Methods for manufacturing semiconductor devices and semiconductor devices
By employing an aluminum diffusion barrier layer in the work function matching material layer, the issue of aluminum diffusion between transistors is mitigated, ensuring consistent threshold voltages and improved performance in integrated circuits.
Patent Information
- Application Number
- DE102020115422
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2020-05-29
- Filing Date
- 2020-06-10
- Publication Date
- 2024-11-28
- Estimated Expiration
- 2040-06-10
AI Technical Summary
As integrated circuits become smaller, the diffusion of metals like aluminum between adjacent transistors with different threshold voltages causes threshold voltage shifts and performance degradation due to the close proximity of devices with varying metal gate films.
Incorporation of an aluminum diffusion barrier layer on the upper and/or lower surfaces of the aluminum-containing work function matching material layer to prevent aluminum diffusion into adjacent devices, using Ta-rich, Ta-doped, Si-doped, or Ti-rich layers to maintain device isolation and performance.
The diffusion barrier layer effectively suppresses aluminum diffusion, maintaining consistent threshold voltages and improving the electrical performance of adjacent transistors, enhancing device reliability and efficiency.
Smart Images

Figure 00000000_0000_ABST
Abstract
Description
BACKGROUND
[0001] With the increasing miniaturization of integrated circuits and the ever-greater demands on their speed, transistors must deliver higher control currents in ever smaller dimensions. Therefore, three-dimensional field-effect transistors (3D FETs) have been developed. Three-dimensional FETs (3D FETs) feature vertical semiconductor nanostructures (such as fins, nanowires, nanosheets, etc.) over a substrate. These semiconductor nanostructures are used to form source and drain regions, as well as channel regions between them. Trench insulation regions (STI regions) are formed to define the semiconductor nanostructures. 3D FETs also feature gate stacks, which are formed on the sidewalls and top surfaces of the semiconductor fins or on all sides of nanowires or nanosheets.Since 3D-FETs have a three-dimensional channel structure, ion implantation processes of the channel require special care to reduce any kind of geometric effect.
[0002] US patent 2020 / 0105894A1 describes a metal-gate structure of a field-effect transistor which has several TiAl layers with different aluminum-to-titanium ratios, with the layer with more titanium being intended to trap oxygen.
[0003] In US 2017 / 0117191A1, various structuring designs for NMOS and PMOS FinFETs for reducing a threshold voltage are discussed. An n-type exit layer with aluminum and titanium features two sublayers, one with a higher titanium content and one with a higher aluminum content. This is intended to reduce the diffusion of aluminum into the high-k dielectric.
[0004] US 2014 / 0167187A1 describes material compositions for work matching layers of a metal oxide semiconductor transistor.
[0005] In US 2012 / 0012937A1 and US 2015 / 0014780A1, connecting structures between adjacent metal gate stacks are proposed. BRIEF DESCRIPTION OF THE DRAWINGS
[0006] The present disclosure is best understood with reference to the following detailed description in conjunction with the accompanying drawings. It should be noted that, in accordance with industry practice, various elements are not shown to scale and are for illustrative purposes only. In fact, the dimensions of the various elements may have been arbitrarily enlarged or reduced for the sake of clarity. Fig. Figure 1A shows a cross-sectional view and Fig. Figure 1B shows a perspective view of a semiconductor device in accordance with an embodiment of the present disclosure. Fig. 2A, Fig. 2B, Fig. 2C, Fig. 2D, Fig. 2E and Fig. Figures 2F show cross-sectional views of various stages of a sequential manufacturing process of a semiconductor device in accordance with an embodiment of the present disclosure. Fig. 3A, Fig. 3B, Fig. 3C, Fig. 3D, Fig. 3E and Fig. Figure 3F shows cross-sectional views of various stages of a sequential manufacturing process of a semiconductor device in accordance with an embodiment of the present disclosure. Fig. Figure 3G shows a process flow for the manufacture of a semiconductor device in accordance with an embodiment of the present disclosure. Fig. Figure 4A shows gate structures of several FETs with different threshold voltages in accordance with embodiments of the present disclosure. Fig. 4B and Fig. Figure 4C shows various output work matching material layers and high k dielectric gate layers for multiple FETs with different threshold voltages in accordance with embodiments of the present disclosure. Fig. Figure 5A shows a top view (layout) of a CMOS circuit, Fig. 5B shows a cross-sectional view corresponding to area A1 of Fig. 5A, and Fig. Figure 5C shows an enlarged view of area B1 of Fig. 5B in accordance with an embodiment of the present disclosure. Fig. Figure 6A shows a top view (layout) of a CMOS circuit, Fig. Figure 6B shows a cross-sectional view corresponding to area A1 of Fig. 6A, and Fig. Figure 6C shows an enlarged view of area B1 of Fig. 6B in accordance with an embodiment of the present disclosure. Fig. 7A, Fig. 7B, Fig. 7C, Fig. 7D, Fig. 7E and Fig. Figure 7F shows various cross-sectional views of an n-output work matching material layer (WFM layer) in accordance with embodiments of the present disclosure. Fig. 8A and Fig. Figure 8B shows cross-sectional views and Ti concentrations of n-output work matching material (WFM) layers in accordance with embodiments of the present disclosure. Fig. Figure 9 shows a process flow for the manufacture of a semiconductor device in accordance with embodiments of the present disclosure. Fig. 10A, Fig. 10B, Fig. 10C, Fig. 10D, Fig. 10E and Fig. Figure 10F shows cross-sectional views of various stages of a sequential manufacturing process of a semiconductor device in accordance with embodiments of the present disclosure. Fig. Figure 11 shows a suppression of Al diffusion by a diffusion barrier layer in accordance with embodiments of the present disclosure. Fig. 12A and Fig. Figure 12B shows thickness reduction effects of an interface layer in accordance with embodiments of the present disclosure. DETAILED DESCRIPTION
[0007] It is understood that the following disclosure provides many different embodiments for implementing various features of the invention. Specific embodiments of components and arrangements are described below. For example, the dimensions are not limited to the disclosed range or values, but may depend on process conditions and / or desired properties of the device. Furthermore, the formation of a first element above or on top of a second element in the following description may include embodiments in which the first and second elements are in direct contact with each other, and may also include embodiments in which additional elements are formed between the first and second elements, so that the first and second elements do not have to be in direct contact with each other.For the sake of simplicity and clarity, various elements may be drawn at different scales. Layers / elements may have been omitted from the accompanying drawings for the sake of simplicity.
