-terminal SRAM device with a gate-all-around (GAA) transistor, and using different channel widths for different transistors. The present disclosure also configures a back-side via that crosses the gate and cuts the dielectric component, so that it can retain a relatively large back contact area between the source / drain component and the back power rail, effectively reducing the contact resistance and expanding the process window. In addition, the relatively large back contact area provides better overlay control between the back-side via and the source / drain component. In addition, the embodiments of the present disclosure can be easily integrated into existing semiconductor manufacturing processes.
[0249] In one exemplary aspect, the present disclosure relates to a memory cell. The memory cell includes: a first active region, which provides a plurality of first nanostructures for a write-end channel gate transistor, the first active region having a first width; a second active region, which provides a plurality of second nanostructures for a write-end pull-up transistor, the second active region having a second width, and a third active region, which provides a plurality of third nanostructures for a read-end pull-down transistor, the third active region having a third width, the third active region width being greater than the first width, and the first width being greater than the second width. In some embodiments, the ratio of the first width to the second width is in a range of about 1.5:1 to about 5:1. In some embodiments, the ratio of the third width to the first width is in a range of about 1.1:1 to about 3:1. In some embodiments, the memory cell further includes a gate structure line extending over the first, second, and third active regions, and a gate-cut dielectric feature, the gate-cut dielectric feature dividing the gate structure line into a first gate structure surrounding each first nanostructure and a second gate structure surrounding each second nanostructure and each third nanostructure. In some embodiments, the gate-cut dielectric feature extends vertically deeper in a region in the gate structure line than in a region outside the gate structure line. In some embodiments, the gate cut dielectric component is a first gate cut dielectric component, and the memory cell further includes a second gate cut dielectric component adjacent to the second gate structure. In some embodiments, the memory cell further includes a back contact across the second gate cut dielectric component. In some embodiments, the memory cell further includes a fourth active region providing a plurality of fourth nanostructures for a write-end pull-down transistor, the fourth active region having a fourth width, the third width being greater than the fourth width, and the fourth width being greater than the second width. In some embodiments, the fourth width is different from the first width. In some embodiments, the memory cell further includes a fourth active region providing a plurality of fourth nanostructures for a write-end pull-up transistor, the fourth active region having a fourth width, the fourth width being less than the first width, and the fourth width being different from the second width.
[0250] In another exemplary embodiment, the present disclosure relates to a semiconductor structure. The semiconductor structure includes a first active region and a second active region extending longitudinally along a first direction, a gate stack extending longitudinally along a second direction perpendicular to the first direction, a dielectric component disposed between the first active region and the second active region, the dielectric component dividing the gate stack into a first section above the first active region and a second section above the second active region, a first epitaxial component disposed in the first active region, a second epitaxial component disposed on the second active region, the first epitaxial component and the second epitaxial component disposed on two opposite sides of the dielectric component, and a back conductive component electrically coupled to the first epitaxial component and the second epitaxial component and spanning the dielectric component. In some embodiments, the dielectric component extends longitudinally along the first direction. In some embodiments, the back conductive component includes a first foot located on the bottom surface of the first epitaxial component, a second foot located on the bottom surface of the second epitaxial component, and a middle portion physically connecting the first foot and the second foot, the middle portion being located directly below the dielectric component. In some embodiments, the middle portion is physically in contact with the dielectric component. In some embodiments, the semiconductor structure also includes a front contact disposed on the first epitaxial component and the second epitaxial component, the dielectric component being vertically stacked between the front contact and the back conductive component. In some embodiments, the front contact is wider than the back conductive feature in the second direction. In some embodiments, the first segment and the second segment of the gate stack are configured to fit the first active region and the second active region to form the first transistor and the second transistor in the first read end and the second read end of the first memory cell and the second memory cell in the memory array, respectively.
