Semiconductor device and method of forming the same

CN122909887APending Publication Date: 2026-10-13TAIWAN SEMICONDUCTOR MANUFACTURING CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610632678.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2025-06-06
Filing Date
2026-05-09
Publication Date
2026-10-13

AI Technical Summary

Technical Problem

然而这种调整可能降低偶极强度而造成不平衡的临界电压调整

Benefits of technology

[0088]本发明实施例提供多种优点以用于n型场效晶体管与p型场效晶体管的临界电压调整,但不限于此。优点的一例为采用相同的p型偶极材料以同时调整n型场效晶体管与p型场效晶体管。优点的另一例为主要注入p型偶极材料于界面层中,而非高介电常数的介电层中,以改善n型场效晶体管的临界电压调整。优点的另一例采用n型偶极以进一步调整n型场效晶体管,并采用不同的p型偶极调整p型场效晶体管。在此方式中,可达多临界电压装置所用的较大调整范围。

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122909887A_ABST
    Figure CN122909887A_ABST
Patent Text Reader

Abstract

A method of forming a semiconductor device includes forming a first active region in an n-type transistor region on a substrate; forming a second active region in a p-type transistor region on the substrate; depositing a gate dielectric layer to cover a plurality of semiconductor channels of the first active region and the second active region, the gate dielectric layer including an interface layer on the semiconductor channels and a high-k dielectric layer on the interface layer; depositing a p-type dipole layer on the gate dielectric layer in the n-type transistor region and the p-type transistor region; depositing a fill layer on the p-type dipole layer; and performing a thermal treatment to drive a plurality of dopants in the p-type dipole layer into the gate dielectric layer. A first concentration of the dopants in the interface layer after the thermal treatment is greater than a second concentration of the dopants in the high-k dielectric layer.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to semiconductor devices, and more particularly to a doped layer formed between an interface layer and a dielectric layer with a high dielectric constant. Background Technology

[0002] The electronics industry continues to demand smaller, faster electronic devices capable of supporting a large number of complex functions simultaneously. To meet these demands, the ongoing trend in the integrated circuit (IC) industry is to manufacture low-cost, high-performance, and low-power ICs. The primary method for achieving these goals is to reduce IC size (such as the minimum IC architecture size), thereby improving production capacity and reducing related costs. However, size reduction also increases the complexity of IC manufacturing processes. Therefore, for IC devices and their performance to continue to advance, IC manufacturing processes and technologies must also advance in a similar manner.

[0003] One advancement is the development of multi-critical-voltage devices, including varying the metal gate thickness or material to generate multiple critical voltages across the same and / or different types of devices. Dipole dopants can also be integrated into the gate dielectric layer to adjust the critical voltage. For example, to form devices with extremely low critical voltages (such as high-speed and / or low standby devices), n-type dipole dopants can be driven into the gate dielectric layer to lower the critical voltage of n-type transistors, such as n-type field-effect transistors. For p-type devices, p-type dipole dopants can be driven into the gate dielectric layer to lower the critical voltage of p-type transistors, such as p-type field-effect transistors. In some cases, it is necessary to increase the critical voltage to form devices with extremely low leakage currents (such as high-density memory or logic devices). In these cases, n-type and p-type transistors can each use dopants with opposite morphologies. However, this adjustment may reduce dipole strength, resulting in unbalanced critical voltage adjustments. Consequently, the adjustment range used to increase and decrease the critical voltage may differ significantly, thus limiting the required process tolerances.

[0004] Therefore, although the critical voltage adjustment method used in existing integrated circuit devices is generally suitable for the intended purpose, it cannot meet all the requirements. Summary of the Invention

[0005] An embodiment of the present invention relates to a method for forming a semiconductor device. The method includes forming a first active region in an n-type transistor region on a substrate; forming a second active region in a p-type transistor region on the substrate; depositing a gate dielectric layer to cover a plurality of semiconductor channels in the first and second active regions, the gate dielectric layer including an interface layer on the semiconductor channels and a high-dielectric-constant dielectric layer on the interface layer; depositing a p-type dipole layer on the gate dielectric layer in the n-type and p-type transistor regions; depositing a fill layer on the p-type dipole layer; performing a heat treatment to drive a plurality of dopants in the p-type dipole layer into the gate dielectric layer, wherein the dopants after the heat treatment have a first concentration in the interface layer greater than a second concentration in the high-dielectric-constant dielectric layer; removing the p-type dipole layer and the fill layer; and depositing an n-type gate metal on the gate dielectric layer in the n-type transistor region and depositing a p-type gate metal on the gate dielectric layer in the p-type transistor region.

[0006] Another embodiment relates to a method for forming a semiconductor device. The method includes forming an active region in an n-type transistor region on a substrate; depositing an interface layer to cover a plurality of semiconductor channels in the active region; depositing a high-dielectric-constant dielectric layer on the interface layer; depositing a zinc oxide dipole layer on the high-dielectric-constant dielectric layer; depositing a fill layer on the zinc oxide dipole layer, wherein the fill layer fills the space between adjacent semiconductor channels; performing a heat treatment to drive a plurality of zinc dopants in the zinc oxide dipole layer into the high-dielectric-constant dielectric layer and the interface layer, wherein the zinc dopants reduce the thickness of the interface layer; removing the zinc oxide dipole layer and the fill layer; and depositing an n-type gate metal on the high-dielectric-constant dielectric layer.

[0007] Another embodiment of the present invention relates to a semiconductor device. The semiconductor device includes a first stack and a second stack of multiple semiconductor channels located on a substrate; a first interface layer and a second interface layer respectively covering the first stack and the second stack of semiconductor channels; a first high-dielectric-constant dielectric layer and a second high-dielectric-constant dielectric layer respectively covering the first interface layer and the second interface layer; and a first gate work function metal and a second gate work function metal respectively located on the first high-dielectric-constant dielectric layer and the second high-dielectric-constant dielectric layer. The first gate work function metal and the second gate work function metal are work function metals with opposite morphologies. The first interface layer and the second interface layer are doped with a first concentration of p-type dopant, and the first high-dielectric-constant dielectric layer and the second high-dielectric-constant dielectric layer are doped with a second concentration of p-type dopant, wherein the first concentration is greater than the second concentration. Attached Figure Description

[0008] Figure 1 This is a device layout in the memory and logic device area having n-type field-effect transistors and p-type field-effect transistors in one embodiment of the present invention.

[0009] Figure 2This is a flowchart of part or all of the method for adjusting n-type field-effect transistors and p-type field-effect transistors using the same p-type dipole material in one embodiment of the present invention.

[0010] Figure 3A , 4A A, B, C, D, E, F, G, F, G, F, G, F, G, F, G, F, and E are examples of embodiments of the present invention, based on... Figure 2 Methods for fabricating and processing intermediate stages of semiconductor devices along Figure 1 The sectional view with section line A-A' in the figure.

[0011] Figure 3B , 4B 5B, 6B, 7B, 8B, 9B, 10B, and 11B are examples of one embodiment of the present invention, based on... Figure 2 Methods for fabricating and processing intermediate stages of semiconductor devices along Figure 1 The sectional view with section line B-B' in the figure.

[0012] Figure 12 This is a flowchart, in part or in whole, of a method for adjusting an n-type field-effect transistor and a p-type field-effect transistor using various p-type and n-type dipole materials, according to one embodiment of the present invention.

[0013] Figure 13A , 14A 15A, 16A, 17A, 18A, and 19A are embodiments of the present invention, based on Figure 12 Methods for fabricating and processing intermediate stages of semiconductor devices along Figure 1 The sectional view with section line C-C' in the figure.

[0014] Figure 13B , 14B 15B, 16B, 17B, 18B, and 19B are embodiments of the present invention, based on Figure 12 Methods for fabricating and processing intermediate stages of semiconductor devices along Figure 1 The sectional view with section line D-D' in the figure.

[0015] Figure 20 In one embodiment of the present invention, aluminum dopant is driven to form a dopant distribution profile in an n-type field-effect transistor and a p-type field-effect transistor.

[0016] Figure 21 In one embodiment of the present invention, zinc dopant is driven to form a dopant distribution profile in an n-type field-effect transistor.

[0017] Figure 22 This is an embodiment of the present invention, which describes the critical voltage regulation efficiency between a zinc-based p-type dipole and an aluminum-based p-type dipole.

[0018] Figure 23 In one embodiment of the present invention, the zinc-dominant dopant profile and the aluminum-dominant dopant profile are in the gate dielectric layer.

[0019] Explanation of reference numerals in the attached figures: A-A',B-B',C-C',D-D': Section lines CR: Channel Area PD: Pull-down transistor PG: Through-gate transistor PU: Pull-up transistor SDR: Source / Drain Region 100: Device 100a: n-type field-effect transistor 100b: p-type field-effect transistor 150a: Memory device area 150b: Logic Device Area 201: Substrate 203: Interface Layer 204: Active Zone 204a: Channel 204b: Source / Drain Epitaxial Structure 205: High dielectric constant dielectric layer 206: Isolation Structure 207: Gate dielectric layer 210a: n-type gate metal 210b: p-type gate metal 211: Gate spacer 213: Interlayer dielectric layer 215: Etching Stop Layer 216: Inner spacer 218: Gate structure 219: Gate Trench 250a, 250b: p-type dipole layers 252: n-type dipole layer 255, 257: Doped layer 275, 277: Patterned masking layers 280: Fill layer 600: Heat treatment 1000, 1100: Method 1002, 1004, 1006, 1008, 1010, 1012, 1014, 1016, 1102, 1104, 1106, 1108, 1110, 1112, 1114: Steps Detailed Implementation

[0020] The following detailed description is illustrated with accompanying drawings to facilitate understanding of various aspects of the invention. It is worth noting that the various structures are for illustrative purposes only and are not drawn to scale, as is customary in the art. In practice, the dimensions of various structures may be arbitrarily increased or decreased for clarity. It should also be emphasized that the drawings illustrate only general embodiments of the invention and should not be construed as limiting the scope of the embodiments, as the embodiments of the invention are equally applicable to other embodiments. Furthermore, the drawings may imply structures not specifically described in the detailed description.