[0008] Furthermore, terms of spatial relationships, such as "below," "underneath," "lower," "above," "upper," and the like, may be used herein for the purpose of simplifying the description of the relationship of one element or feature depicted in the figures to another element(s) or feature(s). These terms of spatial relationships are intended to encompass various orientations of the device during use or operation, in addition to the orientation shown in the figures. The component / device may also be oriented differently (rotated by 90 degrees or in other orientations), and the terms of spatial relationships used herein may be interpreted accordingly. Moreover, the term "(made) of" may mean either "comprising / having" or "consisting of / made of."Furthermore, in the following manufacturing process, one or more additional operations may be arranged between the described operations, and the sequence of operations may be changed. In this disclosure, a phrase such as "either A, B, or C" means "A, B, and / or C" (A, B, C, A and B, A and C, B and C, or A, B, and C), and does not mean one of the elements of A, one of the elements of B, and one of the elements of C, unless otherwise specified. Throughout this disclosure, a source and a drain are used interchangeably, and source / drain refers either to a source or a drain, or both together. In the following embodiments, materials, arrangements, dimensions, processes, and / or operations described with respect to one embodiment (for example, one or more figures) may be used in the other embodiments, and a detailed description thereof may have been omitted.
[0009] Disclosed embodiments relate to a semiconductor device, in particular a gate structure of a field-effect transistor (FET), and its fabrication or manufacturing process. The embodiments, such as those disclosed herein, are generally applicable not only to planar FETs, but also to a fin FET (FinFET), a dual-gate FET, an encapsulated gate FET, an omega-gate FET, or a wraparound gate FET (GAA FET) (such as a side-encircling gate FET or a vertical wraparound gate FET), and / or nanowire transistors, nanosheet transistors, nanofork-blade transistors, nanostructure transistors, nanoplate transistors, or any suitable devices comprising one or more output function matching material (WFM) layers in the gate structure.
[0010] As integrated circuits become increasingly miniaturized, the space between adjacent components decreases, and components with different threshold voltages move very close together. This causes a threshold voltage shift due to the diffusion of metals (such as aluminum) from one component to another. In FET structures, the design of multiple low-voltage devices (Vt) is crucial for low power consumption and increased device performance. The composition and thickness of metal gate films play a critical role in defining the device's work function (Vt). Multiple FETs exhibiting different threshold voltages can be achieved by matching the materials and / or thicknesses of one or more work function matching material (WFM) layers placed between a dielectric gate layer and a body metal gate electrode layer (e.g., a W-layer).For an n-type FET having a Si channel and / or a p-type FET having a SiGe channel, an aluminum-containing layer, such as TiAl, TiAlC, TaAl, and / or TaAlC, is used as a WFM layer. However, if the aluminum-containing layer is formed as a WFM layer over an underlying layer, such as a WCN, WN, and / or TiN layer, the diffusion of aluminum into the WCN, WN, and / or TiN layer of adjacent devices causes a change in the threshold voltage and / or other deterioration of the FET characteristics.
[0011] The present disclosure relates to the use of an aluminum diffusion barrier layer on an upper surface and / or a lower surface of an aluminum-containing WFM layer. As discussed below, the present disclosure provides devices and methods that can protect the underlying layer on the same FET device and can also protect the high k layers and / or the WFM layers in adjacent FET devices from Al diffusion from the aluminum-containing WFM layer.
[0012] Fig. Figure 1A shows a cross-sectional view of a semiconductor device in accordance with an embodiment of the present disclosure.
[0013] In some embodiments, a semiconductor device has a gate stack 80 arranged over a channel region of a fin structure 20. The gate stack 80 comprises an interface layer 81, a dielectric gate layer 82, a first conductive layer 83 as a cap / cover layer, a second conductive layer 84 as a first barrier layer, an output function matching material layer or an output function matching material layer (a WFM layer) 86, an adhesive layer 87, and a body-gate electrode layer 88, as shown in Fig. Figure 1A shows that in some embodiments, the fin structure 20 is arranged over a substrate 10 and protrudes from an insulating layer 30. Furthermore, gate sidewall spacers 46 are arranged on opposite side faces of the gate stack 80, and one or more dielectric layers 50 are formed to cover the gate sidewall spacers 46. In some embodiments, a piece of insulating material 42 is arranged between the gate sidewall spacer 46 and the insulating layer 30. As shown in Fig. As shown in Figure 1B, epitaxial source / drain layers 60 are also formed over depressed fin structures. Although Fig. 1A shows two fin structures and Fig. 1B shows three fin structures; the number of fin structures is not limited to those in the Fig. 1A and Fig. 1B shown limited.