[0251] In another exemplary embodiment, the present disclosure relates to a method. The method includes forming a first fin and a second fin protruding from the front of a substrate, forming a gate stack on the first fin and the second fin, forming a dielectric component, the dielectric component divides the gate stack into a first section that fits the first fin to form a first transistor and a second section that fits the second fin to form a second transistor, wherein the first transistor and the second transistor are read-end transistors, growing a first epitaxial component on the first fin and growing a second epitaxial component on the second fin, wherein the dielectric component is disposed between the first epitaxial component and the second epitaxial component, performing an etching process on the back side of the substrate to form a back side groove, the back side groove exposing the dielectric component and the first epitaxial component and the second epitaxial component, and forming a back side guide hole in the back side groove, the back side guide hole spanning the dielectric component and electrically connected to the first epitaxial component and the second epitaxial component. In some embodiments, the execution step of the etching process includes etching the first fin and the second fin from the back side of the substrate. In some embodiments, the first fin and the second fin are etched to form a first via and a second via extending from the back groove and respectively exposing the first epitaxial component and the second epitaxial component, the first via and the second via are arranged on opposite sides of the dielectric component, and the formation of the back via includes filling the first via and the second via and the back groove with a conductive material.
[0252] The above summarizes the components of several embodiments so that those with ordinary knowledge in the technical field to which the present disclosure belongs can more easily understand the concepts of the embodiments of the present disclosure. Those with ordinary knowledge in the technical field to which the present disclosure belongs should understand that they can design or modify other processes and structures based on the embodiments of the present disclosure to achieve the same purposes and / or advantages as the embodiments introduced herein. Those with ordinary knowledge in the technical field to which the present disclosure belongs should also understand that such equivalent processes and structures do not deviate from the spirit and scope of the present disclosure, and they can make various changes, substitutions and replacements without violating the spirit and scope of the present disclosure.
Claims
1. A memory cell, characterized in that: include: a first active region, the first active region providing a plurality of first nanostructures for a write-end channel gate transistor, the first active region having a first width; a second active region, the second active region providing a plurality of second nanostructures for a write-end pull-up transistor, the second active region having a second width; as well as a third active region, the third active region providing a plurality of third nanostructures for a read-end pull-down transistor, the third active region having a third width, The third width is greater than the first width, and the first width is greater than the second width.
2. The memory cell according to claim 1, wherein: A ratio of the first width to the second width is between about 1.5:1 and about 5:
1.
3. The memory unit according to claim 1 or 2, characterized in that A ratio of the third width to the first width is between about 1.1:1 and about 3:
1.
4. The memory cell according to claim 1, wherein: Also includes: A gate structure line extending across the first active region, the second active region and the third active region; and A gate cutting dielectric component is provided, wherein the gate cutting dielectric component divides the gate structure line into a first gate structure surrounding each of the first nanostructures and a second gate structure surrounding each of the second nanostructures and each of the third nanostructures.
5. The memory cell according to claim 4, wherein: The gate cutting dielectric feature extends vertically deeper in an inner region of the gate structure line than in an outer region of the gate structure line.
6. The memory cell according to claim 1, wherein: Also includes: a fourth active region, the fourth active region providing a plurality of fourth nanostructures for a write-end pull-up transistor, the fourth active region having a fourth width, The fourth width is smaller than the first width, and the fourth width is different from the second width.
7. A semiconductor structure, characterized in that: include: A first active region and a second active region, the first active region and the second active region longitudinally extending along a first direction; a gate stack extending longitudinally along a second direction perpendicular to the first direction; a dielectric component disposed between the first active region and the second active region, wherein the dielectric component divides the gate stack into a first section above the first active region and a second section above the second active region; a first epitaxial component, disposed on the first active region; a second epitaxial component disposed on the second active region, wherein the first epitaxial component and the second epitaxial component are disposed on opposite sides of the dielectric component; and A back conductive component is electrically coupled to the first epitaxial component and the second epitaxial component and straddles the dielectric component.
8. The semiconductor structure according to claim 7, wherein: The back conductive component includes a first foot located on a bottom surface of the first epitaxial component, a second foot located on a bottom surface of the second epitaxial component, and a middle portion physically connecting the first foot and the second foot, wherein the middle portion is located directly below the dielectric component.
9. The semiconductor structure according to claim 7, wherein: Also includes: A front contact is disposed on the first epitaxial component and the second epitaxial component, wherein the dielectric component is vertically stacked between the front contact and the back conductive component.
10. The semiconductor structure according to claim 7, wherein: The first section and the second section of the gate stack are configured to respectively match the first active region and the second active region to form a plurality of first transistors and a plurality of second transistors in a plurality of first read ends of a plurality of first memory cells and a plurality of second read ends of a plurality of second memory cells in a memory array.