[0021] The different embodiments or examples provided below can implement different structures of the embodiments of the present invention. The embodiments of specific components and arrangements are intended to simplify this disclosure and not to limit the invention. For example, a description of forming a first component on a second component includes direct contact between the two, or the two being separated by other additional components rather than in direct contact. Various embodiments of the invention may use the same reference numerals repeatedly for brevity, but elements with the same reference numerals in various embodiments and / or arrangements do not necessarily have the same correspondence.

[0022] In addition, spatial relative terms such as “below,” “below,” “lower,” “above,” “higher,” or similar terms are used to describe the relationship between some elements or structures in the accompanying drawings and other elements or structures. These spatial relative terms include different orientations of the device in use or operation, as well as the orientations described in the accompanying drawings. When the device is turned in a different orientation (rotated 90 degrees or other orientations), the spatial relative adjectives used will also be interpreted according to the orientation after the turn.

[0023] Furthermore, when numerical values ​​or ranges are described using terms such as "about," "approximately," or similar expressions, they are intended to encompass values ​​within a reasonable range, such as those inherent to the manufacturing process that are considered by those skilled in the art. For example, based on known manufacturing tolerances related to the numerical value, the numerical value or range encompasses a reasonable range including the stated number, such as within + / - 10% of the stated number. For example, if the thickness of a material layer is about 5 nm and those skilled in the art know that the manufacturing tolerance for deposited material layers is 10%, the included size range is 4.5 nm to 5.5 nm. When the size of one structure is compared to another, terms such as "substantially identical," "basically the same," or "similar in size" can be understood as meaning that the difference between the compared structures is within + / - 10%. Furthermore, the disclosure of different structural dimensions may implicitly disclose the size ratios between different structures.

[0024] This invention relates to threshold voltage regulation in semiconductor devices, and more particularly to the selective use of p-type dipole materials to reduce the threshold voltage in p-type transistors (e.g., p-type field-effect transistors) and increase the threshold voltage in n-type transistors (e.g., p-type field-effect transistors). This configuration is necessary for forming extremely high-speed p-type metal-oxide-semiconductor (MOSFET) devices and extremely low-leakage-current n-type metal-oxide-semiconductor (NMOS) devices. However, p-type dipole materials typically suffer from insufficient threshold voltage regulation in NMOS devices, resulting in an unbalanced N / P p-type dipole threshold voltage regulation ratio. In other words, the regulation range required to reduce the threshold voltage of p-type MOSFETs is much larger than the regulation range required to increase the threshold voltage of n-type MOSFETs. Furthermore, when p-type dipoles are doped into gate oxides with high dielectric constants, there is a capacitance equivalent thickness penalty, causing a 0.5% to 1.5% speed reduction in the ring oscillator. Although n-type dipoles can be added to achieve individual regulation targets, they only effectively reduce the threshold voltage in n-type MOSFETs and increase the threshold voltage in p-type MOSFETs, rather than increasing the threshold voltage in n-type MOSFETs.

[0025] To address these and other issues, embodiments of the present invention provide methods and related structures in which the p-type dopant is primarily located in the interface layer, rather than in the high-dielectric-constant dielectric layer of the gate dielectric layer. The p-type dopant is selected such that its stable oxide formation energy in the high-dielectric-constant dielectric layer is higher than that in the interface layer. The p-type dopant may be composed of beryllium, zinc, and / or indium. These dopants provide a strong p-type dipole dopant layer at the interface between the interface layer and the high-dielectric-constant dielectric layer. In n-type field-effect transistors, since the p-type dipole dopant layer is primarily formed in and / or close to the interface layer, the anti-p-type dipole effect of dopants in the n-type work function metal can be reduced, thereby improving the equivalent capacitance thickness. Furthermore, the dielectric constant of the p-type dopant (such as zinc) is greater than that of the interface layer, but less than that of the high-dielectric-constant dielectric layer. If the p-type dopant replaces the interface layer (or a portion thereof), the effect of the equivalent capacitance thickness can be mitigated. Advantageously, the same p-type dopant can be deposited and driven simultaneously in both the n-type and p-type MOSFETs to achieve a balanced N / P critical voltage regulation ratio. In other device regions, different p-type dopant can be used to selectively regulate the p-type MOSFETs without regulating the n-type MOSFETs, and / or n-type dopant can be used to selectively regulate the n-type and / or p-type MOSFETs. In this approach, a larger regulation range can be achieved for multi-critical voltage devices.

[0026] This invention relates to, but is not limited to, fully wound gate field-effect transistors (FETs). A fully wound gate field-effect transistor refers to a transistor having a gate stack (gate and gate dielectric layer) surrounding a transistor channel, such as a metal-oxide-semiconductor (MOS) field-effect transistor device with vertically stacked horizontal nanowires or nanosheets of fully wound gates. Those skilled in the art will understand that this disclosure can be readily used as a basis to design or modify other structures to achieve the same purposes and / or advantages as the embodiments described herein. For example, the embodiments herein can also be implemented in MOS field-effect transistors, fin field-effect transistors, forked field-effect transistors, complementary field-effect transistors (CFETs), and / or combinations thereof.

[0027] Figure 1 This is a device layout in one embodiment of the present invention, having an n-type field-effect transistor 100a and a p-type field-effect transistor 100b in a memory device region 150a and a logic device region 150b. The device layout corresponds to a semiconductor structure, which forms a semiconductor device 100 or a portion thereof. Thus, the two terms are interchangeable. The semiconductor device 100 may be part of an integrated circuit chip or a system-on-a-chip, including various passive and active microelectronic devices such as resistors, capacitors, inductors, diodes, p-type field-effect transistors, n-type field-effect transistors, fin field-effect transistors, nanosheet field-effect transistors, nanowire field-effect transistors, other types of multi-gate field-effect transistors (such as forked field-effect transistors or complementary field-effect transistors), metal-oxide-semiconductor field-effect transistors, complementary metal-oxide-semiconductor transistors, bipolar junction transistors, laterally diffused metal-oxide-semiconductor transistors, high-voltage transistors, high-frequency transistors, other suitable components, or combinations thereof. In some embodiments, the apparatus includes or is comprised of: volatile or non-volatile memory such as volatile static random access memory, non-volatile random access memory, flash memory, electrically erasable and programmable read-only memory, electrically programmable read-only memory, other suitable memory types, or combinations thereof.

[0028] Semiconductor device 100 includes a substrate 201 defining an n-type transistor region for forming an n-type field-effect transistor 100a and a p-type transistor region for forming a p-type field-effect transistor 100b. The n-type transistor region (also considered as an n-type field-effect transistor region) includes p-type wells, which may be portions of the boron-doped substrate 201. The p-type transistor region (also considered as a p-type field-effect transistor region) includes n-type wells, which may be portions of the phosphorus- or arsenic-doped substrate 201. Although not illustrated, semiconductor device 100 may have interleaved n-type and p-type wells to form interleaved n-type field-effect transistors 100a and p-type field-effect transistors 100b. The n-type and p-type wells may be formed by ion implantation or thermal diffusion onto the substrate 201.

[0029] Semiconductor device 100 includes a memory device region 150a for forming a memory device and a logic device region 150b for forming a logic device. Each of these regions may each include one or more n-type field-effect transistors 100a and p-type field-effect transistors 100b. As described below, each individual n-type field-effect transistor 100a and p-type field-effect transistor 100b is defined by the channel region CR between the source / drain regions SDR and the gate on the channel region CR.

[0030] Semiconductor device 100 includes a plurality of active regions 204 (extending along the x-direction in their length direction) and a plurality of gate structures 218 (extending along the y-direction in their length direction). The active regions 204 form the channel region CR and source / drain region SDR of each of the n-type field-effect transistors 100a and p-type field-effect transistors 100b. The gate structures 218 form the gates of each of the n-type and p-type field-effect transistors 100a and 100b. The active regions 204 are formed on a substrate 201. The substrate 201 may be a silicon substrate or a substrate having other semiconductor materials such as germanium, silicon carbide, silicon-germanium, or diamond. The active regions 204 may protrude above the isolation structure 206 on the substrate 201 (see...). Figure 3B The isolation structure 206 may be a shallow trench isolation layer and may provide isolation between adjacent active regions 204 separated along the y-direction. The dielectric material used in the isolation structure 206 may include silicon oxide, silicon oxynitride, fluorosilicate glass, a low dielectric constant dielectric layer, a combination of the above, and / or other suitable materials.

