Semiconductor structure and method for forming the same

By forming a multi-layer stack of alternating SiGe and Si layers on the semiconductor substrate and selectively etching to form a high-k metal gate structure, the problem of WFM layer process complexity in GAA FETs is solved, and the NFET and PFET are realized Vt adjustment is compatible, improving manufacturing efficiency and performance.

CN113224057BActive Publication Date: 2024-11-08TAIWAN SEMICONDUCTOR MANUFACTURING CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202110443667.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2021-01-22
Filing Date
2021-04-23
Publication Date
2024-11-08
Estimated Expiration
2041-04-23

AI Technical Summary

Technical Problem

In forming multi-gate metal oxide semiconductor field effect transistors (MOSFETs), especially the channel region of the full-ring gate (GAA) FETs, the deposition and patterning process of different work function metal (WFM) layers are arranged in complex and difficult to adjust Threshold voltage (Vt), affecting device performance.

Method used

By forming a multi-layer stack of alternating silicon germanium (SiGe) and silicon (Si) layers on the semiconductor substrate, the SiGe and Si layers of different components are arranged as n-type and p-type channel regions, respectively, to form complementary The metal gate structure of the metal gate structure is removed by selective etching process, and a high k metal gate structure is formed to adjust the Vt of the NFET and PFET.

Benefits of technology

The manufacturing process of multiple WFM layers is simplified, processing complexity is reduced, Vt adjustment compatibility between NFET and PFET is achieved, and the manufacturing efficiency and performance of the device is improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113224057B_ABST
    Figure CN113224057B_ABST
Patent Text Reader

Abstract

The semiconductor structure includes a first stack of semiconductor layers disposed above a semiconductor substrate, wherein the first stack of semiconductor layers includes a first SiGe layer and a plurality of Si layers disposed above the first SiGe layer, and the Si layers are substantially free of Ge; and a second stack of semiconductor layers adjacent to the first stack of semiconductor layers, wherein the second stack of semiconductor layers includes a first SiGe layer and a plurality of second SiGe layers disposed above the first SiGe layer, and wherein the first SiGe layer and the second SiGe layer have different compositions. The semiconductor structure also includes a first metal gate stack interlaced with the first stack of semiconductor layers to form a first device and a second metal gate stack interlaced with the second stack of semiconductor layers to form a second device different from the first device. Embodiments of the present invention also relate to methods for forming semiconductor structures.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] Embodiments of the present invention relate to semiconductor structures and methods of forming the same. Background Art

[0002] The semiconductor industry has experienced rapid growth. Technological advances in semiconductor materials and design have produced generations of semiconductor devices, each with smaller and more complex circuits than the previous one. Over the course of integrated circuit (IC) evolution, functional density (i.e., the number of interconnected devices per chip area) has generally increased, while geometry size (i.e., the smallest component (or line) that can be created using a manufacturing process) has decreased. This scaling process generally provides benefits by increasing production efficiency and reducing associated costs. However, these advances have also increased the complexity of processing and manufacturing semiconductor devices.

[0003] Multi-gate transistors, such as gate-all-around (GAA) field-effect transistors (FETs), have been incorporated into various memory and core devices to reduce the footprint of IC chips while maintaining reasonable processing margins. Although methods for forming GAA FETs are generally adequate, they are not completely satisfactory in all respects. For example, when the channel region of a GAA FET is configured with multiple nano-components (e.g., nanosheets, nanorods, etc.) that are closely arranged in a vertical stack, the process of adjusting the threshold voltage of the metal gate structure involving the deposition and patterning of different work function metal (WFM) layers becomes challenging. Therefore, for at least this reason, improvements in methods for forming a metal gate structure with a suitable threshold voltage in a GAA FET are desired. Summary of the Invention

[0004] An embodiment of the present invention provides a semiconductor structure, comprising: a first stack of semiconductor layers, arranged above a semiconductor substrate, wherein the first stack of semiconductor layers includes a first silicon germanium (SiGe) layer and a plurality of silicon (Si) layers arranged above the first silicon germanium layer, and wherein the silicon layer does not contain germanium; a second stack of semiconductor layers, arranged adjacent to the first stack of semiconductor layers, wherein the second stack of semiconductor layers includes the first silicon germanium layer and a plurality of second silicon germanium layers arranged above the first silicon germanium layer, and wherein the first silicon germanium layer and the second silicon germanium layer have different compositions; a first metal gate stack, interlaced with the first stack of semiconductor layers to form a first device; and a second metal gate stack, interlaced with the second stack of semiconductor layers to form a second device, wherein the first device and the second device have different conductivity types.

[0005] Another embodiment of the present invention provides a semiconductor structure comprising: a first semiconductor fin protruding from a substrate, wherein a channel region of the first semiconductor fin comprises a stack of silicon (Si) layers disposed above a first silicon germanium (SiGe) layer; a second semiconductor fin protruding from the substrate, wherein a channel region of the second semiconductor fin comprises a stack of second silicon germanium layers disposed above the first silicon germanium layer, wherein an amount of germanium in the second silicon germanium layer is greater than an amount of germanium in the first silicon germanium layer, and wherein the number of the second silicon germanium layers in the second semiconductor fin is one more than the number of the silicon layers in the first semiconductor fin; a first metal gate stack bonded to the channel region of the first semiconductor fin; and a second metal gate stack bonded to the channel region of the second semiconductor fin.

[0006] Yet another embodiment of the present invention provides a method for forming a semiconductor structure, comprising: forming a first silicon-germanium-based (SiGe-based) layer over a substrate; forming a multilayer stack (ML) of alternating silicon-based (Si-based) layers and second silicon-germanium-based layers over the first silicon-germanium-based layer, wherein the first silicon-germanium-based layer and the second silicon-germanium-based layer have different compositions, and wherein the multilayer stack includes one more second silicon-germanium-based layer than the silicon-based layer; forming a first fin and a second fin adjacent to the first fin in the multilayer stack; forming a dummy gate stack over a channel region of the first fin and a channel region of the second fin; forming a dummy gate stack in the first fin; forming a dummy gate stack over the channel region of the first fin and the channel region of the second fin; forming a dummy gate stack in the first fin; forming a dummy gate stack over ... in the first fin; forming a dummy gate stack in the first fin; forming a dummy gate stack in the first fin; forming a dummy gate stack in the first fin; forming a dummy gate stack forming n-type source / drain (S / D) features in the second fin; forming p-type source / drain features in the second fin; removing the dummy gate stack between the n-type source / drain features and between the p-type source / drain features to form a first gate trench and a second gate trench, respectively; removing the second silicon-germanium-based layer in the first fin to form a first opening between the silicon-based layers; removing the silicon-based layer in the second fin to form a second opening between the second silicon-germanium-based layers and between the second silicon-germanium-based layer and the first silicon-germanium-based layer; and forming a metal gate stack in the first gate trench, the second gate trench, the first opening, and the second opening. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0008] Figure 1A and Figure 1B A flowchart illustrating an example method for fabricating a semiconductor device according to various embodiments of the present invention is shown.

[0009] Figure 2A is a three-dimensional perspective view of an example semiconductor device according to various embodiments of the present invention.

[0010] Figure 2B According to various embodiments of the present invention Figure 2A A planar top view of the semiconductor device shown in FIG.

[0011] Figure 3A 、 Figure 4A 、 Figure 5A 、 Figure 6A 、 Figure 7A 、 Figure 8A 、 Figure 9A 、 Figure 10A 、 Figure 11A 、 Figure 12A 、 Figure 13A 、 Figure 14A 、 Figure 15A 、 Figure 16A and Figure 17A According to various embodiments of the present invention Figure 1A and Figure 1B An example of a method of the intermediate stage taken along line AA' Figure 2A and Figure 2B A cross-sectional view of a semiconductor device is shown.

[0012] Figure 3B 、 Figure 4B 、 Figure 5B 、 Figure 6B 、 Figure 7B 、 Figure 8B 、 Figure 9B 、 Figure 10B 、 Figure 11B 、 Figure 12B 、 Figure 13B 、 Figure 14B 、 Figure 15B 、 Figure 16B and Figure 17B According to various embodiments of the present invention Figure 1A and Figure 1B An example of a method of the middle stage taken along line BB' Figure 2A and Figure 2B A cross-sectional view of a semiconductor device is shown.