[0014] In some embodiments, the fin structure (a channel region) for an n-FET is made of Si and for a p-FET of SiGe. The Ge concentration of SiGe ranges from approximately 20 atomic percent to 60 atomic percent in some embodiments and from approximately 30 atomic percent to 50 atomic percent in others. In some embodiments, the channel region of the n-FET contains Ge in a lower quantity than that in the SiGe channel of the p-FET. In other embodiments, both the channel region of a p-FET and that of an n-FET are made of Si or a composite semiconductor.
[0015] In some embodiments, the first conductive layer 83 contains a metal nitride, such as WN, TaN, TiN, and TiSiN. In some embodiments, TiN is used. The thickness of the first conductive layer 83 is in the range of approximately 0.3 nm to approximately 30 nm in some embodiments and in the range of approximately 0.5 nm to approximately 25 nm in other embodiments. In some embodiments, the first conductive layer 83 is crystalline, for example, having columnar crystal grains. In some embodiments, the first conductive layer 83 is not formed. In some embodiments, the first conductive layer 83 is formed and then removed after a tempering process using a wet etching process.
[0016] In some embodiments, the second conductive layer 84 contains a metal nitride, such as WN, TaN, TiN, and TiSiN. In some embodiments, TaN is used. The thickness of the second conductive layer 84 is in the range of approximately 0.3 nm to approximately 30 nm in some embodiments and in the range of approximately 0.5 nm to approximately 25 nm in other embodiments. In some embodiments, the second conductive layer 84 serves as a barrier layer or an etch stop layer. In some embodiments, the second conductive layer 84 is thinner than the first conductive layer 83. In some embodiments, the second conductive layer 84 is not formed.
[0017] In some embodiments, the WFM layer 86 is made of a conductive material, such as a single layer of TiN, WN, WCN, Ru, W, TaAlC, TiC, TaAl, TaC, Co, Al, or TiAl, or a multilayer of two or more of these materials. For an n-FET having a Si channel, an aluminum-containing layer, such as TiAl, TiAlC, TaAl, and / or TaAlC, is used. In some embodiments, one or more of the materials TaN, TiN, WN, TiC, WCN, MoN, and / or Co are optionally used, formed beneath the aluminum-containing layer. For a p-FET having a SiGe channel, one or more of the materials TaN, TiN, WN, TiC, WCN, MoN, and Co are used. In some embodiments, one or more of the materials formed thereon, TiAl, TiAlC, TaAl, and TaAlC, are used.
[0018] In some embodiments, the adhesive layer 87 is made of one or more of the materials TiN, Ti, and Co. In some embodiments, the body-gate electrode layer 88 comprises one or more layers of conductive material, such as polysilicon, aluminum, copper, titanium, tantalum, tungsten, cobalt, molybdenum, tantalum nitride, nickel silicide, cobalt silicide, TiN, WN, TiAl, TiAlN, TaCN, TaC, TaSiN, metal alloys, other suitable materials, and / or combinations thereof.
[0019] As explained above, in some embodiments the first conductive layer 83 and the second conductive layer 84 are not formed. In such a case, one or more WFM layers are formed directly on the dielectric gate layer 82.
[0020] Fig. Figures 2A - 2F show cross-sectional views of different stages of a sequential manufacturing process of the semiconductor device in accordance with an embodiment of the present disclosure. Fig. Figure 3G shows a process flow for the manufacture of a semiconductor device in accordance with an embodiment of the present disclosure. It is understood that in the sequential manufacturing process, one or more additional operations are performed before, during, and after the operations described in the Fig. Stages 2A-3F can be provided, and some of the operations described below can be substituted or omitted for additional embodiments of the method. The sequence of operations / processes can be interchangeable.
[0021] As in Fig. As shown in Figure 2A, one or more fin structures 20 are formed over a substrate 10. The substrate 10 is, for example, a p-silicon substrate with an impurity concentration in the range of approximately 1 × 10⁻⁶. 15 cm -3 up to approximately 1 × 10 18 cm -3 In other embodiments, the substrate 10 is an n-silicon substrate with an impurity concentration in the range of approximately 1 × 10 15 cm -3 up to approximately 1 × 10 18 cm -3Alternatively, the substrate 10 can comprise another elemental semiconductor, such as germanium; a compound semiconductor comprising group IV-IV compound semiconductors, such as SiC and SiGe; group III-V compound semiconductors, such as GaAs, GaP, GaN, InP, InAs, InSb, GaAsP, AlGaN, AlInAs, AlGaAs, GaInAs, GaInP, and / or GaInAsP; or combinations thereof. In one embodiment, the substrate 10 is a silicon layer of a silicon-on-insulator (SOI) substrate. Amorphous substrates, such as amorphous Si or amorphous SiC, or insulating materials, such as silicon oxide, can also be used as the substrate 10. The substrate 10 can have various regions that have been appropriately doped with impurities (for example, with p- or n-type conductivity).
[0022] In some embodiments, part of the substrate 10 is recessed for p-FETs by etching, and a SiGe layer is formed over the recesses. Fig. 2A - 3F show the case of an n-FET, but most of the manufacturing process is essentially the same for a p-FET.
[0023] The fin structures 20 can be structured by any suitable method. For example, the fin structures 20 can be structured using one or more photolithography processes, including dual or multiple structuring processes. Typically, dual and multiple structuring processes combine photolithography and self-aligning processes, enabling the creation of structures with, for example, spacing dimensions that are smaller than those achievable using a single, direct photolithography process. In one embodiment, for example, a sacrificial layer is formed over a substrate and structured using a photolithography process. Spacing elements are formed along with the structured sacrificial layer using a self-aligning process.Then the sacrificial layer is removed, and the remaining spacer elements can be used to structure the fin structures 20.