[0031] Each active region 204 includes a channel region CR located between the source / drain regions SDR. The channel region CR includes vertically stacked channels 204a (see...). Figure 3A and 3B The source / drain SDR includes the source / drain epitaxial structure 204b (see...). Figure 3AThe channel 204a extends laterally along the X-direction between the source / drain epitaxial structures 204b. The source / drain epitaxial structure 204b may include an n-type source / drain structure to correspond to an n-type transistor region, or a p-type source / drain structure to correspond to a p-type transistor region. The source / drain epitaxial structure 204b may be formed by an epitaxial process, employing chemical vapor deposition techniques such as vapor phase epitaxy and / or ultra-high vacuum chemical vapor deposition, molecular beam epitaxy, other suitable epitaxial growth processes, or combinations thereof. The source / drain epitaxial structure 204b is doped with n-type dopants and / or p-type dopants. In some embodiments, for an n-type transistor (i.e., an n-type field-effect transistor 100a), the source / drain epitaxial structure 204b comprises silicon and may be doped with carbon, phosphorus, arsenic, other n-type dopants, or combinations thereof, such as forming a carbon-doped silicon epitaxial source / drain structure, a phosphorus-doped silicon epitaxial source / drain structure, or a carbon and phosphorus-doped silicon epitaxial source / drain structure. In some embodiments, for a p-type transistor (i.e., a p-type field-effect transistor 100b), the source / drain epitaxial structure 204b comprises silicon-germanium or germanium and may be doped with boron, other p-type dopants, or combinations thereof, such as forming a boron-doped silicon-germanium epitaxial source / drain structure.

[0032] Gate structures 218 (or gate stacks) are located on the channel regions CR of the active region 204. Each gate structure 218 is respectively bonded to and covers each channel 204a in the channel region CR (see Figure 11A and 11B As detailed in the accompanying drawings, each gate structure 218 includes a gate dielectric layer 207, with the gate located on the gate dielectric layer. The gate dielectric layer 207 includes an interface layer 203, and a high-dielectric-constant dielectric layer 205 located on the interface layer 203. The gate may include one or more conductive material layers, such as a capping layer, a work function metal layer, a barrier layer, a metal-filled layer, and / or other suitable conductive material layers. The work function layers may be the same or different, and may be n-type or p-type work function layers, depending on the corresponding transistor configuration. The gate dielectric layer includes a high-dielectric-constant dielectric material, such as a material with a dielectric constant greater than that of silicon oxide (approximately 3.9). The metal-filled layer includes suitable conductive materials such as aluminum, tungsten, and / or copper. The metal-filled layer may additionally or collectively include other metals, metal oxides, metal nitrides, other suitable materials, or combinations thereof.

[0033] like Figure 1As shown, a single gate structure 218 may extend in the y-direction and cross a single active region 204 or multiple active regions 204. Some gate structures 218 may extend across multiple active regions of different shapes (i.e., across the regions used by n-type field-effect transistors 100a and p-type field-effect transistors 100b), as shown. Some gate structures 218 may extend across a single active region or multiple active regions of the same shape (i.e., across multiple n-type field-effect transistors 100a or across multiple p-type field-effect transistors 100b, not shown). In the embodiment, the memory device region 150a includes a static random access memory device having one or more pull-up transistors PU, one or more pull-down transistors PD, and one or more bypass transistors PG. The pull-down transistors PD and bypass transistors PG are n-type field-effect transistors 100a, while the pull-up transistors PU are p-type field-effect transistors 100b. Figure 1 The arrangement of the pull-up transistor PU, pull-down transistor PD, and bypass transistor PG is shown (indicated on the metal gate structure 218 of each transistor). As shown, the pull-up transistor PU in the p-type transistor region and the pull-down transistor PD in the n-type transistor region can share the same gate structure 218. The gate structure 218 of the bypass transistor PG is only in the n-type transistor region and does not extend into the p-type transistor region.

[0034] In some regions of device 100, such as memory device region 150a, the threshold voltage of n-type field-effect transistor 100a needs to be increased to form a low-leakage-current, high-density bypass transistor PG and pull-down transistor PD device, and the threshold voltage of p-type field-effect transistor 100b needs to be decreased to form a high-speed, low-threshold pull-up transistor PU device. In embodiments of the present invention, the same p-type dipole material (e.g., zinc dopant) is applied to individual n-type and p-type transistor regions to achieve the above objective. However, in some regions of device 100, such as logic device region 150b, the threshold voltages of n-type and p-type field-effect transistors need to be decreased. The present invention can apply different p-type dipole materials (e.g., aluminum dopant) to individual p-type field-effect transistors 100b and apply n-type dipole materials (e.g., lanthanum dopant) to individual n-type field-effect transistors 100a to achieve the above objective.

[0035] Figure 2 This is a flowchart of part or all of a method 1000 for adjusting n-type field-effect transistor 100a and p-type field-effect transistor 100b using the same p-type dipole material, according to an embodiment of the present invention.

[0036] Figure 3A , 4A 5A, 6A, 7A, 8A, 9A, 10A, and 11A are based on Figure 2 Method 1000 is an intermediate stage in the fabrication and processing of semiconductor device 100, along with Figure 1The cross-sectional view is shown by section line A-A'. In the embodiment described, section line A-A' cuts through the active region 204, which is used to form a plurality of n-type field-effect transistors 100a in the memory device region 150a. It should be noted that in other embodiments, section line A-A' may cut through the logic device region 150b, or through both the memory device region 150a and the logic device region 150b.

[0037] Figure 3B , 4B 5B, 6B, 7B, 8B, 9B, 10B, and 11B are based on Figure 2 Method 1000 is an intermediate stage in the fabrication and processing of semiconductor device 100, along with Figure 1 A cross-sectional view of section line B-B'. In this embodiment, section line B-B' cuts through gate structure 218, which is used to form n-type field-effect transistors 100a and p-type field-effect transistors 100b in memory device region 150a. It should be noted that in other embodiments, section line B-B' may cut through logic device region 150b.

[0038] Figure 3A and 3B During the same production stage Figure 4A and 4B In the same production stage, and Figure 5A and 5B In the same production stage, this process continues. Method 1000 will be paired with... Figure 1 , 3A The following are explanations for 11A and 3B to 13B.

[0039] like Figure 3A and 3B As shown, step 1002 of method 1000 involves receiving or providing a semiconductor workpiece having an active region 204 formed in an n-type transistor region (for forming an n-type field-effect transistor 100a) and a p-type transistor region (for forming a p-type field-effect transistor 100b). Already paired Figure 1 This section describes various structures of semiconductor workpieces. Some structures will not be repeated for simplicity. As shown in the figure, active regions 204 are each formed on substrate 201. Each active region 204 includes a channel region CR sandwiched between source / drain epitaxial structures 204b, and the source / drain epitaxial structures 204b are formed in the source / drain region SDR. Each active region 204 includes individual stacks of semiconductor channels 204a within the channel region CR. Figure 3A The single n-type active region 204 shown has two channel regions CR sandwiched between the source / drain epitaxial structure 204b to form two n-type field-effect transistors 100a. Figure 3B The adjacent n-type active region 204 and p-type active region 204 shown are isolated from each other by an isolation structure 206.

[0040] During this fabrication stage, the gate trench 219 exposes various stacks of semiconductor channels 204a. For example, prior to this fabrication stage, a process to remove dummy gates and release channels can be performed to remove the dummy gates and dummy semiconductor layers, thereby forming the gate trench 219 and exposing the stack of suspended semiconductor channels 204a. Note that the upper and side surfaces of the substrate portion of the channel region are also exposed (see...). Figure 3B Although the substrate portion is not suspended, it protrudes from the substrate 201 above the isolation structure 206 and also serves as a semiconductor channel 204a.

[0041] exist Figure 3A In the cross-sectional view, the semiconductor workpiece includes an inner spacer 216 in the channel region CR, which is located vertically between semiconductor channels 204a along the sidewalls of the source / drain epitaxial structure 204b. The inner spacer 216 provides isolation between the gate and the source / drain and may comprise a dielectric material formed of silicon, oxygen, carbon, nitrogen, other suitable materials, or combinations thereof (e.g., silicon oxide, silicon nitride, silicon oxynitride, silicon carbide, or silicon carbonitride). The channel region CR also includes a gate spacer 211 vertically aligned with the inner spacer 216 and landing on the topmost semiconductor channel 204a. The gate spacer 211 may be composed of silicon oxide, silicon nitride, silicon carbonitride, silicon carbonitride, silicon carbonitride, metal nitride, or a suitable dielectric material. The gate spacer 211 provides isolation for the subsequently formed gate.

[0042] like Figure 3A As shown, the interlayer dielectric layer 213 is located on the source / drain epitaxial structure 204b. The interlayer dielectric layer 213 can be formed by first compliantly depositing a dielectric pad such as an etch stop layer 215 (the deposition method can be chemical vapor deposition, atomic layer deposition, or other suitable processes), followed by the interlayer dielectric layer 213 on the etch stop layer 215. A planarization process such as chemical mechanical polishing can then be performed to planarize the upper surfaces of the interlayer dielectric layer 213 and the etch stop layer 215. The etch stop layer 215 may include silicon nitride, and the interlayer dielectric layer 213 may include dielectric materials such as silicon oxide, silicon nitride, silicon oxynitride, oxides formed from tetraethoxysilanes, phosphosilicate glass, borosilicate glass, low dielectric constant dielectric materials, other suitable dielectric materials, or combinations thereof. Exemplary low dielectric constant dielectric materials include silicon or polymer-based materials such as fluorosilicate glass, carbon-doped silicon oxide, Black Diamond® (available from Applied Materials, Santa Clara, California), degel, aerogel, amorphous fluorinated carbon, parylene, benzocyclobutene, SiLK (available from Dow Chemical, Midland, Michigan), polyimide, or combinations thereof.