[0013] Figure 3C 、 Figure 4C 、 Figure 5C 、 Figure 6C 、 Figure 7C 、 Figure 8C 、 Figure 9C 、 Figure 10C 、 Figure 11C 、 Figure 12C 、 Figure 13C 、 Figure 14C 、 Figure 15C 、 Figure 16C 、 Figure 17C 、 Figure 17E and Figure 17F According to various embodiments of the present invention Figure 1A and Figure 1B A view taken along line CC' at an intermediate stage of the exemplary method Figure 2A and Figure 2B A cross-sectional view of the semiconductor device shown.

[0014] Figure 3D 、 Figure 4D 、 Figure 5D 、 Figure 6D 、 Figure 7D 、 Figure 8D 、 Figure 9D 、 Figure 10D 、 Figure 11D 、 Figure 12D 、 Figure 13D 、 Figure 14D 、 Figure 15D 、 Figure 16D and Figure 17D According to various embodiments of the present invention Figure 1A and Figure 1B A view taken along line DD' at an intermediate stage of the exemplary method Figure 2A and Figure 2B A cross-sectional view of a semiconductor device is shown.

[0015] Figure 18 FIG. 4 is a diagram illustrating the relationship between current and work function of a semiconductor device according to various embodiments of the present invention. DETAILED DESCRIPTION

[0016] The following disclosure provides many different embodiments or examples for realizing the different features of the present invention. Specific examples of components and arrangements are described below to simplify the present invention. Of course, these are merely examples and are not intended to be limiting. For example, in the following description, forming a first component above or on a second component may include an embodiment in which the first component and the second component are directly in contact with each other, and may also include an embodiment in which an additional component may be formed between the first component and the second component so that the first component and the second component may not be in direct contact. In addition, for ease of description, the present invention may use spatial relative terms such as "lower," "upper," "lateral," "vertical," "above," "above," "below," "below," "upward," "downward," "top," "bottom," and their derivatives (e.g., "laterally," "downwardly," "upwardly," etc.) to describe the relationship between one component and another. The spatial relative terms are intended to encompass different orientations of the device comprising the components.

[0017] In addition, when "about," "approximately," or the like is used to describe a value or range of values, the term is intended to encompass values ​​within a reasonable range including the described value, such as within + / - 10% of the described value or other values ​​understood by those skilled in the art. For example, the term "about 5 nm" encompasses a size range from 4.5 nm to 5.5 nm. In addition, the present invention may repeat reference numbers and / or letters in various examples. This repetition is for the purpose of simplicity and clarity and does not in itself indicate a relationship between the various embodiments and / or configurations discussed.

[0018] The present invention generally relates to the structure and formation method of multi-gate metal oxide semiconductor field effect transistors (MOSFETs or FETs in the present invention) such as all-around gate (GAA) FETs. More specifically, the present invention relates to the structure and formation method of multi-layer channel regions in n-channel or n-type GAA FETs (GAA NFETs) and p-channel or p-type GAA FETs (GAA PFETs) that together form a complementary MOSFET (CMOSFET). The GAA FETs provided herein can be nanosheet-based FETs, nanowire-based FETs, and / or nanorod-based FETs. In other words, the present invention does not limit the GAA FETs to having a specific configuration.

[0019] Typically, the channel region of a GAA NFET and the channel region of a GAA PFET each include a stack of silicon-based channel layers (Si layers) interposed between metal gate structures. While such structures are generally sufficient to maintain the performance of GAA devices, they are not completely satisfactory in all aspects. For example, since NFETs and PFETs are typically configured with different work functions (WFs) in their respective gate structures in order to achieve a specific threshold voltage (V t ) requirements, and therefore include different work function metal (WFM) layers as part of the corresponding metal gate structure of the device. However, the fabrication of the WFM layer can include a series of deposition and patterning processes, which become complicated in the multilayer structure of the GAA FET due to the reduced length scale. The present embodiment provides a method for forming a GAA CMOSFET, wherein the channel region of the NFET and the channel region of the PFET are configured with different materials (Si layer and SiGe layer, respectively), allowing the V of the NFET and PFET to be adjusted using WFM layers of the same composition t , thereby reducing the processing complexity associated with manufacturing multiple WFMs.

[0020] refer to Figure 1A and Figure 1B, a flow chart of a method 100 of forming a semiconductor device (hereinafter referred to as a device) 200 according to various aspects of the present invention is shown. Method 100 is merely an example and is not intended to limit the present invention to that expressly recited in the claims. Additional operations may be provided before, during, and after methods 100 and 300, and some of the operations described may be replaced, eliminated, or rearranged for additional embodiments of the method. Figures 2A to 18 A method 100 is described, wherein Figure 2A It is a three-dimensional stereogram. Figure 2B is a flat top view, and Figures 3A to 17D It is a passage in the middle stage of the method Figure 2A and Figure 2B sectional views taken from various regions of the device 200 shown in FIG. Figure 3A 、 Figure 4A 、 Figure 5A 、 Figure 6A 、 Figure 7A 、 Figure 8A 、 Figure 9A 、 Figure 10A 、 Figure 11A 、 Figure 12A 、 Figure 13A 、 Figure 14A 、 Figure 15A 、 Figure 16A and Figure 17A 2 is a cross-sectional view taken along line AA′ through a fin active area (hereinafter referred to as fin) 204 of the device 200 . Figure 3B 、 Figure 4B 、 Figure 5B 、 Figure 6B 、 Figure 7B 、 Figure 8B 、 Figure 9B 、 Figure 10B 、 Figure 11B 、 Figure 12B 、 Figure 13B 、 Figure 14B 、 Figure 15B 、 Figure 16B and Figure 17B is a cross-sectional view taken along line BB′ through the fin 206 of the device 200 . Figure 3C 、 Figure 4C 、 Figure 5C 、 Figure 6C 、 Figure 7C 、 Figure 8C 、 Figure 9C 、 Figure 10C 、 Figure 11C 、 Figure 12C 、 Figure 13C 、 Figure 14C 、 Figure 15C 、 Figure 16C 、 Figure 17C 、 Figure 17E and Figure 17F is a cross-sectional view taken along line CC′ through the channel region of fin 204 and fin 206 . Figure 3D 、 Figure 4D 、 Figure 5D 、 Figure 6D 、 Figure 7D 、 Figure 8D 、 Figure 9D 、 Figure 10D 、 Figure 11D 、 Figure 12D 、 Figure 13D 、 Figure 14D 、 Figure 15D 、 Figure 16D and Figure 17D is a cross-sectional view taken along line DD′ through the source / drain (S / D) regions of fin 204 and fin 206, and Figure 18 is a schematic diagram illustrating the relationship between current and work function in an embodiment of a GAA FET provided herein.

[0021] Device 200 may be an intermediate device fabricated during processing of an IC or portion thereof, which may include static random access memory (SRAM) and / or other logic circuits, passive components (such as resistors, capacitors, and inductors), and active components (such as GAA FETs, FinFETs, MOSFETs, CMOSFETs, bipolar transistors, high-voltage transistors, high-frequency transistors, and / or other transistors). The present invention is not limited to any particular number of devices or device regions, or any particular device configuration. Additional components may be added to device 200, and some of the components described below may be replaced, modified, or eliminated in other embodiments of device 200.

[0022] refer to Figure 2A and Figures 3A to 3D , method 100 provides a semiconductor substrate (hereinafter referred to as "substrate") 202 at operation 102, and then forms a multilayer structure (ML) over substrate 202. Substrate 202 may include: an elemental (i.e., having a single element) semiconductor, such as silicon (Si), germanium (Ge), or other suitable materials; a compound semiconductor, such as silicon carbide, gallium arsenide, gallium phosphide, indium phosphide, indium arsenide, indium antimonide, other suitable materials, or combinations thereof; an alloy semiconductor, such as SiGe, GaAsP, AlInAs, AlGaAs, GaInAs, GaInP, GaInAsP, other suitable materials, or combinations thereof. Substrate 202 may be a single layer of material having a uniform composition. Alternatively, substrate 202 may include multiple layers of materials having similar or different compositions suitable for fabricating device 200.

[0023] In some examples where substrate 202 includes a FET, various doped regions may be provided in or on substrate 202. Depending on design requirements, the doped regions may be doped with n-type dopants (such as phosphorus or arsenic) and / or p-type dopants (such as boron or BF2). The doped regions may be formed directly on substrate 202 in a p-well structure, an n-well structure, a double-well structure, or a raised structure. The doped regions may be formed by implanting dopant atoms, epitaxial growth with in-situ doping, and / or other suitable techniques. Of course, these examples are for illustrative purposes only and are not intended to be limiting.