[0024] As in Fig. As shown in Figure 2A, two fin structures 20, extending in the Y direction, are arranged side by side in the X direction. However, the number of fin structures is not limited to two. The number can be one, two, three, four, five, or more. Furthermore, one or more dummy fin structures can be arranged adjacent to both sides of the fin structures 20 to improve the structural accuracy in the structuring processes. The width of the fin structure 20 ranges from approximately 5 nm to approximately 40 nm in some embodiments and from approximately 7 nm to approximately 15 nm in others. The height of the fin structure 20 ranges from approximately 100 nm to approximately 300 nm in some embodiments and from approximately 50 nm to approximately 100 nm in others.The space between the fin structures 20 lies in a range of approximately 5 nm to approximately 80 nm in some embodiments, and in a range of approximately 7 nm to approximately 15 nm in other embodiments. However, those skilled in the art will recognize that the dimensions and values given in these descriptions are only examples and can be modified to fit integrated circuits of different sizes.
[0025] After the fin structures 20 have been formed, an insulating layer 30 is formed over the fin structures 20, as shown in Fig. 2B shown.
[0026] The insulating layer 30 comprises one or more layers of insulating materials, such as silicon oxide, silicon oxynitride, or silicon nitride, formed by LPCVD (low-pressure chemical vapor deposition), plasma CVD, or flowable CVD. In flowable CVD, flowable dielectric materials can be deposited instead of silicon oxide. Flowable dielectric materials, as the name suggests, can "flow" during deposition to fill gaps or spaces with a high aspect ratio. Typically, various chemicals are added to silicon-containing precursors to enable the flow of the deposited film. In some embodiments, nitrogen hydride bonds are added.Examples of flowable dielectric precursors, particularly flowable silicon oxide precursors, include a silicate, a siloxane, a methylsilsesquioxane (MSQ), a hydrogen silsesquioxane (HSQ), a mixture of MSQ and HSQ, a perhydrosilazane (TCPS), a perhydropolysilazane (PSZ), a tetraethyl orthosilicate (TEOS), or a silylamine, such as trisilylamine (TSA). These flowable silicon oxide materials are formed in a multi-step process. After the flowable film has been applied, it is hardened and then annealed to remove unwanted elements from the silicon oxide. The flowable film can be doped with boron and / or phosphorus. The insulating layer 30 can, in some embodiments, be formed by one or more layers of spin-on glass (SOG), SiO₂, SiON₂, SiOCN₂, and / or fluoride-doped silicate glass (FSG).
[0027] After the formation of the insulating layer 30 over the fin structures 20, a planarization process is performed to remove a portion of the insulating layer 30 and the mask layer (e.g., the contact oxide layer and the silicon nitride mask layer formed on the contact oxide layer). The planarization process can include chemical-mechanical polishing (CMP) and / or a back-etching process. Subsequently, the insulating layer is removed further, exposing an upper portion of the fin structure 20, which is intended to become a channel layer, as shown in Fig. 2B shown.
[0028] In certain embodiments, the partial removal of the insulating layer 30 is carried out using a wet etching process, for example by immersing the substrate in hydrofluoric acid (HF). In another embodiment, the partial removal of the insulating layer 30 is carried out using a dry etching process. For example, a dry etching process can be used with CHF3 or BF3 as etching gases.
[0029] After the formation of the insulating layer 30, a thermal process, for example a tempering process, can be carried out to improve the quality of the insulating layer 30. In certain embodiments, the thermal process is carried out using rapid thermal tempering (RTA) at a temperature in the range of approximately 900°C to approximately 1050°C for approximately 1.5 seconds to approximately 10 seconds in an inert gas environment, such as an N2, Ar or He environment.
[0030] Then a dummy gate structure 40 is formed over a part of the fin structures 20, as shown in Fig. 2C shown.
[0031] A dielectric layer and a polysilicon layer are formed over the insulating layer 30 and the exposed fin structures 20 before structuring operations are performed to obtain a dummy gate structure comprising a dummy gate electrode layer 44 made of polysilicon and a dielectric dummy gate layer 42. The structuring of the polysilicon layer is carried out using a hard mask, which in some embodiments comprises a silicon nitride layer and an oxide layer. The dielectric dummy gate layer 42 can be silicon oxide formed by CVD, PVD, ALD, electron beam evaporation, or other suitable processes. In some embodiments, the dielectric dummy gate layer 42 comprises one or more layers of silicon oxide, silicon nitride, silicon oxynitride, or high-k dielectrics.In some embodiments, the thickness of the dielectric dummy gate layer is in the range of approximately 1 m to approximately 5 nm.
[0032] In some embodiments, the dummy gate electrode layer 44 is doped with polysilicon by uniform or non-uniform doping. In the present embodiment, the width of the dummy gate electrode layer 44 is in the range of approximately 30 nm to approximately 60 nm. In some embodiments, the thickness of the dummy gate electrode layer is in the range of approximately 30 nm to approximately 50 nm. Furthermore, one of several dummy gate structures can be arranged adjacent to both sides of the dummy gate structure 40 to improve the structural accuracy in the structuring processes. The width of the dummy gate structure 40 is in the range of approximately 5 nm to approximately 40 nm in some embodiments, and in certain embodiments, it is in the range of approximately 7 nm to approximately 15 nm.