[0043] like Figure 4A and 4B As shown, in step 1004 of method 1000, a gate dielectric layer 207 is formed to cover the respective semiconductor channel 204a of the active region 204. The gate dielectric layer 207 includes an interface layer 203 and a high dielectric constant dielectric layer 205 formed on the interface layer 203.

[0044] In the illustrated embodiment, the interface layer 203 is formed by an oxidation process such as thermal oxidation or chemical oxidation, which oxidizes the surface of the exposed semiconductor channel 204a in the gate trench 219. Thus, the interface layer 203 is not formed on the surface of the isolation structure 206. In other embodiments, the interface layer 203 is formed by a deposition process such as atomic layer deposition or chemical vapor deposition, which deposits on the exposed surfaces of the semiconductor channel 204a in the gate trench 219 and the isolation structure 206. The interface layer 203 may include silicon oxide, silicon dioxide, silicon oxynitride, or other suitable materials. In the embodiment described, the interface layer 203 includes silicon dioxide.

[0045] A high-dielectric-constant dielectric layer 205 is compliantly deposited on the interface layer 203 in the gate trench 219, thereby covering the interface layer 203 and the semiconductor channel 204a. For example... Figure 4B As shown, a high-dielectric-constant dielectric layer 205 is also deposited on the isolation structure 206. The high-dielectric-constant dielectric layer 205 can be formed by chemical oxidation, thermal oxidation, atomic layer deposition, chemical vapor deposition, and / or other suitable methods. The high-dielectric-constant dielectric layer 205 comprises a high-dielectric-constant dielectric material with a dielectric constant greater than that of silicon oxide (approximately 3.9). High dielectric constant dielectric materials may include hafnium oxide, hafnium silicon oxide, hafnium silicate, hafnium silicon oxynitride, hafnium lanthanum oxide, hafnium tantalum oxide, hafnium titanium oxide, hafnium zirconium oxide, hafnium aluminum oxide, zirconium oxide, zirconium dioxide, zirconium silicate, aluminum oxide, aluminum silicon oxide, aluminum oxide, titanium dioxide, lanthanum oxide, lanthanum silicon oxide, tantalum trioxide, tantalum pentoxide, yttrium oxide, strontium titanate, barium zirconium oxide, barium titanate, barium strontium titanate, silicon nitride, hafnium oxide-aluminum oxide alloy, other suitable high dielectric constant dielectric materials, or combinations thereof. In the described embodiment, the high dielectric constant dielectric layer 205 comprises hafnium oxide.

[0046] like Figure 5A and 5B As shown, in step 1006 of method 1000, a p-type dipole layer 250a is deposited on the gate dielectric layer 207 in the n-type transistor region and the p-type transistor region. The p-type dipole layer 250a is compliantly deposited on the high-dielectric-constant dielectric layer 205 in the gate trench 219, thereby covering the high-dielectric-constant dielectric layer 205, the interface layer 203, and the semiconductor channel 204a. Figure 5BAs shown, a p-type dipole layer 250a is also deposited on the lower surface of a dielectric layer 205 with a high dielectric constant.

[0047] As described herein, the p-type dipole layer 250a is used to increase the threshold voltage used by the n-type transistor (i.e., n-type field-effect transistor 100a) and decrease the threshold voltage used by the p-type transistor (i.e., p-type field-effect transistor 100b). To achieve a balanced adjustment ratio ΔN / ΔP, the p-type dipole layer 250a includes a first p-type dipole material, the choice of which specifically improves the adjustment range of the n-type field-effect transistor 100a. The first p-type dipole material has a higher stable oxide formation energy in the high-dielectric-constant dielectric layer 205 than in the interface layer 203. As detailed below, this allows more first p-type dipole material to penetrate the interface layer 203 than to penetrate the high-dielectric-constant dielectric layer 205. A stronger p-type dipole needs to be formed in the n-type field-effect transistor 100a. In these embodiments, the first p-type dipole material includes zinc, beryllium, or indium. In these embodiments, the p-type dipole layer 250a includes oxides, nitrides, carbides, or metal compounds of zinc, beryllium, or indium. For example, the p-type dipole layer 250a may include zinc oxide, zinc nitride, or zinc carbide. In another example, the p-type dipole layer 250a may include beryllium oxide, beryllium nitride, or beryllium carbide. In yet another example, the p-type dipole layer 250a may include indium oxide, indium nitride, or indium carbide.

[0048] like Figure 6A and 6B As shown, in step 1008 of method 1000, a fill layer 280 is deposited on the p-type dipole layer 250a. The fill layer 280 serves as a gap filler protective layer in preparation for subsequent annealing steps. The deposition method of the fill layer 280 can be any suitable method, such as chemical vapor deposition or atomic layer deposition. The fill layer 280 completely fills the remaining gaps between the semiconductor channels 204a, such that the fill layer 280 completely fills the gate trench 219 below the topmost semiconductor channel 204a. In this embodiment, the fill layer 280 comprises aluminum. The peripheral thickness of the fill layer 280 (i.e., the portion surrounding the p-type dipole layer 250a) is greater than the peripheral thickness of the p-type dipole layer 250a (i.e., the portion surrounding the high-dielectric-constant dielectric layer 205). In one embodiment, the peripheral thickness of the fill layer 280 is between about 20 Å and about 40 Å. In one embodiment, the peripheral thickness of the p-type dipole layer 250a is between about 2 Å and about 10 Å. The thickness of the p-type dipole layer 250a is only sufficient to drive the first p-type dipole material (not the material of the fill layer 280) into the gate dielectric layer.

[0049] In other embodiments, the p-type dipole layer 250a completely fills the gaps between the semiconductor channels 204a, thus omitting step 1006 (i.e., the filler layer 280 is not required). In this example, the peripheral thickness of the p-type dipole layer 250a is between about 22 Å and about 50 Å. It should be noted that there may be trade-offs between including the filler layer 280 and omitting the filler layer 280. For example, depositing the filler layer 280 may be a lower-cost and / or faster process than depositing the p-type dipole layer 250a. In this way, only the minimum amount of p-type dipole layer 250a required is deposited, with the remainder filled by the filler layer 280. On the other hand, omitting the filler layer 280 eliminates the need for additional deposition steps. Furthermore, there is no risk that depositing an excessively thin p-type dipole layer 250a could result in insufficient dopant material during subsequent annealing steps.

[0050] like Figure 7A and 7B and Figure 8A and 8B As shown, step 1010 of method 1000 involves patterning the p-type dipole layer 250a to selectively remove a portion of the p-type dipole layer 250a on the gate dielectric layer 207. It should be noted that step 1010 may be performed as needed in examples where patterning is not required. For example, it may be necessary to drive the first p-type dipole dopant to any location where the p-type dipole layer 250a is deposited. However, method 1000 includes step 1010 to provide additional threshold voltage adjustment flexibility.

[0051] Step 1010 can be a patterning process, which includes photolithography and etching. For example... Figure 7A and 7B As shown, a photolithography process is performed to form a patterned mask layer 275, which selectively covers one of the two channel regions CR of the n-type field-effect transistor region used to form the n-type field-effect transistor 100a, while maintaining the other of the two channel regions CR exposed. Figure 7A At the same time, it also covers the channel region in the p-type field-effect transistor region used to form the p-type field-effect transistor 100b. Figure 7B In other words, the patterned mask layer 275 covers the two active regions 204 cut by the section line B-B', while maintaining the exposure of the left side portion of the active region 204 cut by the section line A-A' (see...). Figure 1 ).

[0052] The patterned mask layer 275 can be formed by transferring a pattern from a patterned photoresist (i.e., a soft mask) onto the patterned mask layer 275. An anisotropic etching step can be used to transfer the pattern. In some embodiments, the etching step performs a dry etching process such as plasma etching, reactive ion etching, another suitable dry etching method, or a combination thereof. In summary, the result of the dry etching process can form the patterned mask layer 275. In other embodiments, the patterned mask layer 275 itself is a patterned photoresist (i.e., a soft mask).

[0053] In this embodiment, the patterned mask layer 275 comprises a bottom anti-reflective coating. The bottom anti-reflective coating can be formed by spin coating, which has a lower process cost than chemical vapor deposition. In some examples, the bottom anti-reflective coating may include a silicon-containing polymer, a carbon-containing polymer, or spin-coated carbon.

[0054] like Figure 8A and 8B As shown, a patterned mask layer 275 is used as an etching mask for the etching process. In one embodiment, the etching process is a wet etching process to selectively remove the p-type dipole layer 250a and the filler layer 280 from exposed areas (such as one of the two channel regions CR of an n-type field-effect transistor region). The etching process may be performed using ammonium hydroxide. The patterned mask layer 275 can then be removed using any suitable method. In some embodiments, the method for removing the patterned mask layer 275 may include an ashing step at a temperature of about 100°C to about 300°C. If the temperature is too low, such as below about 100°C, residues from the patterned mask layer 275 may cause problems with subsequent deposition. If the temperature is too high, such as above about 300°C, the process cost may outweigh any advantages.