[0024] In this embodiment, the ML includes alternating silicon germanium (SiGe) and silicon (Si) layers arranged in a vertical stack along the Z-axis and configured to provide a channel region suitable for forming at least one GAA NFET and at least one GAA PFET. In the embodiment shown, the bottommost layer of the ML is SiGe layer 203, and subsequent layers of the ML include alternating Si layers 205 and SiGe layers 207, wherein, as discussed in detail below, the SiGe layers 207 are configured as channel layers for the GAA PFET. In this embodiment, the ML includes the same number of Si layers 205 as the combined number of SiGe layers 203 and SiGe layers 207. In other words, the number of SiGe layers 207 is one less than the number of Si layers 205. In some examples, the ML can include three to ten Si layers 205, and thus two to nine SiGe layers 207.

[0025] In the present embodiment, each Si layer 205 comprises elemental Si and is substantially free of Ge, while SiGe layer 203 and each SiGe layer 207 comprise substantially both Si and Ge, although the amount of Ge in SiGe layer 203 is less than the amount of Ge in SiGe layer 207. In some embodiments, the amount of Ge in SiGe layer 207 is from about 20% to about 50%, and the amount of Ge in SiGe layer 203 is at least about 10% but less than about 20%. Thus, in the present embodiment, the amount of Si in SiGe layer 207 is from about 50% to about 80%, and the amount of Si in SiGe layer 203 is at least about 80% but less than about 90%. In contrast, the amount of silicon in each Si layer 205 is at least about 90%.

[0026] In this embodiment, the Ge with a minimum composition of about 20% increases the hole mobility of SiGe in the SiGe layer 207 relative to the Si layer 205, thereby reducing the V t regulated with the NFET’s V t Compatible with the subsequently formed metal gate structure required (i.e. Figure 17B 、 Figure 17C and Figure 18The WF of the high-k metal gate structure 280B shown in FIG. A composition less than about 20% may not be sufficient to cause the V t This shift in the desired WF. On the one hand, Ge at a minimum composition of approximately 10% relative to SiGe layer 203 ensures sufficient etch selectivity between Si layer 205 and SiGe layer 203, such that when Si layer 205 (i.e., a non-channel layer) is removed to form an opening between SiGe layer 207 (i.e., a channel layer) of the PFET, SiGe layer 203 may not be etched or not completely etched. On the other hand, Ge in SiGe layer 203 not exceeding approximately 20% (i.e., less than the minimum composition of Ge in SiGe layer 207) ensures sufficient etch selectivity between SiGe layer 203 and SiGe layer 207, such that when SiGe layer 207 (i.e., a non-channel layer) is removed to form an opening between Si layer 205 (i.e., a channel layer) of the NFET, SiGe layer 203 may not be etched or substantially not etched. In other words, SiGe layer 203 is configured as a stop layer to protect substrate 202 from being accidentally damaged during multiple subsequent manufacturing processes.

[0027] Along similar lines, a minimum composition of Ge of approximately 20% in the SiGe layer 207 ensures sufficient etching selectivity relative to the SiGe layer 203. On the other hand, Ge exceeding approximately 50% in the SiGe layer 207 (i.e., the channel layer of the PFET) may induce defects in subsequently formed epitaxial S / D features, exacerbate problems related to leakage current, and / or worsen the drain-induced barrier lowering (DIBL) effect due to a reduced bandgap.

[0028] In the present embodiment, forming the ML includes alternately growing SiGe layers (i.e., SiGe layer 203 or SiGe layer 207) and Si layers (i.e., Si layer 205) in a series of epitaxial growth processes, the epitaxial growth processes using chemical vapor deposition (CVD) techniques (e.g., vapor phase epitaxy (VPE), ultra-high vacuum CVD (UHV-CVD), low pressure (LP-CVD) and / or plasma enhanced CVD (PE-CVD)), molecular beam epitaxy, other suitable selective epitaxial growth (SEG) processes, or combinations thereof. The epitaxial processes may use gaseous and / or liquid precursors that interact with components of the underlying substrate. For example, a substrate 202 comprising Si may interact with a Ge-containing precursor to form the SiGe layer 203. In some examples, the SiGe layer 203, the Si layer 205, and the SiGe layer 207 may be formed as nanosheets, nanowires, or nanorods. In the present embodiment, each of the SiGe layer 203, the Si layer 205, and the SiGe layer 207 is formed along Figure 3A The Z-axis shown measures substantially the same thickness T.

[0029] In this embodiment, the Si layer 205 is configured as a channel layer for forming an NFET, while the SiGe layer 203 and the SiGe layer 207 are configured as channel layers for forming a PFET. Therefore, the SiGe layer 207 disposed between the Si layers 205 of the NFET is configured as a non-channel layer of the NFET, and the Si layer 205 disposed between the SiGe layers 207 is configured as a non-channel layer of the PFET. A sheet (or line) release process can then be performed to form a plurality of openings between the corresponding channel layers, and subsequently metal gate structures are formed in the openings to complete the fabrication of the corresponding FET. Notably, as discussed herein, due to their compositional differences, the SiGe layer 207 is selectively removed relative to the SiGe layer 203 during the subsequent sheet formation (or sheet release) process for forming the PFET.

[0030] Now Figure 2A 、 Figure 2B and Figures 4A to 4D , method 100 forms fins 204 and fins 206 over substrate 202 at operation 104. In the embodiment shown, fins 204 and 206 are disposed adjacent to each other and substantially parallel, i.e., oriented longitudinally along the X-axis and spaced apart along the Y-axis. As discussed in detail below, although fins 204 and 206 are fabricated from the same ML and substrate 202, they are configured to provide GAA FETs of different conductivity types, i.e., one of fins 204 and 206 is configured to provide an NFET and the other of fins 204 and 206 is configured to provide a PFET. In the embodiment shown, fin 204 is configured to provide an NFET and fin 206 is configured to provide a PFET. Thus, fin 204 can be formed in a region of substrate 202 doped with p-type dopants (i.e., a p-well structure), and fin 206 can be formed in a region of substrate 202 doped with n-type dopants (i.e., an n-well structure). Note that embodiments of the device 200 may include additional fins (semiconductor fins) disposed above the substrate 202 , the additional fins configured to provide one or more NFETs and / or PFETs.

[0031] In this embodiment, still refer to Figure 2A 、 Figure 2B and Figures 4A to 4DEach fin 204 includes an ML disposed above a base fin 204', and each fin 206 includes an ML disposed above a base fin 206', wherein the base fins 204' and 206' protrude from the substrate 202. The fins 204 and 206 may be fabricated using suitable processes including photolithography and etching processes. The photolithography process may include forming a mask element having a hard mask layer 220 disposed above the ML, a hard mask layer 222 disposed above the hard mask layer 220, and a photoresist layer (photoresist; not shown) disposed above the hard mask layer 222, exposing the photoresist to a pattern, performing a post-exposure bake process on the photoresist, and developing the photoresist to form a patterned mask element that exposes portions of the ML. The patterned mask element is then used to etch recesses in the ML and portions of the substrate 202, leaving the fins 204 and 206 protruding from the substrate 202. The hard mask layers 220 and 222 have different compositions and can each include silicon oxide, silicon nitride, silicon carbide, silicon oxynitride, silicon oxycarbide, other suitable materials, or combinations thereof. The etching process can include dry etching, wet etching, reactive ion etching (RIE), other suitable processes, or combinations thereof.

[0032] Many other embodiments of methods for forming fins 204 and 206 may be suitable. For example, fins 204 and 206 may be patterned using a double patterning or multiple patterning process. Typically, double patterning or multiple patterning processes combine photolithography and self-aligned processes, allowing for the creation of patterns with a pitch smaller than that achievable using a single direct photolithography process, for example. For example, in one embodiment, a sacrificial layer is formed over a substrate and patterned using a photolithography process. Spacers are formed adjacent to 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 to pattern fins 204 and 206.

[0033] refer to Figure 2A and 5A to 5D , the method 100 forms an isolation structure 208 over the substrate 202 at operation 106, and separates the bottoms of the fins 204 and 206. The isolation structure 208 may include silicon oxide, fluoride-doped silicate glass (FSG), low-k dielectric material, other suitable materials, or combinations thereof. In the present embodiment, the isolation structure 208 includes a shallow trench isolation (STI) feature. Figure 5C and Figure 5DAs shown, in some embodiments, the isolation structure 208 is formed by depositing a dielectric layer over the substrate 202, thereby filling the trench between the fins 204 and 206, and then recessing the dielectric layer so that the top surface of the isolation structure 208 is below the top surface of the fins 204 and 206. Other isolation structures such as field oxide, local oxidation of silicon (LOCOS), other suitable structures, or combinations thereof may also be implemented as the isolation structure 208. In some embodiments, the isolation structure 208 may include a multi-layer structure, for example, with one or more thermal oxide liner layers. The isolation structure 208 may be deposited by any suitable method, such as CVD, flowable CVD (FCVD), spin-on-glass (SOG), other suitable methods, or combinations thereof.