[0033] As in Fig. 2C and Fig. As shown in 2D, the side wall spacer elements 46 are further formed on opposite side walls of the dummy gate structures 40. Fig. 2D is a cross-section in the yx-plane. An insulating layer for the sidewall spacers 46 is formed over the dummy gate structure 40. The insulating layer is conformally applied / deposited so that it has substantially the same thickness on vertical surfaces, such as the sidewalls, horizontal surfaces, and the top of the dummy gate structure 40. In some embodiments, the insulating layer has a thickness in the range of approximately 5 nm to approximately 20 nm. The insulating layer comprises one or more of the materials SiN, SiON, and SiCN, or any other suitable dielectric material. The insulating layer can be formed by ALD, CVD, or any other suitable method.Next, lower sections of the insulating material layer are removed by anisotropic etching, forming the gate sidewall spacers 46. In some embodiments, the sidewall spacers 46 have two to four layers of different insulating materials. In some embodiments, a portion of the dielectric dummy gate layer 42 is arranged between the sidewall spacers 46 and the insulating layer 30. In other embodiments, no portion of the dielectric dummy gate layer 42 is arranged between the sidewall spacers 46 and the insulating layer 30.
[0034] Subsequently, in some embodiments, a source / drain region of the fin structure 20, which is not covered by the dummy gate structure 40, is etched (recessed) to form a source / drain depression. After the formation of the source / drain depression, one or more epitaxial source / drain layers 60 are formed within the source / drain depression (see Fig. 1B). In some embodiments, a first epitaxial layer, a second epitaxial layer, and a third epitaxial layer are formed. In other embodiments, no depression is formed, and the epitaxial layers are formed over the fin structure.
[0035] In some embodiments, the first epitaxial layer contains SiP or SiCP for an n-FinFET, and SiGe or B-doped Ge for a p-FinFET. The amount of P (phosphorus) in the first epitaxial layer is in the range of approximately 1 × 10⁻⁶ in some embodiments. 18atoms / cm² 3 up to approximately 1 × 10 20 atoms / cm² 3 The thickness of the first epitaxial layer ranges from approximately 5 nm to approximately 20 nm in some embodiments, and from approximately 5 nm to approximately 15 nm in other embodiments. If the first epitaxial layer is SiGe, the Ge content is approximately 25 atomic percent to approximately 32 atomic percent in some embodiments, and approximately 28 atomic percent to approximately 30 atomic percent in other embodiments. In some embodiments, the second epitaxial layer contains SiP or SiCP for an n-FinFET, and B-doped SiGe for a p-FinFET. In some embodiments, the phosphorus content in the second epitaxial layer is greater than that of the first epitaxial layer, and is in the range of approximately 1 × 10⁻⁶. 20 atoms / cm² 3 up to approximately 2 × 10 20 atoms / cm² 3The thickness of the second epitaxial layer ranges from approximately 20 nm to approximately 40 nm in some embodiments and from approximately 25 nm to approximately 35 nm in other embodiments. If the second epitaxial layer is SiGe, the amount of Ge is approximately 35 atomic percent to approximately 55 atomic percent in some embodiments and approximately 41 atomic percent to approximately 46 atomic percent in other embodiments. The third epitaxial layer comprises a SiP epitaxial layer in some embodiments. The third epitaxial layer is a sacrificial layer for the formation of silicide in the source / drain. In some embodiments, the amount of phosphorus in the third epitaxial layer is less than the amount of phosphorus in the second epitaxial layer and is in the range of approximately 1 × 10⁻⁶. 18 atoms / cm² 3 up to approximately 1 × 10 21 atoms / cm² 3If the third epitaxial layer is SiGe, the amount of Ge in some embodiments is less than approximately 20 atomic percent, and in other embodiments is approximately 1 atomic percent to approximately 18 atomic percent.
[0036] In at least one embodiment, the epitaxial layers 60 are epitaxially grown by an LPCVD process, molecular beam epitaxy, atomic layer deposition, or any other suitable method. The LPCVD process is carried out at a temperature of approximately 400 to 850°C and under a pressure of approximately 1 Torr to 200 Torr using silicon source gas, such as SiH4, Si2H6, or Si3H8; germanium source gas, such as GeH4 or G2H6; carbon source gas, such as CH4 or SiH3CH3; and phosphorus source gas, such as PH3.
[0037] Then, as in Fig. 2C and Fig. As shown in 2D, an interlayer dielectric (ILD) 50 is formed over the epitaxial S / D layer 60 and the dummy gate structure 40. Materials for the ILD layer 50 include compounds containing Si, O, C and / or H, such as silicon dioxide, SiCOH and SiOC. Organic materials, such as polymers, can also be used for the ILD layer 50.
[0038] After the ILD layer 50 has been formed, a planarization process, such as CMP, is performed so that the upper section of the dummy gate electrode layer 44 is exposed, as shown in Fig. 2C and Fig. 2D shown. Before the ILD layer 50 is formed, in some embodiments a contact etch stop layer, such as a silicon nitride layer or a silicon oxynitride layer, is formed.
[0039] Then the dummy gate electrode layer 44 and the dielectric dummy gate layer 42 are removed, forming a gate spacer element 47, as shown in Fig. 2E and Fig. 2F shown. Fig. 2F is a cross-section in the yx plane. The dummy gate structures can be removed using plasma dry etching and / or wet etching. If the dummy gate electrode layer 44 is polysilicon and the ILD layer 40 is silicon dioxide, a wet etchant, such as a TMAH solution, can be used to selectively remove the dummy gate electrode layer 44. The dielectric dummy gate layer 42 is then removed using plasma dry etching and / or wet etching.