[0055] In this process stage, the p-type dipole layer 250a can be effectively patterned. For example... Figure 1 As shown, the p-type dipole layer 250a is currently present in some, but not all, portions of the n-type field-effect transistor region in the memory device region 150a, while the p-type dipole layer 250a is present in all portions of the p-type field-effect transistor region in the memory device region 150a. It should be noted that in other embodiments where no patterning is performed, the p-type dipole layer 250a extends across all portions of the memory device region 150a. It should also be noted that in other embodiments where subsequent patterning is performed, the p-type dipole layer 250a is present in some, but not all, portions of the p-type field-effect transistor region.

[0056] like Figure 9A and 9BAs shown, step 1012 of method 1000 involves a heat treatment 600 (or annealing process) to drive the dopant of the p-type dipole layer 250a into the gate dielectric layer 207. The heat treatment 600 provides the energy required for the elements of the p-type dipole layer 250a to migrate into the high-dielectric-constant dielectric layer 205 and interface layer 203. In other words, it drives the first p-type dipole material of the p-type dipole layer 250a into the high-dielectric-constant dielectric layer 205 and interface layer 203 of the gate dielectric layer 207. It should be noted that due to the patterning process in step 1010, the regions without the p-type dipole layer 250a remain unchanged.

[0057] The heat treatment 600 can be performed using any suitable method. For example, the heat treatment 600 can perform immersion annealing, peak annealing, laser annealing, other processing methods, or combinations thereof. In some embodiments, the temperature at which the heat treatment 600 is performed is from about 500°C to about 1200°C. If the temperature is too low, the dipole migration efficiency may be too low. If the temperature is too high, migration becomes difficult to control.

[0058] It has been found that for successful formation of a p-type dipole (as envisioned herein), the oxygen areal density σ of the dipole layer (e.g., p-type dipole layer 250a) needs to be higher than the oxygen areal density σ of the interface layer (e.g., interface layer 203). In this example, interface layer 203 comprises silicon oxide, therefore the ratio σ / σSiO2 > 1 must be followed. Furthermore, the electronegativity should be between the electronegativity of the high-dielectric-constant dielectric layer atoms (e.g., from the high-dielectric-constant dielectric layer 205) and the electronegativity of the interface layer atoms (e.g., from interface layer 203). In this embodiment, the electronegativity required for the high-dielectric-constant dielectric layer atoms is between about 1.3 and about 1.5 (e.g., for hafnium). In this embodiment, the electronegativity required for the interface layer atoms is about 1.8 (e.g., for silicon). Thus, the electronegativity of the dipole layer (e.g., p-type dipole layer 250a) in one embodiment needs to be between about 1.6 and about 1.8.

[0059] However, further investigation revealed that merely meeting the above conditions is insufficient to form a strong p-type dipole in a specific n-type transistor (i.e., n-type field-effect transistor 100a). For example, while aluminum as a p-type dipole material meets the above conditions, using aluminum to increase the critical voltage in n-type field-effect transistor 100a may have a counterproductive effect, unexpectedly reducing the required effective work function. For instance, the n-type gate metal in n-type field-effect transistor 100a may include aluminum, which causes oxygen atoms to diffuse outward from the high-dielectric-constant dielectric layer 205. That is, the n-type gate metal itself may include aluminum, which can attract oxygen from the high-dielectric-constant dielectric layer 205, thereby unexpectedly reducing the effective work function. This is not the case for p-type field-effect transistor 100b, where the p-type gate metal typically acts as an oxygen source and does not contain aluminum. This phenomenon also causes inconsistencies in the tuning between n-type and p-type field-effect transistors. At least for these reasons, the critical voltage tuning of p-type dipole n-type field-effect transistors is less favorable than that of p-type field-effect transistors with p-type dipoles.

[0060] As described here and in combination Figures 21 to 23 The aforementioned problem can be addressed by additional conditions, where the energy required for the formation of a stable oxide of the first p-type dipole material in the high-dielectric-constant dielectric layer 205 is higher than that in the interface layer 203. In this approach, the p-type dipole material driven into the interface layer 203 after annealing, such as heat treatment 600, is greater than that driven into the high-dielectric-constant dielectric layer 205. This may reduce oxygen diffusion from the high-dielectric-constant dielectric layer 205 to the interface layer 203, since the p-type dipoles are primarily located on the interface layer side, thus maintaining a higher oxygen concentration in the high-dielectric-constant dielectric layer 205. Furthermore, by pushing the dopant closer to the interface layer side, the comparison of potential decrease is no longer between the original high-dielectric-constant material (i.e., hafnium) and the substituted dipole material (i.e., aluminum), but rather between the dipole material (i.e., zinc) and the interface layer material (i.e., silicon). In this approach, a higher potential decrease can be maintained. Ultimately, by pushing the dopant closer to the interface layer, the separation distance can be increased, thereby reducing the effect of oxygen diffusing outward into the n-type gate metal.

[0061] In other embodiments, p-type dipole dopants may be driven into the interface layer 203 before the high-k dielectric layer 205 is formed on the interface layer 203. While this may provide a larger initial distribution of p-type dopants in the interface layer 203, subsequent thermal processing during semiconductor fabrication may cause these dopants to diffuse further into the high-k dielectric layer 205, resulting in more p-type dopants in the high-k dielectric layer 205 than in the interface layer 203. This problem can be avoided by selecting p-type dopants with a higher stable oxide formation energy in the high-k dielectric layer 205 than in the interface layer 203. In this manner, annealing of the interface layer 203 is not required before depositing the high-k dielectric layer 205 because the dopant distribution depends on its own oxide formation energy. In other words, the rate at which dopants are driven into the interface layer 203 is greater than the rate at which dopants are driven into the high-k dielectric layer 205. As described herein, these p-type dopants can be considered as first p-type dopant materials. These dopants may include zinc, beryllium, indium, or combinations thereof. For example, zinc has a stable oxide formation energy of 3.25 eV in a high-dielectric-constant dielectric layer 205 with hafnium and 1.34 eV in an interface layer 203 with silicon. Meanwhile, aluminum has a stable oxide formation energy of 1 eV in a high-dielectric-constant dielectric layer 205 with hafnium and 3.41 eV in an interface layer 203 with silicon. Therefore, zinc is superior to aluminum in forming the p-type dipole layer 250a used in the n-type field-effect transistor 100a.

[0062] like Figure 10A and 10B As shown, step 1014 of method 1000 removes the p-type dipole layer 250a and the filler layer 280 (if present). Step 1014 may be performed using a wet etching process similar to the aforementioned wet etching process. As shown, the first p-type dopant material of the p-type dipole layer 250a is implanted into the annealed interface layer 203 (in certain regions). Although not shown, the first p-type dopant material can also be implanted into the high-dielectric-constant dielectric layer 205, although at a lower concentration than in the interface layer 203. The dopant material of the first p-type dopant material can form the doped layer 255 (see...). Figure 23 The doped layer 255 may be formed at or around the interface between the interface layer 203 and the high-dielectric-constant dielectric layer 205. The portion of the doped layer 255 replacing the interface layer 203 is larger than the portion of the high-dielectric-constant dielectric layer 205 replaced by the doped layer 255. For example... Figure 23As shown, the reduction in the thickness of the interface layer by the doped layer 255 is greater than the reduction in the thickness of the high-dielectric-constant dielectric layer 205 by the doped layer 255. In one embodiment, the atomic percentage of the dopant (such as zinc) in the doped layer 255 is between 0.1% and 10%. In one embodiment, the doped layer 255 comprises silicon, oxygen, hafnium, zinc, and aluminum. In one embodiment, the doped layer 255 comprises silicon, oxygen, hafnium, beryllium, and aluminum. In one embodiment, the doped layer 255 comprises silicon, oxygen, hafnium, indium, and aluminum.

[0063] like Figure 11A and 11B As shown, in step 1016 of method 1000, n-type gate metal 210a and p-type gate metal 210b are deposited on the gate dielectric layer 207 in the n-type transistor and p-type transistor regions, respectively. The deposition method for n-type gate metal 210a and p-type gate metal 210b can be any suitable process. The n-type and p-type gate metals (which can also be considered as gates) may each include a capping layer, a work function metal layer, and a fill metal layer. The capping layer may include titanium nitride, tantalum nitride, or other suitable materials, and its formation method may be a suitable deposition technique such as atomic layer deposition. The fill metal layer may include aluminum, tungsten, copper, other suitable metals, or metal alloys, and its deposition method may be physical vapor deposition or other suitable deposition techniques. Step 1016 forms individual gate structures 218 for individual n-type field-effect transistors 100a and p-type field-effect transistors 100b.

[0064] The work function metal layer included in the n-type gate metal 210a may differ from the work function metal layer of the p-type gate metal 210b. The choice of the work function metal depends on the field-effect transistor formed on the active region. For example, the n-type work function metal is a metal with a first work function to reduce the threshold voltage of the associated n-type field-effect transistor. For example, the work function of the n-type work function metal is less than or equal to about 4.2 eV. The p-type work function metal is a metal with a second work function to reduce the threshold voltage of the associated p-type field-effect transistor. For example, the work function of the p-type work function layer is greater than or equal to about 5.2 eV. The metal included in the n-type work function metal has a sufficiently low effective work function, such as titanium, aluminum, tantalum carbide, tantalum carbonitride, silicon tantalum nitride, or a combination thereof. The metal included in the p-type work function metal may have a sufficiently large effective work function, such as titanium nitride, tantalum nitride, ruthenium, molybdenum, tungsten, platinum, or a combination thereof. In this embodiment, the n-type gate metal 210a comprises an n-type work function metal such as titanium aluminum nitride or other aluminum-containing metals. The deposition method of the n-type work function metal can be global deposition to cover all n-type device regions, since these specific n-type regions have higher threshold voltages due to the annealed p-type dipole layer 250a as described herein. In this embodiment, the p-type gate metal 210b comprises a p-type work function metal such as titanium nitride or other aluminum-free metals. The deposition method of the p-type work function metal can be global deposition to cover all p-type device regions. In some embodiments, in addition to using or additionally employing work function metals to adjust the individual n-type field-effect transistors 100a and p-type field-effect transistors 100b, the fill metal itself can be specific to n-type and p-type.