[0034] refer to 6A to 6D , method 100 forms a dielectric fin 223 above the isolation structure 208 at operation 108, so that each of the fins 204 and 206 is disposed between two dielectric fins 223. Each dielectric fin 223 can be a single-layer structure or a multi-layer structure. In the present embodiment, the dielectric fin 223 is a three-layer structure, including a first layer 225 disposed on the isolation structure 208, a second layer 227 surrounded by the first layer 225, and a third layer 229 disposed above the first layer 225 and the second layer 227. Each of the first layer 225, the second layer 227, and the third layer 229 can include one or more dielectric materials, such as silicon oxide, silicon nitride, silicon carbide, silicon oxynitride, silicon oxycarbide, other suitable materials, or combinations thereof. In some embodiments, the components of the first layer 225, the second layer 227, and the third layer 229 are different. The dielectric fins 223 may be formed by any suitable process, including depositing and planarizing a first layer 225 over the device 200 to fill the space around the fins 204 and 206, patterning (e.g., by photolithography) the first layer 225 to form a trench between the fins 204 and 206, depositing and planarizing a second layer 227 in the trench, depositing a third layer 229 over the first and second layers 225, 227, patterning the third layer 229 to expose portions of the first layer 225, and removing the exposed portions of the first layer 225 using the patterned third layer 229 as a hard mask. The first, second, and third layers 225, 227, and 229 may be formed by any suitable deposition process, such as CVD, FCVD, ALD, other suitable processes, or combinations thereof. In the present embodiment, the dielectric fins 223 are configured to control the subsequent formation of n-type and p-type epitaxial S / D features over the fins 204 and 206, respectively. In some examples, dielectric fins 223 can prevent overgrowth of epitaxial S / D features that could unintentionally cause shorts in device 200 .

[0035] Now refer to Figure 2A 、 Figure 2B and 7A to 7DAt operation 110, method 100 forms a dummy gate stack (i.e., a placeholder gate) 210 over the channel region of each of the fins 204 and 206. In this embodiment, after forming the other components of device 200, a portion of the dummy gate stack 210 comprising polysilicon is replaced with a high-k (referring to a dielectric material having a dielectric constant greater than the dielectric constant of silicon oxide, which has a dielectric constant of approximately 3.9) metal gate structure (HKMG). The dummy gate stack 210 can be formed by a series of deposition and patterning processes. For example, the dummy gate stack 210 can be formed by depositing a polysilicon layer over the fins 204 and 206 and then performing an anisotropic etching process (e.g., a dry etching process) to leave portions of the polysilicon over the channel regions of the fins 204 and 206. In this embodiment, the device 200 further includes an interfacial layer 209, which is formed on the fins 204 and 206 by a suitable method such as thermal oxidation, chemical oxidation, other suitable methods, or combinations thereof before depositing the dummy gate stack 210. In the embodiment shown, a hard mask layer 211 and a hard mask layer 213 are formed over the dummy gate stack 210 to protect the dummy gate stack 210 from being etched during subsequent operations. Each of the hard mask layers 211 and 213 can include any suitable dielectric material discussed above with respect to the hard mask layers 220 and 222, and can be formed by a suitable deposition process such as CVD, ALD, PVD, other suitable processes, or combinations thereof. Thereafter, the hard mask layers 211 and 213 are removed before removing the dummy gate stack 210 to form the HKMG.

[0036] Afterwards, refer to Figures 8A to 8D , the method 100 forms an S / D recess 230A in the S / D region of the fin 204 and forms an S / D recess 230B in the S / D region of the fin 206 at operation 112 . Figure 8A and Figure 8B Before forming the S / D recesses 230A and 230B, the method 100 first forms a top spacer 212 on the sidewalls of the dummy gate stack 210. The top spacer 212 can be a single-layer structure or a multi-layer structure and can include silicon oxide, silicon nitride, silicon carbide, silicon oxynitride, silicon oxycarbide, other suitable materials, or combinations thereof. Each spacer layer of the top spacer 212 can be formed by first depositing a dielectric layer over the dummy gate stack 210 and then removing a portion of the dielectric layer in an anisotropic etching process (e.g., a dry etching process), leaving a portion of the dielectric layer located on the sidewalls of the dummy gate stack 210 as the top spacer 212.

[0037] Then, still refer to Figure 8A and Figure 8B , and further reference Figure 8DMethod 100 removes portions of the ML in the S / D regions of fins 204 and 206 by etching process 302, stopping at SiGe layer 203. Etching process 302 may be a dry etching process, a wet etching process, RIE, or a combination thereof. In this embodiment, method 100 employs an etchant configured to remove SiGe layer 203, Si layer 205, and SiGe layer 207 at operation 112. In other words, etching process 302 is not selective for a particular material layer of the ML. In some examples, method 100 may employ a dry etching process using a chlorine-containing etchant (e.g., Cl2, SiCl4, BCl3, other chlorine-containing gases, or combinations thereof), a bromine-containing etchant (e.g., HBr), other suitable etchants, or combinations thereof. In some embodiments, the depth of S / D recesses 230A and 230B is controlled by adjusting the duration, temperature, pressure, source power, bias voltage, bias power, etchant flow rate, other suitable parameters, or combinations thereof, of etching process 302. In the embodiment shown, the etching process 302 is controlled such that the S / D recesses 230A and 230B expose portions of the substrate 202 , and a cleaning process may then be performed using hydrofluoric acid (HF) and / or other suitable solvents to remove any etching residues in the S / D recesses 230A and 230B.

[0038] Common Reference 9A to 11D , method 100 forms internal spacers 240 on the sidewalls of the non-channel layer in the portions of the ML exposed in the S / D recesses 230A and 230B, respectively, at operations 114-118. In this embodiment, the internal spacers 240 are configured to separate the epitaxial S / D features of the NFET and PFET from their respective HKMGs formed between the channel layers. In this embodiment, as discussed in detail below, the internal spacers 240 are formed separately for the NFET and the PFET.

[0039] refer to 9A to 9D , the method 100 selectively removes a portion of the SiGe layer 207 configured as a non-channel layer of the NFET at operation 114 to form a recess 234. To prevent unintentional damage to the SiGe layer 207 configured as a channel layer of the PFET, a mask element 232A is formed over the device 200 and then patterned to expose a portion of the fin 204 while covering a portion of the fin 206. The patterned mask element 232A may include at least a photoresist layer that can be patterned by photolithography, similar to the above description of, for example, forming a fin 206. Figures 4A to 4DThe method 100 then uses an etching process 304 to selectively remove the portion of the SiGe layer 207 exposed in the S / D recess 230A without removing or substantially removing portions of the Si layer 205 and the SiGe layer 203. In this embodiment, the etching process 304 has a selectivity for Ge content of at least about 20%, such that the SiGe layer 207 is etched at a higher rate than both the Si layer 205, which is substantially free of Ge, and the SiGe layer 203, which has a Ge content of less than about 20%. In some embodiments, the etching process 304 is a wet etching process using hydrogen peroxide (H2O2), a hydroxide (e.g., ammonium hydroxide (NH4OH), tetramethylammonium hydroxide (TMAH), etc.), acetic acid (CH3COOH), other suitable etchants, or combinations thereof. In some embodiments, the etching process 304 is a dry etching process using a fluorine-containing etchant, such as HF, F2, NF3, other fluorine-containing etchants, or combinations thereof. In this embodiment, the duration of the etching process 304 is controlled to ensure that only a portion of each SiGe layer 207 is etched to form the recesses 234. In some embodiments, various parameters of the etching process 304 (e.g., the etchant used) are adjusted to ensure high etching uniformity between the recesses 234 so that the gate length L of the NFET can be reduced. N is controlled to a desired value between the Si layers 205. After performing the etching process 304 to form the recesses 234, the patterned masking element 232A is removed from the device 200 by any suitable method, such as plasma ashing and / or photoresist stripping.