[0040] Fig. Figure 3A shows the structure after the channel area of the fin structures 20 in the gate spacer element 47 has been exposed. Fig. In 3A - 3F, the side wall spacer elements 46 and the ILD layer 50 have been omitted.
[0041] As in Fig. 3B is shown, is shown at S301 by Fig. 3G formed an interface layer 81 on the fin structure 20, and at S303 of Fig. In 3G, a dielectric gate layer 82 is formed on the interface layer 81. In some embodiments, the interface layer is formed using chemical oxidation. In some embodiments, the interface layer 81 contains one of the materials silicon oxide, silicon nitride, or silicon germanium oxide. If the channel is made of Si, the interface layer in some embodiments is a silicon oxide layer 81N, and if the channel is made of SiGe, the interface layer is a silicon germanium oxide layer 81P (see Fig. 4A). In some embodiments, the thickness of the interface layer 81 is in the range of approximately 0.6 nm to approximately 2 nm. In some embodiments, the dielectric gate layer 82 comprises one or more layers of a dielectric material, such as silicon oxide, silicon nitride, or a high-k dielectric material, other suitable dielectric materials, and / or combinations thereof. Examples of high-k dielectric materials include HfO₂, HfSiO₂, HfSiON₄, HfTaO, HfTiO₂, HfZrO₂, zirconium oxide, aluminum oxide, titanium oxide, hafnium dioxide-alumina alloy (HfO₂-Al₂O₃ alloy), La₂O₃, HfO₂-La₂O₃, La₂O₃, Y₂O₃, Dy₂O₃, Sc₂O₃, MgO, or other suitable high-k dielectric materials and / or combinations thereof. The dielectric gate layer 82 can be formed by CVD, ALD or any suitable method.In one embodiment, the dielectric gate layer 82 is formed using a highly conformal deposition process, such as ALD, to ensure the formation of a dielectric gate layer with a uniform thickness around each of the channel layers. In one embodiment, the thickness of the dielectric gate layer 82 is in the range of approximately 1 nm to approximately 30 nm.
[0042] Then, as in Fig. 3C shown, at S305 of Fig. 3G a first conductive layer 83 is formed. The first conductive layer 83 can be formed in some embodiments by CVD, ALD or any suitable method. In some embodiments, the first conductive layer 83 is made of TiN or TiSiN. In some embodiments, no first conductive layer 83 is formed.
[0043] In some embodiments, S307 is used by Fig. 3G performs an initial annealing process after the formation of the first conductive layer 83 for approximately 1 ns (pulse annealing, such as laser annealing and / or isothermal annealing) to approximately 360 seconds at a temperature of approximately 600°C to approximately 950°C. This initial annealing can help to densify the dielectric gate layer 82 and incorporate nitrogen into it. Nitrogen helps to passivate oxygen vacancies, reducing leakage currents and improving device reliability. The initial annealing can also help to form a stable intermixing layer, which helps to provide a stable platform for subsequent metal gate film deposition onto the dielectric layer.If the temperature is too high, the initial annealing can cause crystallization and grain boundary formation in the high-k dielectric gate layer 82, which impairs the leakage current characteristics and the regrowth of the interface layer 81, thus slowing down the device speed. Conversely, if the temperature is too low, the initial annealing may not provide sufficient densification and / or nitriding in the high-k dielectric gate layer, causing instability / deviations of the device during subsequent metal-gate deposition processes. If no initial conductive layer 83 has been formed, some embodiments do not perform an annealing operation at this stage. In other embodiments, the initial conductive layer 83 is formed, and then an annealing operation is performed before the initial conductive layer 83 is removed by a wet etching process.
[0044] In some embodiments, the stacked structure comprising the interface layer 81, the dielectric gate layer 82, and the first conductive layer 83 is soaked in fluorine-containing gas (e.g., F2 and / or NF3) for approximately 4 seconds to approximately 15 minutes at a temperature ranging from approximately room temperature (25°C) to approximately 55°C. The incorporation of fluorine contributes to appropriately improving the work function matching, reducing the Vt of a PMOS device, passivating oxygen vacancies in the dielectric gate layer 82, reducing leakage currents, and decreasing unpaired bonds in the dielectric gate layer.In some embodiments, a cap layer, for example made of crystalline, polycrystalline, or amorphous silicon, is then formed over the first conductive layer 83, and a second annealing process is carried out for approximately 1 nanosecond (pulse annealing, such as laser annealing) to approximately 360 seconds at a temperature of approximately 550°C to approximately 1300°C. In some embodiments, the annealing temperature is between 900°C and 1100°C. This leads, in some embodiments, to the diffusion of fluorine into the cap layer, the first conductive layer 83, and the dielectric gate layer 82. After the second annealing process, the cap layer is removed. The second annealing with the silicon cap layer also contributes to improving the quality of the dielectric gate layer 82.A dielectric gate layer, such as a high-k dielectric layer, is formed at a relatively low temperature to avoid crystallization and grain boundary formation, while metal gate films are deposited at relatively high temperatures. Consequently, it is desirable to make the high-k dielectric layer more thermally stable before metal gate deposition. The second annealing with the cap layer in the temperature range described above can densify the high-k dielectric layer and make it more thermally stable without any thermal oxide inversion during metal gate deposition. The second annealing also helps to thermally diffuse the fluorine from the outer layers (for example, the cap layer) into the first conductive layer 83, the dielectric gate layer 82, and the interface layer 81.The cap layer is used to protect the dielectric gate layer 82 and the first conductive layer 83 from unwanted oxidation damage and to insulate these films from the annealing atmosphere. After the dielectric gate layer has been thermally stabilized, the cap layer is no longer needed in the final device structure and is therefore removed.