[0065] Additional steps may be provided before, during, and after method 1000, and additional embodiments of method 1000 may substitute, replace, or omit some of the steps. Additional steps may include forming contacts of a device layer to connect to the source / drain epitaxial structure 204b and / or the gate structure 218. Additional steps may also include forming interconnect structures on the contacts of the device layer. The interconnect structures may include one or more interconnect layers having lines and vias buried in a dielectric layer. One or more interconnect layers connect the gates, sources, and drains of various transistors and other circuitry in device 100 to form part or all of an integrated circuit. Additional steps may also include forming a passivation layer on the interconnect layers.

[0066] like Figure 1 As shown, method 1000 is implemented in memory device region 150a, but not in logic device region 150b. However, in other embodiments, method 1000 is also applied to logic device region 150b, or to a mixed region containing portions of memory device region 150a and logic device region 150b. Figure 1In the illustrated embodiment, method 1000 may reduce the threshold voltage of one or more pull-up transistors PU used by one or more static random access memory devices in memory device region 150a. Simultaneously, method 1000 may increase the threshold voltage of one or more pull-down transistors PD used by one or more static random access memory devices. Due to step 1010, method 1000 does not increase the threshold voltage of one or more bypass transistors PG used by one or more static random access memory devices. However, in an embodiment where step 1010 is omitted, method 1000 may also increase the threshold voltage of one or more bypass transistors PG.

[0067] As described herein, method 1000 can be implemented to combine the formation of n-type dipoles and / or p-type dipoles on the same device 100. To demonstrate this idea, Figure 12 The flowchart of method 1100 of the combination method 1000 is shown. In one embodiment of the present invention, some or all of method 1100 employs a variety of p-type dipole materials and n-type dipole materials to adjust the critical voltage in n-type field-effect transistor 100a and p-type field-effect transistor 100b.

[0068] Figure 13A , 14A 15A, 16A, 17A, 18A, and 19A are embodiments of the present invention, based on Figure 12 Method 1100 involves the intermediate stages of fabricating and processing the semiconductor device 100 along... Figure 1 A cross-sectional view of section line C-C'. In the embodiment described, section line C-C' cuts through active region 204, which is used to form n-type field-effect transistor 100a in logic device region 150b. It should be noted that in other embodiments, section line C-C' may cut through memory device region 150a, or through both memory device region 150a and logic device region 150b.

[0069] Figure 13B , 14B 15B, 16B, 17B, 18B, and 19B are embodiments of the present invention, based on Figure 12 Method 1100 involves the intermediate stages of fabricating and processing the semiconductor device 100 along... Figure 1 A cross-sectional view of section line D-D'. In this embodiment, section line D-D' cuts through two separate gate structures 218, which are used to form n-type field-effect transistors 100a and p-type field-effect transistors 100b in the logic device region 150b. It should be noted that in other embodiments, section line D-D' may cut through the memory device region 150a.

[0070] Figure 13A and 13B During the same production stage Figure 14A and 14BIn the same production stage, and Figure 15A and 15B In the same production stage, this process is repeated. Method 1100 combination Figure 1 , 13A The steps of method 1100 and 13B to 20 are explained below. Method 1100 will be further explained in conjunction with the various steps of method 1000 and their corresponding figures.

[0071] like Figure 1 , Figure 4A and 4B ,and Figure 13A and 13B As shown, in step 1102 of method 1100, a gate dielectric layer 207 is formed on the first and second stacks of semiconductor channels 204a in the n-type transistor region (for forming the n-type field-effect transistor 100a), and on the third and fourth stacks of semiconductor channels 204a in the p-type transistor region (for forming the p-type field-effect transistor 100b). Step 1102 may conform to or include step 1004 of method 1000, which corresponds to... Figure 4A and 4B . Figure 4B The first and third stacks of semiconductor channels 204a in individual n-type transistor regions and p-type transistor regions can be displayed (e.g., Figure 4B (The left and right sides are stacked in the middle), and Figure 13B The second and fourth stacks of semiconductor channels 204a in individual n-type transistor regions and p-type transistor regions can be displayed (e.g., Figure 13B (The left and right sides are stacked). Figure 13A and 13B and Figure 4A and 4B Similarly, without restating the same structure to simplify the content.

[0072] like Figure 5A and 5B As shown, in step 1104 of method 1100, a first p-type dipole layer 250a is formed on the gate dielectric layer 207 of the first and third stacks of the semiconductor channels 204a in the n-type and p-type transistor regions, respectively. Step 1104 corresponds to step 1006 of method 1000. Step 1104 may further include step 1008 to form a fill layer 280 on the first p-type dipole layer 250a (see...). Figure 6A and 6B (and corresponding explanations).

[0073] like Figure 14A and 14B and Figure 15A and 15BAs shown, step 1106 of method 1100 forms a second stack of n-type dipole layer 252 in the semiconductor channel 204a in the n-type transistor region (as shown). Figure 15B The gate dielectric layer 207 is stacked on the left side of the structure. Step 1106 may include depositing a fill layer 280 to fill the gaps between semiconductor channels 204a, as previously described. The n-type dipole layer 252 may include lanthanum, scandium, erbium, strontium, yttrium, or a combination thereof. In this embodiment, the n-type dipole layer 252 includes lanthanum to reduce the threshold voltage of the n-type field-effect transistor 100a (such as the n-type field-effect transistor 100a in logic device region 150b). In some embodiments, the n-type dipole layer 252 may also be used to increase the threshold voltage of the p-type field-effect transistor 100b. However, in the embodiment described above, the n-type dipole layer 252 is removed from the p-type field-effect transistor region before thermal processing, as described below.

[0074] An n-type dipole layer 252 can be deposited across the entire area first. For example, Figure 14B The n-type dipole layer 252 shown is also deposited on the fourth stack (as shown on the right) of the semiconductor channel 204a in the p-type transistor region. However, as... Figure 15B As shown, step 1106 includes a patterning process to remove the n-type dipole layer 252 from the fourth stack (such as the right-hand stack) of semiconductor channel 204a. The patterning process includes photolithography and etching. Figure 15B For example, a photolithography process is performed to form a patterned mask layer 277, which selectively covers the channel region CR of the n-type field-effect transistor region (such as the second stack of semiconductor channels 204a) while maintaining exposure of the channel region CR in the p-type field-effect transistor region (such as the fourth stack of semiconductor channels 204a). In other words, the patterned mask layer 277 maintains exposure of the entirety of the two active regions 204 cut by section line D-D', and covers the left side portion of the active region 204 cut by section line C-C' (see...). Figure 1 ).

[0075] The patterned mask layer 277 can be formed by transferring a pattern from a patterned photoresist (i.e., a soft mask) onto the patterned mask layer 277. An anisotropic etching step can be used to transfer the pattern. In some embodiments, the etching step performs a dry etching process such as plasma etching, reactive ion etching, another suitable dry etching method, or a combination thereof. In summary, the result of the dry etching process can form the patterned mask layer 277. In other embodiments, the patterned mask layer 275 itself is a patterned photoresist (i.e., a soft mask).

[0076] In this embodiment, the patterned mask layer 277 comprises a bottom anti-reflective coating. The bottom anti-reflective coating can be formed by spin coating, which is less expensive than chemical vapor deposition. In some examples, the bottom anti-reflective coating may comprise a silicon-containing polymer, a carbon-containing polymer, or spin-coated carbon.

[0077] like Figure 15B As shown, a patterned mask layer 277 is used as an etching mask for the etching process. In one embodiment, the etching process is a wet etching process to selectively remove the n-type dipole layer 252 and filler layer 280 from exposed areas (such as the channel region CR of a p-type field-effect transistor). The etching process may be performed using ammonium hydroxide. The patterned mask layer 277 can then be removed using any suitable method. In some embodiments, the method for removing the patterned mask layer 277 may include an ashing step at a temperature of about 100°C to about 300°C. If the temperature is too low, such as below about 100°C, residues from the patterned mask layer 277 may cause problems with subsequent deposition. If the temperature is too high, such as above about 300°C, the process cost may outweigh any advantages.

[0078] In summary, this process stage can effectively pattern the n-type dipole layer 252. For example... Figure 1 As shown, the n-type dipole layer 252 is currently present in some portions, but not all portions, of the n-type field-effect transistor region. For example, the n-type dipole layer 252 is present only in the logic device region 150b, and not in the memory device region 150a (which at this stage has the aforementioned p-type dipole layer 250a). Furthermore, the patterning process ensures that the n-type dipole layer 252 is not present in any portion of the p-type field-effect transistor region. It is noteworthy that in other embodiments without patterning, the n-type dipole layer 252 may also be present in the p-type field-effect transistor region (such as the p-type field-effect transistor region of the logic device region 150b). It is noteworthy that in other embodiments with subsequent patterning, the n-type dipole layer 252 is present in some portions, but not all portions, of the n-type field-effect transistor region.