[0040] Now refer to 10A to 10D, method 100 selectively removes a portion of the Si layer configured as a non-channel layer of the PFET at operation 116 to form recess 236. A mask element 232B substantially similar to mask element 232A may be applied over device 200, and mask element 232B may be patterned to protect portions of fin 204 while exposing portions of fin 206. Subsequently, method 100 employs an etching process 306 to selectively remove portions of Si layer 205 exposed in S / D recess 230B without removing or substantially removing portions of SiGe layer 203 and SiGe layer 207. In the present embodiment, etching process 306 has an etching selectivity to Si such that Si layer 205, which includes elemental Si and is substantially free of other elements (e.g., Ge), is etched at a greater rate than SiGe layer 203 and SiGe layer 207. After performing the etching process 306 to form the recesses 236, the patterned mask element 232B is removed from the device 200 by any suitable method, such as plasma ashing and / or photoresist stripping. In some embodiments, the etching process 306 is a dry etching process that uses hydrogen, a fluorine-containing etchant (such as F2, CF4, other fluorine-containing etchants), a nitrogen-containing etchant such as NH3, other nitrogen-containing etchants, or a combination thereof. In some embodiments, the etching process 306 is a wet etching process that uses a hydroxide (such as NH4OH), other suitable etchants, or a combination thereof. In this embodiment, the duration of the etching process 306 is controlled to ensure that only a portion of the Si layer 205 is etched to form the recesses 236. In some embodiments, various parameters of the etching process 306 (e.g., the etchant used) are adjusted to ensure high etching uniformity between the recesses 236 so that the gate length L of the PFET can be reduced. P is controlled to a desired value between SiGe layers 207 and 203. After performing the etching process 306 to form recess 236, the patterned mask element 232B is removed from the device 200 by any suitable method (such as plasma ashing and / or photoresist stripping). Note that the present embodiment does not limit the order in which recesses 234 and 236 are formed in SiGe layer 207 and Si layer 205, respectively. In other words, although as shown herein, recess 234 is first formed at operation 114 and recess 236 is subsequently formed at operation 116, in some embodiments of the present invention, recess 236 may alternatively be formed before recess 234 is formed.

[0041] refer to 11A to 11DAt operation 118, method 100 forms inner spacers 240 in recesses 234 and 236. The inner spacers 240 may include any suitable dielectric material including silicon, carbon, oxygen, nitrogen, other elements, or combinations thereof. For example, the inner spacers 240 may include silicon nitride, silicon carbide, silicon oxide, carbon-containing silicon nitride (SiCN), carbon-containing silicon oxide (SiOC), oxygen-containing silicon nitride (SiON), carbon and oxygen-doped silicon nitride (SiOCN), a low-k dielectric material, tetraethyl orthosilicate (TEOS), doped silicon oxide (e.g., borophosphosilicate glass (BPSG), fluorine-doped silicate glass (FSG), phosphosilicate glass (PSG), boron-doped silicate glass (BSG), etc.), air, other suitable dielectric materials, or combinations thereof. The inner spacers 240 may each be configured as a single-layer structure or a multi-layer structure including a combination of dielectric materials provided herein. In some embodiments, the inner spacers 240 have a different composition than the top spacer 212. In some embodiments, the inner spacer 240 and the top spacer 212 have substantially the same composition. The method 100 can form the inner spacer 240 by depositing one or more dielectric layers in the recesses 234 and 236 via any suitable deposition process (such as ALD, CVD, other suitable methods, or combinations thereof) and then performing one or more etching processes to remove any excess dielectric material formed on the sidewalls of the channel layer (i.e., the Si layer 205 in the fin 204 and the SiGe layer 207 in the fin 206).

[0042] Now refer to 12A to 12DAt operation 120 , method 100 forms an n-type epitaxial S / D feature 250 in each S / D recess 230A. Each n-type epitaxial S / D feature 250 is configured to form an NFET in conjunction with a subsequently formed HKMG. The n-type epitaxial S / D feature 250 may include one or more epitaxial layers of silicon (epitaxial Si) or silicon carbon (epitaxial SiC) doped with an n-type dopant, such as arsenic, phosphorus, other n-type dopants, or combinations thereof. In this embodiment, a hard mask layer 244A is formed over the device 200 and then patterned to protect the fins 206 while exposing the fins 204 prior to forming the n-type epitaxial features 250. The hard mask layer 244A may include a suitable dielectric material and is patterned by a photolithography process using a mask element (not shown) similar to the mask element 232A discussed in detail above with respect to operation 114 . In this embodiment, one or more epitaxial growth processes are performed to grow epitaxial material in each S / D recess 230A. For example, method 100 can employ the epitaxial growth process discussed above with respect to forming the Si layer 205 and the SiGe layer 207 of the ML. In some embodiments, the epitaxial material is doped in situ by adding dopants to the source material during epitaxial growth. In some embodiments, the epitaxial material is doped by an ion implantation process after the deposition process. In some embodiments, an annealing process is subsequently performed to activate the dopants in the n-type epitaxial S / D features 250. Thereafter, the patterned hard mask layer 244A is removed from the device 200 by a suitable process, such as a dry etch process, a wet etch process, or a combination thereof.

[0043] Then, refer to 13A to 13B , method 100 forms a p-type epitaxial S / D feature 252 in each S / D recess 230B at operation 122. Each p-type epitaxial S / D feature 252 is configured to form a PFET together with a subsequently formed HKMG. The p-type epitaxial S / D feature 252 may include one or more epitaxial layers of silicon germanium (epitaxial SiGe) doped with a p-type dopant such as boron, germanium, indium, other p-type dopants, or combinations thereof. In this embodiment, a hard mask layer 244B is formed over the device 200 and is subsequently patterned to protect the fins 204 while exposing the fins 206 prior to forming the p-type epitaxial S / D features 252. The hard mask layer 244B may include a suitable dielectric material and is patterned by a photolithography process using a mask element (not shown) similar to the mask element 232A discussed in detail above with respect to operation 114. In the present embodiment, p-type epitaxial S / D features 252 are formed and doped during one or more of the epitaxial growth and doping processes discussed above with respect to forming n-type epitaxial features 250. The patterned hard mask layer 244B is then removed from the device 200 by a suitable etching process similar to that discussed above with respect to operation 120.

[0044] refer to 14A to 14D , the method 100 removes the dummy gate stack 210 at operation 124 to form a gate trench 262A exposing the channel region of the fin 204 and a gate trench 262B exposing the channel region of the fin 206. Before removing the dummy gate stack 210, the method 100 forms an interlayer dielectric (ILD) layer 260 over the n-type epitaxial S / D feature 250 and the p-type epitaxial S / D feature 252 by, for example, CVD, FCVD, other suitable methods, or combinations thereof. The ILD layer 260 may include silicon oxide, a low-k dielectric material, TEOS, doped silicon oxide (e.g., BPSG, FSG, PSG, BSG, etc.), other suitable dielectric materials, or combinations thereof. In some embodiments, as Figure 14A 、 Figure 14B and Figure 14D As shown, the method 100 first forms an etch stop layer (ESL) 261 over the n-type epitaxial S / D features 250 and the p-type epitaxial S / D features 252 before forming the ILD layer 260. The ESL 261 may include silicon nitride, silicon carbide, carbon-containing silicon nitride (SiCN), oxygen-containing silicon nitride (SiON), carbon and oxygen-doped silicon nitride (SiOCN), aluminum nitride, high-k dielectric material, other suitable materials, or combinations thereof, and may be formed by CVD, ALD, PVD, other suitable methods, or combinations thereof. Subsequently, the method 100 may planarize the ESL 261 and the ILD layer 260 in one or more CMP processes to expose the top surface of the dummy gate stack 210. Thereafter, referring to 14A to 14C At least a portion of the dummy gate stack 210 is removed from the device 200 by any suitable etching process, such as a dry etching process, to form gate trenches 262A and 262B. In this embodiment, after removing the dummy gate stack 210, the interface layer 209 remains above the ML.

[0045] Common reference 15A to 16D , the method 100 performs a sheet forming process on the fin 204 and the fin 206 at operations 126 and 128 , respectively, thereby forming an opening 264 between the Si layer 205 in the fin 204 and an opening 266 between the SiGe layer 207 in the fin 206 . 15A to 15D , method 100 forms a mask element 270A over device 200 at operation 126, and then patterns the mask element 270A to protect the channel region of fin 206 and expose the channel region of fin 204. The mask element 270A may be substantially similar in composition to that described above with respect to, for example, forming a mask element 270A. 9A to 9DThe masking element 232A is shown as discussed with respect to the recess 234. Thereafter, the SiGe layer 207 is selectively removed from the channel region of the fin 204 by an etching process 308 that does not or substantially does not remove the SiGe layer 203 and the Si layer 205 from the channel region of the fin 204, thereby forming an opening 264.