[0045] In other embodiments, no fluorine impregnation process is carried out together with the formation of a Si cap layer and a second tempering process.
[0046] As a result, S309 will be discussed. Fig. 3G formed a second conductive layer as a first barrier layer 84, before then at S311 from Fig. 3G one or more WFM layers 86 are formed. A metal gate layer comprising an adhesive layer 87 and a body metal layer (gate electrode layer) 88 is formed in S313 by Fig. 3G formed above the exit work adaptation material layer 86.
[0047] In some embodiments, the second conductive layer 84 is made of TaN and serves as an etch-stop barrier layer. The barrier layer 86 serves as a wet etch stop layer during the structuring of p- and n-WFM layers, which are subsequently formed to create several Vt devices. In some embodiments, no second conductive layer 84 is formed.
[0048] The output work-matching material (WFM) layer 86 can be formed by ALD, PVD, CVD, electro-beam evaporation, or another suitable process. Furthermore, the WFM layer can be formed separately for the n-channel FET and the p-channel FET, which may use different metal layers. The gate electrode layer (body metal layer) 88 and the adhesive layer 87 can be formed by CVD, ALD, electroplating, or another suitable method. If the first and second conductive layers are not formed, the WFM layer 86 is formed directly on the dielectric gate layer 82. In some embodiments, the first conductive layer 83 is formed and removed after the S307 annealing process, while the second conductive layer is not formed thereafter, and the WFM layer 86 is formed directly on the dielectric gate layer 82.
[0049] Fig. Figure 4A shows a cross-sectional view of gate structures for FETs with different threshold voltages in accordance with an embodiment of the present disclosure. Fig. 4B and Fig. Figure 4C shows various output work matching material layers for multiple FETs with different threshold voltages in accordance with embodiments of the present disclosure.
[0050] In some embodiments, a semiconductor device comprises a first n-FET N1 having a WFM layer structure WF1, a second n-FET N2 having a WFM layer structure WF2, a third n-FET N3 having a WFM layer structure WF3, a first p-FET P1 having the WFM layer structure WF3, a second p-FET P2 having the WFM layer structure WF2, and a third p-FET P3 having the WFM layer structure WF1. The absolute threshold voltage of the first n-FET N1 (ultra-low voltage FET) is lower than the threshold voltage of the second n-FET N2 (low voltage FET), and the absolute threshold voltage of the second n-FET N2 is lower than the threshold voltage of the third n-FET N3 (standard voltage FET).Similarly, the absolute threshold voltage of the first p-FET P1 (ultra-low voltage FET) is lower than the threshold voltage of the second p-FET P2 (low voltage FET), and the absolute threshold voltage of the second p-FET P2 is lower than the threshold voltage of the third p-FET P3 (standard voltage FET). The absolute threshold voltage of the first n-FET N1 is designed to be the same as the threshold voltage of the first p-FET P1, the absolute threshold voltage of the second n-FET N2 is designed to be the same as the threshold voltage of the second p-FET P2, and the absolute threshold voltage of the third n-FET N3 is designed to be the same as the threshold voltage of the third p-FET P3.
[0051] In some embodiments, the WFM layer structure WF1 has a first WFM layer 100, the WFM layer structure WF2 has a second WFM layer 89-2 and the first WFM layer 100 closer to the dielectric gate layer 82, and the third WFM layer structure WF3 has a third WFM layer 89-1, the second WFM layer 89-2 and the first WFM layer 100 closer to the dielectric gate layer 82, as shown in Fig. 4A shown.
[0052] In Fig. 4B, the semiconductor device has three different threshold voltage levels. In other embodiments, such as in Fig. As shown in Figure 4C, more than three, for example eight different threshold voltages are used for an n-FET or a p-FET. Fig. In 4C, not only the WFM layer structures but also the arrangements HK1, HK2, and HK3 of the dielectric gate layer 82 (e.g., material, thickness, etc.) are adapted to achieve a desired threshold voltage. HK1, HK2, and HK3 are composed of various materials, such as HfO2, HfSiO, HfSiON, HfTaO, HfTiO, HfZrO, zirconium oxide, aluminum oxide, titanium oxide, hafnium dioxide-alumina alloy (HfO2-Al2O3 alloy), La2O3, HfO2-La2O3, La2O3, Y2O3, Dy2O3, Sc2O3, MgO, or other suitable high k-dielectric materials and / or combinations thereof. In some embodiments, HK1, HK2 and HK3 are composed of a high k dielectric with different concentrations of rare earth metal and / or dopants of group III (such as La, Al, Mg, Sc, Dy, Y, Ti, Lu, Sr, etc.).In some embodiments, HK3 is composed of HfOx, HK2 is composed of HfLaOx (or HfYOx, HfLuOx, HfSrOx, HfScOx, HfDyOx), and HK1 is composed of HfLaOx (or HfYOx, HfLuOx, HfSrOx, HfScOx, HfDyOx), such that the amount of La (or Y, Lu, Sr, Sc, Dy) in HK1 is greater than in HK2. In some embodiments, HK1 is composed of HfOx, HK2 is composed of HfAlOx (or HfZrOx, HITiOx), and HK3 is composed of HfAlOx (or HfZrOx, HITiOx), such that the amount of Al (or Zr, Ti) in HK3 is greater than in HK2. In some embodiments, HK2 is composed of HfOx, HK1 is composed of HfLaOx (or HfYOx, HfLuOx, HfSrOx, HfScOx, HfDyOx), and HK3 is composed of HfAlOx (or HfZrOx, HffiOx). In some embodiments, HK1 contains a larger amount of La than HK2, and HK3 contains no La. In some embodiments, HK3 contains a larger amount of Al than HK2, and HK1 contains no Al.In some embodiments, HK1 contains La, HK3 contains Al, and HK2 contains neither Al nor La. In some embodiments, the thicknesses of HK1, HK2, and HK3 range from approximately 0.6 nm to approximately 30 nm. In some embodiments, more than three different high-k dielectric films are used.