[0079] like Figure 16A and 16B As shown, step 1108 of method 1100 forms a second p-type dipole layer 250b in the fourth stack of the semiconductor channel 204a in the p-type transistor region (i.e., Figure 16BThe second p-type dipole layer 250b is stacked on the right side of the gate dielectric layer 207. The second p-type dipole layer 250b includes a second p-type dipole material that differs from the first p-type dipole material used in the first p-type dipole layer 250a. For example, the second p-type dipole material includes aluminum, niobium, titanium, boron, or phosphorus. Unlike the first p-type dipole material, the second p-type dipole material is not limited to materials where the stable oxide formation energy in the high-dielectric-constant dielectric layer 205 is higher than that in the interface layer 203. This is because the second p-type dipole layer 205b is only used to adjust the critical voltage of the p-type field-effect transistor 100b (not the n-type field-effect transistor 100a). Therefore, it is meaningless to address the issue of using a p-type dipole to adjust the n-type field-effect transistor 100a. Step 1108 may further form a fill layer 280 on the second p-type dipole layer 250b, the formation of which may be similar to the aforementioned method.

[0080] like Figure 9A and 9B and Figure 17A and 17B As shown, step 1110 of method 1100 involves annealing as described above, such as heat treatment 600. Annealing can drive the dopants in the first p-type dipole layer 250a, n-type dipole layer 252, and second p-type dipole layer 250b into the gate dielectric layer 207.

[0081] like Figure 10A and 10B and Figure 18A and 18B As shown, step 1112 of method 1100 removes individual dipole layers (and filling layer 280, if present), for example, as removed in step 1014 above. Figure 18A and 18B As shown, the dopants in the n-type dipole layer 252 and the second p-type dipole layer 250b are primarily driven into and implanted into the high-dielectric-constant dielectric layer 205 (rather than the interface layer 203). Although not shown, the interface layer 203 may also be implanted with individual n-type and second p-type dopants, although at a lower concentration than in the high-dielectric-constant dielectric layer 205. Figure 23 As shown, a second p-type doped material can form a doped layer 257. The doped layer 257 can be formed at and around the interface between the interface layer 203 and the high-dielectric-constant dielectric layer 205. The portion of the doped layer 257 that replaces the high-dielectric-constant dielectric layer 205 is larger than the portion of the doped layer 257 that replaces the interface layer 203. For example... Figure 23 As shown, the reduction in the thickness of the high-dielectric-constant dielectric layer 205 by the doped layer 257 is greater than the reduction in the thickness of the interface layer 203 by the doped layer 257. In one embodiment, the doped layer 257 comprises silicon, oxygen, hafnium, and aluminum. In one embodiment, the doped layer 257 does not contain zinc, beryllium, and / or indium.

[0082] like Figure 11A and 11B and Figure 19A and 19B As shown, in step 1114 of method 1100, n-type gate metal 210a and p-type gate metal 210b are deposited on the gate dielectric layer 207 in the n-type transistor region and the p-type transistor region, respectively. Step 1114 corresponds to the aforementioned step 1016.

[0083] Additional steps may be provided before, during, and after method 1100, and additional embodiments of method 1100 may interchange, substitute, or omit some of the steps. Additional steps may include performing a gate dicing process to form Figure 1 The isolated gate structure 218 is shown. Additional steps may include forming contacts in the device layer to connect to the source / drain epitaxial structure 204b and / or the gate structure 218. Further steps may include forming interconnect structures on the contacts in the device layer. The interconnect structures may include one or more interconnect layers having wiring and vias buried in a dielectric layer. One or more interconnect layers may connect the gates, sources, and drains of various transistors, as well as other circuitry in device 100, to form part or all of an integrated circuit. Further steps may include forming a passivation layer on the interconnect layers.

[0084] Figure 20 This describes the dopant distribution profile when driving aluminum dopant to form dipoles in an n-type field-effect transistor 100a (or an n-type metal-oxide-semiconductor device) and a p-type field-effect transistor 100b (or a p-type metal-oxide-semiconductor device). As shown in the p-type field-effect transistor 100b, a strong p-type dipole can be formed at the interface between the interface layer 203 and the high-dielectric-constant dielectric layer 205. This is because of the peak aluminum distribution at the interface, and the oxygen atoms in the high-dielectric-constant dielectric layer 205 are sufficient to form the necessary oxygen gradient at the interface (i.e., the oxygen content in the aluminum dopant layer at the interface is greater than the oxygen content in the interface layer 203). For example, the p-type gate metal 210b can serve as an additional oxygen source, providing oxygen atoms to diffuse towards the high-dielectric-constant dielectric layer 205. On the other hand, as shown in the n-type field-effect transistor 100a, a weak p-type dipole can be formed at the interface between the interface layer 203 and the high-dielectric-constant dielectric layer 205. This is because additional aluminum diffuses from the n-type gate metal 210a towards the high-dielectric-constant dielectric layer 205 to form an anti-p-type dipole. Furthermore, aluminum atoms in the n-type gate metal 210a may pull out or absorb oxygen atoms from the high-dielectric-constant dielectric layer 205. This causes a narrowing of the concentration gap at the interface between the interface layer 203 and the high-dielectric-constant dielectric layer 205, resulting in a reduced threshold voltage regulation capability. Therefore, while aluminum is suitable for forming a strong p-type dipole for the p-type field-effect transistor 100b, it may not be suitable for forming a strong p-type dipole for the n-type field-effect transistor 100a.

[0085] like Figure 21 As shown, zinc dopant (instead of aluminum dopant) is used to form the p-type dipole in the n-type field-effect transistor 100a (or n-type metal-oxide-semiconductor device). Using zinc instead of aluminum dopant allows for the formation of a strong p-type dipole at the interface between the interface layer 203 and the high-dielectric-constant dielectric layer 205 used in the n-type field-effect transistor 100a. This is partly due to the peak zinc distribution at the interface. Furthermore, as previously mentioned, the energy for forming a stable oxide of zinc in the high-dielectric-constant dielectric layer 205 is higher than that for zinc in the interface layer 203. Consequently, the zinc dipole dopant layer is primarily formed closer to the interface layer 203 and further away from the n-type gate metal 210a. This reduces the oxygen and aluminum diffusion effects caused by the n-type gate metal 210a, and weakens the anti-p-type dipole, even though some aluminum still diffuses from the n-type gate metal 210a into the high-dielectric-constant dielectric layer 205.

[0086] Figure 22 This refers to the critical voltage regulation performance between zinc-based and aluminum-based p-type dipoles. As shown in the figure, when adjusting n-type MOSFETs 100a and 100b using the same p-type dipole (e.g., zinc-based p-type dipoles instead of aluminum-based dipoles), the ratio ΔN / ΔP can be improved from 0.2 to 1 (or greater than 1). For example, the regulation range of zinc-based doped n-type MOSFETs 100a and 100b is at least 80 mV. Furthermore, using zinc-based p-type dipoles results in a much lower penalty for the equivalent capacitance thickness under the same critical voltage regulation conditions, as shown by the increase in flat-band voltage with the same change in equivalent capacitance thickness.

[0087] Figure 23 This is the outline of the zinc-dominant and aluminum-dominant dopants in the gate dielectric layer 207 in one embodiment of the present invention. (Matching) Figure 10A and 10B and Figure 18A and 18B illustrate Figure 23 As described herein, heat treatment 600 may form a doped layer 255 (e.g., a zinc doped layer) and / or a doped layer 257 (e.g., an aluminum doped layer), each perpendicularly located at the interface between the high-dielectric-constant dielectric layer 205 and the interface layer 203 in different regions. For the reasons stated above, the doped layer 255 may be primarily formed in the interface layer 203 (i.e., replacing a large portion of the interface layer 203). The doped layer 257 may be primarily formed in the high-dielectric-constant dielectric layer 205 (i.e., replacing a large portion of the high-dielectric-constant dielectric layer 205).

[0088] This invention provides various advantages for threshold voltage regulation of both n-type and p-type field-effect transistors (FETs), but is not limited thereto. One advantage is the use of the same p-type dipole material to simultaneously regulate both n-type and p-type FETs. Another advantage is the primary implantation of p-type dipole material into the interface layer, rather than a high-dielectric-constant dielectric layer, to improve threshold voltage regulation of the n-type FET. Yet another advantage is the use of n-type dipoles for further regulation of the n-type FET, and the use of different p-type dipoles to regulate the p-type FET. In this approach, a larger regulation range can be achieved for multi-threshold voltage devices.

[0089] An embodiment of the present invention relates to a method for forming a semiconductor device. The method includes forming a first active region in an n-type transistor region on a substrate; forming a second active region in a p-type transistor region on the substrate; depositing a gate dielectric layer to cover a plurality of semiconductor channels in the first and second active regions, the gate dielectric layer including an interface layer on the semiconductor channels and a high-dielectric-constant dielectric layer on the interface layer; depositing a p-type dipole layer on the gate dielectric layer in the n-type and p-type transistor regions; depositing a fill layer on the p-type dipole layer; performing a heat treatment to drive a plurality of dopants in the p-type dipole layer into the gate dielectric layer, wherein the dopants after the heat treatment have a first concentration in the interface layer greater than a second concentration in the high-dielectric-constant dielectric layer; removing the p-type dipole layer and the fill layer; and depositing an n-type gate metal on the gate dielectric layer in the n-type transistor region and depositing a p-type gate metal on the gate dielectric layer in the p-type transistor region.