[0046] As discussed above, the SiGe layer 207 in the channel region of the fin 204 is considered a non-channel layer, while the silicon layer 205 is considered a channel layer configured to form an NFET. In the present embodiment, the etching process 308 includes dry etching, wet etching, RIE, or a combination thereof, and utilizes an etchant having an etch selectivity for Ge relative to Si, wherein the Ge content is at least approximately 20%. Thus, the etching process 308 removes the SiGe layer 207 without removing or substantially removing the SiGe layer 203, which includes less than approximately 20% Ge, and the Si layer 205, which is substantially free of Ge. In some embodiments, the etching process 308 is similar to the etching process 304 with respect to the etchant used and / or other relevant etching parameters. Of course, other suitable etching processes different from the etching process 304 may also be used, as long as they effectively remove the SiGe layer 207 selectively relative to the Si layer 205 and the SiGe layer 203. In this embodiment, the etching process 308 is controlled to ensure that all of the SiGe layer 207 is removed from the fin 204, so that an opening 264 is formed between the silicon layers 205 (and between the SiGe layer 203 and the bottommost Si layer 205), where the silicon layer 205 is the channel layer of the NFET. Thereafter, the patterned mask element 270A is removed from the device 200 by any suitable method (such as plasma ashing and / or photoresist stripping). After or simultaneously with the removal of the SiGe layer 207, the method 100 removes the portion of the interface layer 209 disposed above the channel region of the fin 204 at operation 126.

[0047] Then, refer to 16A to 16D At operation 128, method 100 forms a mask element 270B over device 200, and then patterns mask element 270B, which may be substantially similar in composition to mask element 270A, to protect the channel region of fin 206 and expose the channel region of fin 204. Thereafter, Si layer 205 is selectively removed from the channel region of fin 206 by an etching process 310 that does not or does not substantially remove SiGe layer 203 and SiGe layer 207 from the channel region of fin 206, thereby forming opening 266.

[0048] As discussed above, the Si layer 205 in the channel region of the fin 206 is considered a non-channel layer, while the SiGe layer 207 is considered a channel layer configured to form a PFET. In this embodiment, the etching process 310 includes dry etching, wet etching, RIE, or a combination thereof, and utilizes an etchant that has an etching selectivity for Si relative to Ge. In some embodiments, the etching process 310 is configured to remove a material layer having a Si content greater than approximately 90%. Thus, the etching process 310 removes the Si layer 205 without removing or substantially removing the SiGe layer 203, which includes less than approximately 90% Si, and the SiGe layer 207, which includes less than approximately 80% Si. In some embodiments, the etching process at operation 128 is similar to the etching process 306 with respect to the etchant used and / or other relevant etching parameters. Of course, other suitable etching processes other than the etching process 306 may also be suitable, as long as they effectively and selectively remove the Si layer 205 relative to the SiGe layers 203 and 207. In this embodiment, the etching process 310 is controlled to ensure that all of the Si layer 205 is removed from the fin 206, so that an opening 266 is formed between the SiGe layers 207 (and between the SiGe layer 203 and the bottommost SiGe layer 207), which serve as the channel layer of the PFET. The patterned mask element 270B is then removed from the device 200 by any suitable method, such as plasma ashing and / or photoresist stripping. After or concurrently with the removal of the Si layer 205, the method 100 removes the portion of the interface layer 209 disposed above the channel region of the fin 206 at operation 128. Although, in the illustrated embodiment, the opening 264 is formed before the opening 266, the order in which the openings 264 and 266 are formed is not limited thereto. For example, the method 100 may first form the opening 266 before forming the opening 264.

[0049] Now refer to 17A to 17D At operation 130 , method 100 forms HKMG 280A over the channel region of fin 204 to form an NFET, and forms HKMG 280B over the channel region of fin 206 to form a PFET. In this embodiment, the tops of HKMGs 280A and 280B are formed in gate trenches 262A and 262B, respectively, while the bottoms of HKMGs 280A and 280B are formed in openings 264 and 266, respectively.

[0050] In this embodiment, each of HKMGs 280A and 280B includes at least a high-k dielectric layer 282 disposed above and surrounding the channel layers of the NFET and PFET, and a metal gate electrode disposed above the high-k dielectric layer 282. In this embodiment, the high-k dielectric layer 282 includes any suitable high-k dielectric material, such as hafnium oxide, lanthanum oxide, other suitable materials, or combinations thereof. In this embodiment, the metal gate electrode of HKMG 280A includes at least a work function metal (WFM) layer 284A disposed above the high-k dielectric layer 282 and a conductive layer 286 disposed above the WFM layer 284A, and the metal gate electrode of HKMG 280B includes at least a WFM layer 284B disposed above the high-k dielectric layer 282 and a conductive layer 286 disposed above the WFM layer 284B. Each of the WFM layer 284A and the WFM layer 284B can be a single-layer structure or a multi-layer structure including at least a p-type WFM layer, an n-type WFM layer, or a combination thereof. The conductive layer 286 can include Cu, W, Al, Co, Ru, other suitable materials, or a combination thereof. In the illustrated embodiment, each of the HKMGs 280A and 280B includes an interface layer 281 formed between each channel layer and the high-k dielectric layer 282. The HKMGs 280A and 280B can also include other layers (not shown), such as a capping layer, a barrier layer, other suitable layers, or a combination thereof. In some embodiments, the number of material layers included in each HKMG 280A and 280B is determined by the size of the openings 264 and 266, respectively. The various layers of the HKMGs 280A and 280B can be formed by any suitable method, such as chemical oxidation, thermal oxidation, ALD, CVD, PVD, plating, other suitable methods, or a combination thereof.

[0051] Typically, the WFM layer included in the gate electrode of the device can be adjusted by adjusting the type of the WFM layer and the I 截止_源极 To adjust the device threshold voltage V t , when the gate voltage is 0V and the drain voltage is VDD, the value of the source leakage current is related to V t In a CMOSFET, the material used for the channel regions of the NFET and PFET is the same, for example, both include Si, so I as a function of WF 截止_源极 The variation of is different between NFET and PFET. For example, refer to Figure 18 , will I 截止_源极 The linear correlation 402 described as a function of the WF of the NFET has a negative slope, and the I 截止_源极 The linear correlation 404 described as a function of the WF of the PFET has a positive slope. tIn this embodiment, this difference in WF typically does not allow the NFET and PFET to share a WFM layer of the same composition, thereby increasing the processing complexity associated with the HKMG formation (e.g., operation 130 described above). In this embodiment, the channel region of the PFET is configured with SiGe, which has a different band structure than Si. This difference causes the linear correlation 404 to shift laterally from the linear correlation 402, so that at the intersection 410, both the NFET and the PFET can be adjusted to the same (or substantially the same) WF value, i.e., WF*, which corresponds to the desired V for a given value of I*. t In other words, the incorporation of SiGe into the channel region of the PFET reduces the time required to configure the HKMG 280B to achieve the desired V t Therefore, a WFM layer having the same or substantially similar composition can be formed in the NFET and the PFET, effectively reducing the process complexity and cost of forming the HKMGs 280A and 280B.

[0052] In this embodiment, the SiGe layer 207 in the channel region of the fin 206 allows the WFM layers 284A and 284B to have the same or substantially similar compositions, such that the WF of each of the HKMG 280A and HKMG 280B is tuned to approximately the same value of WF*, which is approximately 4.6 eV ± 0.15 eV, corresponding to an I* of approximately 0.1 A / m ± 2. In some embodiments, each of the WFM layers 284A and 284B includes TiN, TaN, Ru, Mo, Al, WN, ZrSi2, MoSi2, TaSi2, NiSi2, Ti, Ag, TaAl, TaAlC, TiAlN, TaC, TaCN, TaSiN, Mn, Zr, other suitable WFM layers, or combinations thereof. Therefore, instead of independently adjusting the WF of HKMG 280A and 280B by incorporating different WFM layers, the present embodiment provides a method of performing a sheet forming process (i.e., separate selective etching processes) on the NFET and PFET separately so as to obtain channel regions including different semiconductor materials capable of shifting the WF of the NFET and PFET independently of the influence of the WFM layer.