[0090] In one embodiment, a dopant is formed between an interface layer and a high-dielectric-constant dielectric layer, wherein the dopant replaces a first portion of the interface layer in a manner greater than the dopant replaces a second portion of the high-dielectric-constant dielectric layer. In other embodiments, the atomic percentage of the dopant in the dopant layer is between 0.1% and 10%.

[0091] In one embodiment, the p-type dipole layer comprises zinc oxide, zinc nitride, or zinc carbide. In one embodiment, the p-type dipole layer comprises beryllium oxide, beryllium nitride, or beryllium carbide. In one embodiment, the p-type dipole layer comprises indium oxide, indium nitride, or indium carbide.

[0092] In one embodiment, the aluminum concentration and zinc concentration at the interface between the interface layer and the high dielectric constant dielectric layer are substantially the same.

[0093] In one embodiment, the high dielectric constant dielectric layer comprises hafnium oxide, the interface layer comprises silicon oxide, and the first area oxygen density of the p-type dipole layer is greater than the second area oxygen density of the interface layer.

[0094] In one embodiment, the n-type gate metal comprises aluminum nitride, while the p-type gate metal comprises titanium nitride.

[0095] Another embodiment relates to a method for forming a semiconductor device. The method includes forming an active region in an n-type transistor region on a substrate; depositing an interface layer to cover a plurality of semiconductor channels in the active region; depositing a high-dielectric-constant dielectric layer on the interface layer; depositing a zinc oxide dipole layer on the high-dielectric-constant dielectric layer; depositing a fill layer on the zinc oxide dipole layer, wherein the fill layer fills the space between adjacent semiconductor channels; performing a heat treatment to drive a plurality of zinc dopants in the zinc oxide dipole layer into the high-dielectric-constant dielectric layer and the interface layer, wherein the zinc dopants reduce the thickness of the interface layer; removing the zinc oxide dipole layer and the fill layer; and depositing an n-type gate metal on the high-dielectric-constant dielectric layer.

[0096] In one embodiment, the rate at which zinc dopant is driven into the interface layer is greater than the rate at which zinc dopant is driven into a dielectric layer with a high dielectric constant.

[0097] In one embodiment, the active region forms the pass-through transistor and pull-down transistor of the static random access memory device.

[0098] In one embodiment, the method further includes forming a second active region in an n-type transistor region on a substrate, wherein an interface layer covers a plurality of second semiconductor channels of the second active region, wherein a high-dielectric-constant dielectric layer is deposited on the interface layer on the second semiconductor channels, wherein a zinc oxide dipole layer is deposited on the high-dielectric-constant dielectric layer on the second semiconductor channels, wherein a fill layer completely fills the space vertically located between adjacent second semiconductor channels; and prior to thermal processing, patterning the zinc oxide dipole layer and the fill layer to selectively remove portions of the zinc oxide dipole layer and the fill layer on the second semiconductor channels.

[0099] In another embodiment, the method further includes depositing a lanthanum oxide dipole layer on a high-dielectric-constant dielectric layer on a second semiconductor channel; performing a heat treatment to drive a plurality of lanthanum dopants in the lanthanum oxide dipole layer into the high-dielectric-constant dielectric layer; removing the lanthanum oxide dipole layer; and depositing an n-type gate metal on the high-dielectric-constant dielectric layer on the second semiconductor channel. In another embodiment, the second active region forms an n-type logic device.

[0100] In one embodiment, the method further includes forming a third active region in a p-type transistor region on a substrate, wherein an interface layer covers a plurality of third semiconductor channels in the third active region, wherein a high-dielectric-constant dielectric layer is deposited on the interface layer on the third semiconductor channels; wherein a zinc oxide dipole layer is deposited on the high-dielectric-constant dielectric layer on the third semiconductor channels, wherein a filling layer completely fills the space vertically located between adjacent third semiconductor channels, and wherein a heat treatment step drives zinc dopant in the zinc oxide dipole layer into the high-dielectric-constant dielectric layer and the interface layer on the third semiconductor channels. In other embodiments, the third active region forms a pull-up transistor for a static random access memory device.

[0101] Another embodiment of the present invention relates to a semiconductor device. The semiconductor device includes a first stack and a second stack of multiple semiconductor channels located on a substrate; a first interface layer and a second interface layer respectively covering the first stack and the second stack of semiconductor channels; a first high-dielectric-constant dielectric layer and a second high-dielectric-constant dielectric layer respectively covering the first interface layer and the second interface layer; and a first gate work function metal and a second gate work function metal respectively located on the first high-dielectric-constant dielectric layer and the second high-dielectric-constant dielectric layer. The first gate work function metal and the second gate work function metal are work function metals with opposite morphologies. The first interface layer and the second interface layer are doped with a first concentration of p-type dopant, and the first high-dielectric-constant dielectric layer and the second high-dielectric-constant dielectric layer are doped with a second concentration of p-type dopant, wherein the first concentration is greater than the second concentration.

[0102] In one embodiment, a p-type dopant forms a first doped layer at the interface between a first interface layer and a first high dielectric constant dielectric layer.

[0103] In one embodiment, the first doped layer includes oxides, nitrides, carbides, or metal compounds of beryllium, indium, or zinc.

[0104] The features of the above embodiments are beneficial for those skilled in the art to understand the present invention. Those skilled in the art should understand that the present invention can be used as a basis to design and vary other processes and structures to achieve the same objectives and / or advantages as the above embodiments. Those skilled in the art should also understand that these equivalent substitutions do not depart from the concept and scope of the present invention, and changes, substitutions, or modifications can be made without departing from the concept and scope of the present invention.

Claims

1. A method for forming a semiconductor device, comprising: A first active region is formed in an n-type transistor region on a substrate; A second active region is formed in a p-type transistor region on the substrate; A gate dielectric layer is deposited to cover a plurality of semiconductor channels of the first active region and the second active region. The gate dielectric layer includes an interface layer on the semiconductor channels and a high dielectric constant dielectric layer on the interface layer. A p-type dipole layer is deposited on the gate dielectric layer in the n-type transistor region and the p-type transistor region; A filling layer is deposited on the p-type dipole layer; A heat treatment is performed to drive a plurality of dopants in the p-type dipole layer into the gate dielectric layer, wherein the concentration of the dopants in the interface layer after the heat treatment is greater than a second concentration in the high dielectric constant dielectric layer. Remove the p-type dipole layer and the filler layer; and An n-type gate metal is deposited on the gate dielectric layer in the n-type transistor region, and a p-type gate metal is deposited on the gate dielectric layer in the p-type transistor region.

2. The method of forming a semiconductor device as claimed in claim 1, wherein the dopants form a doped layer between the interface layer and the high dielectric constant dielectric layer, wherein the dopants replace a first portion of the interface layer greater than the dopants replace a second portion of the high dielectric constant dielectric layer.

3. The method of forming a semiconductor device as claimed in claim 2, wherein the atomic percentage of the dopants in the doped layer is between 0.1% and 10%.

4. The method for forming a semiconductor device as claimed in claim 1, wherein the p-type dipole layer comprises zinc oxide, zinc nitride, or zinc carbide.

5. A method for forming a semiconductor device, comprising: An active region is formed in an n-type transistor region on a substrate; An interface layer is deposited to cover multiple semiconductor channels in the active region; A dielectric layer with a high dielectric constant is deposited on the interface layer; A zinc oxide dipole layer is deposited on the dielectric layer with a high dielectric constant; A filler layer is deposited on the zinc oxide dipole layer, and the filler layer fills the space between adjacent semiconductor channels; A heat treatment is performed to drive multiple zinc dopants in the zinc oxide dipole layer into the high dielectric constant dielectric layer and the interface layer, wherein these zinc dopants reduce the thickness of the interface layer. Remove the zinc oxide dipole layer and the filler layer; and An n-type gate metal is deposited on the high dielectric constant dielectric layer.

6. The method of forming a semiconductor device as claimed in claim 5, wherein the rate at which the zinc dopants are driven into the interface layer is greater than the rate at which the zinc dopants are driven into the high dielectric constant dielectric layer.

7. The method of forming a semiconductor device as claimed in claim 5, wherein the active region forms a gate-through transistor and a pull-down transistor for a static random access memory device.

8. A semiconductor device, comprising: A first stack and a second stack of multiple semiconductor channels are located on a substrate; A first interface layer and a second interface layer respectively cover the first stack and the second stack of the semiconductor channels; A first high-dielectric-constant dielectric layer and a second high-dielectric-constant dielectric layer respectively cover the first interface layer and the second interface layer; and A first gate work function metal and a second gate work function metal are respectively located on the first high dielectric constant dielectric layer and the second high dielectric constant dielectric layer. The first gate work function metal and the second gate work function metal are work function metals with opposite forms. The first interface layer and the second interface layer are doped with a p-type dopant of a first concentration, and the first high dielectric constant dielectric layer and the second high dielectric constant dielectric layer are doped with the p-type dopant of a second concentration, wherein the first concentration is greater than the second concentration.

9. The semiconductor device of claim 8, wherein the p-type dopant forms a first doped layer at the interface between the first interface layer and the first high dielectric constant dielectric layer.

10. The semiconductor device of claim 9, wherein the first doped layer comprises an oxide, nitride, carbide, or metal compound of beryllium, indium, or zinc.