[0053] Compare Figure 17A and Figure 17B, the portion of HKMG 280A disposed over the uppermost Si layer 205 is above or substantially flush with the top surface of n-type epitaxial S / D component 250 (i.e., above or substantially flush with the bottom surface of ILD layer 260), while HKMG 280B disposed over the topmost SiGe layer 207 extends below the top surface of p-type epitaxial S / D component 252 (i.e., below the bottom surface of ILD layer 260). This offset in gate height is attributed to the arrangement of alternating Si layers 205 and SiGe layers 207 (and 203), which are removed to form the channel regions of the PFET and NFET, respectively. In addition, referring to Figure 17C 、 Figure 17E and Figure 17F ,in Figure 17E and Figure 17F The HKMGs 280A and 280B are shown in more detail, respectively. The channel region of the fin 204 is formed by a height H n The channel region of the fin 206 is defined by the height H p Limit, where H n and H p are measured from the top surface of the SiGe layer 203, and H n Greater than H p The Si layer 205 and the SiGe layer 207 have the same Figures 3A to 3D For the discussed embodiment of substantially the same thickness T, this difference in channel height is attributed to the number of Si layers 205 in the channel region of fin 204 being one greater than the number of SiGe layers 207 in the channel region of fin 206 .

[0054] Thereafter, method 100 may perform additional processing steps on device 200 at operation 132. For example, method 100 may form S / D contacts (not shown) over epitaxial S / D features 250 and 252, respectively. Each S / D contact may comprise any suitable conductive material, such as Co, W, Ru, Cu, Al, Ti, Ni, Au, Pt, Pd, other suitable conductive materials, or combinations thereof. Method 100 may form S / D contact openings in ILD layer 260 via a series of patterning and etching processes, and then deposit conductive material in the S / D contact openings using any suitable method, such as CVD, ALD, PVD, plating, other suitable processes, or combinations thereof. In some embodiments, a silicide layer (not shown) is formed between epitaxial S / D features 250 and 252 and their respective S / D contacts. The silicide layer may comprise nickel silicide, cobalt silicide, tungsten silicide, tantalum silicide, titanium silicide, platinum silicide, silicide, palladium silicide, other suitable silicides, or combinations thereof. A silicide layer may be formed over the device 200 by a deposition process such as CVD, ALD, PVD, or a combination thereof. Subsequently, although not shown, the method 100 may form additional features over the device 200, such as an ESL disposed over the ILD layer 260, an ILD layer disposed over the ESL, a gate contact in the ILD layer to contact the HKMG 280, vertical interconnect features (e.g., vias), horizontal interconnect features (e.g., conductive lines), additional intermetallic dielectric layers (e.g., ESL and ILD layers), other suitable features, or a combination thereof.

[0055] While not intended to be limiting, one or more embodiments of the present invention provide numerous benefits to semiconductor devices and their formation. For example, the present invention provides a method for forming a GAA device comprising an NFET and a PFET, wherein the channel region of the NFET and the channel region of the PFET comprise different materials (a Si layer and a SiGe layer, respectively) configured to adjust the WF of the respective metal gate stacks in the NFET and the PFET, thereby allowing the V of the device to be adjusted using WFM layers of the same composition. t , to simplify the manufacturing process associated with metal gate formation. In this embodiment, the channel regions of the NFET and PFET are formed by separate sheet formation processes that selectively remove the non-channel layers of the device (e.g., removing the SiGe layer relative to the Si layer for the NFET). In some embodiments, an additional SiGe layer with a lower Ge content is formed between the stack of channel layers and the substrate, and the SiGe layer is configured to protect the substrate from unintentional etching during the sheet release process. Embodiments of the disclosed method can be easily integrated into existing processes and technologies for manufacturing GAA FETs.

[0056] In one aspect, the present invention provides a semiconductor structure comprising: a first stack of semiconductor layers disposed above a semiconductor substrate, wherein the first stack of semiconductor layers comprises a first silicon germanium (SiGe) layer and a plurality of silicon (Si) layers disposed above the first SiGe layer, and wherein the Si layer is substantially free of Ge; and a second stack of semiconductor layers disposed adjacent to the first stack of semiconductor layers, wherein the second stack of semiconductor layers comprises a first SiGe layer and a plurality of second SiGe layers disposed above the first SiGe layer, and wherein the first SiGe layer and the second SiGe layer have different compositions. In this embodiment, the semiconductor structure further comprises a first metal gate stack interleaved with the first stack of semiconductor layers to form a first device and a second metal gate stack interleaved with the second stack of semiconductor layers to form a second device, wherein the first device and the second device have different conductivity types.

[0057] In the above semiconductor structure, the first stack of semiconductor layers includes a first number of silicon layers, and the second stack of semiconductor layers includes a second number of silicon germanium layers, and the first number is greater than the second number.

[0058] In the above-mentioned semiconductor structure, the first stack of semiconductor layers includes a first number of silicon layers, and the second stack of semiconductor layers includes a second number of silicon germanium layers, and the first number is greater than the second number, and the first number is one more than the second number.

[0059] In the above semiconductor structure, the germanium content in the first silicon germanium layer is less than 20%.

[0060] In the above semiconductor structure, the germanium content in the first silicon germanium layer is less than 20%, and the germanium content in each of the second silicon germanium layers is at least 20%.

[0061] In the above-mentioned semiconductor structure, the first metal gate stack includes a first work function metal (WFM) layer, the second metal gate stack includes a second work function metal layer, and the first work function metal layer and the second work function metal layer have the same composition.

[0062] In the above-mentioned semiconductor structure, the first metal gate stack includes a first top portion arranged above the first stack portion of the semiconductor layer, the second metal gate stack includes a second top portion arranged above the second stack portion of the semiconductor layer, and the second top portion extends below the bottom surface of the first top portion.

[0063] In another aspect, the present invention provides a semiconductor structure comprising: a first semiconductor fin protruding from a substrate, wherein a channel region of the first semiconductor fin comprises a stack of silicon (Si) layers disposed above a first silicon germanium (SiGe) layer; and a second semiconductor fin protruding from the substrate, wherein a channel region of the second semiconductor fin comprises a stack of second SiGe layers disposed above the first SiGe layer, wherein an amount of Ge in the second SiGe layer is greater than an amount of Ge in the first SiGe layer, and wherein the number of second SiGe layers in the second semiconductor fin is one greater than the number of Si layers in the first semiconductor fin. In this embodiment, the semiconductor structure further comprises a first metal gate stack bonded to the channel region of the first semiconductor fin and a second metal gate stack bonded to the channel region of the second semiconductor fin.

[0064] In the above semiconductor structure, the stack of silicon layers has a first height, and the stack of second silicon germanium layers has a second height, and the first height is greater than the second height.

[0065] In the above-mentioned semiconductor structure, the stack of silicon layers has a first height, and the stack of second silicon germanium layers has a second height, and the first height is greater than the second height, and the difference between the first height and the second height is 5nm to 15nm.

[0066] In the above semiconductor structure, a work function of the first metal gate stack is the same as a work function of the second metal gate stack.

[0067] In the above-mentioned semiconductor structure, the work function of the first metal gate stack is the same as the work function of the second metal gate stack, wherein the first metal gate stack includes a first work function metal (WFM) layer, wherein the second metal gate stack includes a second work function metal layer, and wherein the first work function metal layer and the second work function metal layer have the same composition.

[0068] In the above semiconductor structure, the first silicon germanium layer includes at least 10% but less than 20% germanium.

[0069] In another aspect, the present invention provides a method of forming a semiconductor structure, the method including forming a first silicon-germanium-based (SiGe-based) layer over a substrate, forming a multilayer stack (ML) of alternating silicon (Si)-based layers and second SiGe-based layers over the first SiGe-based layer, wherein the first SiGe-based layer and the second SiGe-based layer have different compositions, and wherein the ML includes one more second SiGe-based layer than the Si layer, forming a first fin and a second fin adjacent to the first fin in the ML, forming a dummy gate stack over a channel region of the first fin and a channel region of the second fin, forming an n-type source / drain (S / D) feature in the first fin, and forming a p-type S / D feature in the second fin. In this embodiment, the method further includes: removing the dummy gate stacks between the n-type S / D components and between the p-type S / D components to form a first gate trench and a second gate trench, respectively, removing the second SiGe-based layer in the first fin to form a first opening between the Si-based layers, removing the Si-based layer in the second fin to form a second opening between the second SiGe-based layers and between the second SiGe-based layer and the first SiGe-based layer, and forming a metal gate stack in the first gate trench, the second gate trench, the first opening, and the second opening.

[0070] In the above method, wherein the first silicon-germanium-based layer includes a first amount of germanium, and each of the second silicon-germanium-based layers includes a second amount of germanium, and wherein the second amount is greater than the first amount.

[0071] In the above method, wherein the first silicon-germanium-based layer includes a first amount of germanium, and each of the second silicon-germanium-based layers includes a second amount of germanium, and wherein the second amount is greater than the first amount, wherein removing the second silicon-germanium-based layer includes selectively etching the second silicon-germanium-based layer relative to the first silicon-germanium-based layer and the silicon-based layer.

[0072] In the above method, wherein the first silicon-germanium-based layer includes a first amount of germanium, and each of the second silicon-germanium-based layers includes a second amount of germanium, and wherein the second amount is greater than the first amount, wherein removing the silicon-based layer includes selectively etching the silicon-based layer relative to the first silicon-germanium-based layer and the second silicon-germanium-based layer.

[0073] In the above method, wherein forming the metal gate stack causes the portion of the metal gate stack formed in the second gate trench to extend below the portion of the metal gate stack formed in the first gate trench.

[0074] In the above method, forming the metal gate stack includes: forming a gate dielectric layer in the first gate trench, the second gate trench, the first opening and the second opening; forming a work function metal (WFM) layer on the gate dielectric layer, wherein a first portion of the work function metal layer is formed in the first gate trench and the first opening, wherein a second portion of the work function metal layer is formed in the second gate trench and the second opening, and wherein the first portion and the second portion have the same composition; and forming a body conductive layer above the work function metal layer.

[0075] In the above method, before forming the dummy gate stack, the method further includes: forming an isolation structure over the substrate to separate the first fin from the second fin; and forming a dielectric fin over the isolation structure and between the first fin and the second fin.

[0076] The features of several embodiments have been summarized above so that those skilled in the art can better understand aspects of the present invention. Those skilled in the art will appreciate that they can easily use the present invention as a basis to design or modify other processes and structures for implementing the same purpose and / or achieving the same advantages as the embodiments described herein. Those skilled in the art will also appreciate that such equivalent configurations do not depart from the spirit and scope of the present invention, and that they may make various changes, substitutions, and modifications herein without departing from the spirit and scope of the present invention.

Claims

1. A semiconductor structure comprising: a first stack of semiconductor layers disposed over a semiconductor substrate, wherein the first stack of semiconductor layers comprises a first silicon germanium (SiGe) layer and a plurality of silicon (Si) layers disposed over the first silicon germanium layer, and wherein the silicon layers do not contain germanium; a second stack of semiconductor layers disposed adjacent to the first stack of semiconductor layers, wherein the second stack of semiconductor layers includes the first silicon germanium layer and a plurality of second silicon germanium layers disposed above the first silicon germanium layer, and wherein the first silicon germanium layer and the second silicon germanium layers have different compositions, wherein the composition of germanium in the first silicon germanium layer is less than 20%, and wherein the composition of germanium in at least one of the second silicon germanium layers is at least 20%; a first metal gate stack interleaved with the first stack of semiconductor layers to form a first device; and A second metal gate stack is interleaved with the second stack of semiconductor layers to form a second device, wherein the first device and the second device have different conductivity types.

2. The semiconductor structure according to claim 1, wherein: The first stack of semiconductor layers includes a first number of silicon layers and the second stack of semiconductor layers includes a second number of silicon germanium layers, and wherein the first number is greater than the second number.

3. The semiconductor structure according to claim 2, wherein: The first number is one greater than the second number.

4. The semiconductor structure according to claim 1, wherein: The second metal gate stack includes a dielectric layer encapsulating each second silicon germanium layer such that the dielectric layer physically contacts each second silicon germanium layer.

5. The semiconductor structure according to claim 4, wherein: The dielectric layer includes an interface layer and a high-k dielectric layer.

6. The semiconductor structure according to claim 1, wherein: The first metal gate stack includes a first work function metal (WFM) layer, wherein the second metal gate stack includes a second work function metal layer, and wherein the first work function metal layer and the second work function metal layer have the same composition.

7. The semiconductor structure according to claim 1, wherein: The first metal gate stack includes a first top portion disposed above the first stack of semiconductor layers, wherein the second metal gate stack includes a second top portion disposed above the second stack of semiconductor layers, and wherein the second top portion extends below a bottom surface of the first top portion.

8. A semiconductor structure comprising: a first semiconductor fin protruding from the substrate, wherein a channel region of the first semiconductor fin includes a stack of silicon (Si) layers disposed above a first silicon germanium (SiGe) layer; a second semiconductor fin protruding from the substrate, wherein a channel region of the second semiconductor fin includes a stack of second silicon germanium layers disposed above the first silicon germanium layer, wherein an amount of germanium in the second silicon germanium layer is greater than an amount of germanium in the first silicon germanium layer, and wherein a number of the second silicon germanium layers in the second semiconductor fin is one less than a number of the silicon layers in the first semiconductor fin; a first metal gate stack coupled to a channel region of the first semiconductor fin; and A second metal gate stack is bonded to the channel region of the second semiconductor fin, wherein the second metal gate stack includes a dielectric layer encapsulating each second silicon germanium layer such that the dielectric layer physically contacts each second silicon germanium layer.

9. The semiconductor structure according to claim 8, wherein: The stack of silicon layers has a first height and the stack of second silicon germanium layers has a second height, and wherein the first height is greater than the second height.

10. The semiconductor structure according to claim 9, wherein: A difference between the first height and the second height is 5 nm to 15 nm.

11. The semiconductor structure according to claim 8, wherein: A work function of the first metal gate stack is the same as a work function of the second metal gate stack.

12. The semiconductor structure according to claim 11, wherein: The first metal gate stack includes a first work function metal (WFM) layer, wherein the second metal gate stack includes a second work function metal layer, and wherein the first work function metal layer and the second work function metal layer have the same composition.

13. The semiconductor structure according to claim 8, wherein: The first silicon germanium layer includes at least 10% but less than 20% germanium.

14. A method of forming a semiconductor structure, comprising: forming a first silicon germanium-based (SiGe-based) layer over the substrate; forming a multilayer stack (ML) of alternating silicon-based (Si-based) layers and second silicon-germanium-based layers over the first silicon-germanium-based layer, wherein the first silicon-germanium-based layers and the second silicon-germanium-based layers have different compositions, and wherein the multilayer stack includes one less second silicon-germanium-based layer than the silicon-based layers, wherein the composition of germanium in the first silicon-germanium-based layers is less than 20%, and wherein the composition of germanium in at least one of the second silicon-germanium-based layers is at least 20%; forming a first fin and a second fin adjacent to the first fin in the multi-layer stack; forming a dummy gate stack over a channel region of the first fin and a channel region of the second fin; forming an n-type source / drain (S / D) feature in the first fin; forming a p-type source / drain feature in the second fin; removing the dummy gate stack between the n-type source / drain components and between the p-type source / drain components to form a first gate trench and a second gate trench, respectively; removing the second silicon germanium-based layer in the first fin to form a first opening between the silicon-based layers; removing the silicon-based layer in the second fin to form a second opening between the second silicon-germanium-based layer and between the second silicon-germanium-based layer and the first silicon-germanium-based layer; and A metal gate stack is formed in the first gate trench, the second gate trench, the first opening, and the second opening.

15. The method according to claim 14, wherein: The first silicon-germanium-based layer includes a first amount of germanium, and each of the second silicon-germanium-based layers includes a second amount of germanium, and wherein the second amount is greater than the first amount.

16. The method according to claim 15, wherein: Removing the second silicon-germanium-based layer includes selectively etching the second silicon-germanium-based layer relative to the first silicon-germanium-based layer and the silicon-based layer.

17. The method according to claim 15, wherein: Removing the silicon-based layer includes selectively etching the silicon-based layer relative to the first silicon-germanium-based layer and the second silicon-germanium-based layer.

18. The method according to claim 14, wherein: Forming the metal gate stack causes the portion of the metal gate stack formed in the second gate trench to extend below the portion of the metal gate stack formed in the first gate trench.

19. The method according to claim 14, wherein: Forming the metal gate stack includes: forming a gate dielectric layer in the first gate trench, the second gate trench, the first opening, and the second opening; forming a work function metal (WFM) layer on the gate dielectric layer, wherein a first portion of the work function metal layer is formed in the first gate trench and the first opening, wherein a second portion of the work function metal layer is formed in the second gate trench and the second opening, and wherein the first portion and the second portion have the same composition; and A bulk conductive layer is formed over the work function metal layer.

20. The method of claim 14, further comprising, before forming the dummy gate stack: forming an isolation structure over the substrate to separate the first fin from the second fin; and A dielectric fin is formed over the isolation structure and between the first fin and the second fin.