[0053] In a CMOS device, one gate electrode is often used (shared) for both an n-FET and a p-FET, and thus an n-FET and a p-FET are selected that have essentially the same threshold voltage. For example, a CMOS device containing an ultra-low voltage FET has the first n-FET N1 and the first p-FET P1. Fig. Figure 5A shows a top view (a layout) of such a CMOS device.
[0054] As in Fig. As shown in Figure 5A, a gate electrode 80 is arranged over one or more fin structures 20 (channel regions). In some embodiments, both the n-FET (NFET) and the p-FET (PFET) have two fin structures. In other embodiments, the number of fin structures per FET is one, three, or more (for example, up to 10). Fig. 5A shows a cross-sectional view of area A1 of Fig. 5A corresponds to, and Fig. Figure 5C shows an enlarged view of area B1 of Fig. 5B. In the Fig. 5B and Fig. In 5C, the adhesive layer 87 and the body metal layer 88 (shown as a dashed line) have been omitted.
[0055] As explained above, the n-FET NFET (e.g., N1) has the WFM layer structure WF1 and (only) the first WFM layer 100, and the p-FET (e.g., P1) has the WFM layer structure WF3 and the second and third WFM layers (89-2 and 89-1, which together are in Fig. 5B (labeled 89) and the first WFM layer 100. Consequently, the second and third WFM layers 89 are formed at the boundary MB of the n-FET NFET and the p-FET PFET.
[0056] Similarly, the Fig. 6A - 6C a CMOS device comprising a threshold voltage Vt3, an n-FET with a WFM layer structure WF2 and a p-FET with a WFM layer structure WF3 (see Fig. 4C). Fig. 6B shows a cross-sectional view of area A2 of Fig. 6A corresponds to, and Fig. Figure 6C shows an enlarged view of area B2 of Fig. 6B. In the Fig. 6B and Fig. In 6C, the adhesive layer 87 and the body metal layer 88 have been omitted.
[0057] As in Fig. As shown in Figure 6A, a gate electrode 80 is arranged over one or more fin structures 20 (channel regions). As described above, the n-FET NFET has the WFM layer structure WF2, comprising the first WFM layer 100 and the second WFM layer 89-2, and the p-FET has the WFM layer structure WF3, comprising the second and third WFM layers and the first WFM layer 100. Consequently, the third WFM layer 89-1 is formed with an interrupt at the boundary MB of the n-FET NFET and the p-FET PFET.
[0058] If the first WFM layer 100 contains aluminum, aluminum can diffuse across the boundary MB into the second and / or third WFM layer 89, and can change the threshold voltage of the p-FET PFET (and can change the threshold voltage of the n-FET NFET). In particular, if diffused Al forms aluminum oxide, which has a high electrical resistivity, the performance of the p-FET PFET would be reduced. Furthermore, lateral diffusion of Al into the second WFM layer 89-1 affects the threshold voltage more than vertical diffusion of Al from the top of the first WFM layer 89-2. As integrated circuits become smaller, the distance between adjacent p-FET and n-FET devices decreases, and these FET devices are positioned closer together.Therefore, lateral diffusion via adjacent devices becomes critical and causes serious changes in the threshold voltage as well as a deterioration in performance.
[0059] In accordance with some embodiments of the present disclosure, the WFM layer 100 contains aluminum (e.g., TiAl and TiAlC) and further comprises at least either a lower surface region and / or an upper surface region of the WFM layer 100. In some embodiments, the diffusion barrier layer comprises at least one layer from the group consisting of a Ta-rich, a Ta-doped, or a Si-doped layer. The diffusion barrier layer suppresses the diffusion of aluminum into the WFM layers 89-1 and 89-2 of adjacent FET devices, which may be made, for example, of WCN, WN, Ru, TiN, or TiSiN. In some embodiments, the diffusion barrier layer of the WFM layer 100 comprises a layer from the group comprising a Ti-rich TiAl layer, a Ti-doped TaAl layer, a Ta-rich TaAl layer, a Ta-doped TiAl layer, a Si-doped TiAl layer and a Si-doped TaAl layer.
[0060] In some embodiments, the body (central region) of the WFM layer 100 is made of TiAl or TiAlC and contains a Ti content of 5 atomic percent to approximately 15 atomic percent and an Al content ranging from approximately 10 atomic percent to approximately 30 atomic percent, while the remainder in some embodiments contains carbon (e.g., approximately 8 atomic percent to approximately 25 atomic percent) and / or oxygen. In some embodiments, the Al / Ti ratio ranges from approximately 1.5 to approximately 3.0.
[0061] In some embodiments, the diffusion barrier layer contains more titanium than the central region adjacent to the diffusion barrier layer(s), with the amount of Ti ranging from approximately 20 atomic percent to approximately 50 atomic percent, and in other embodiments ranging from approximately 30 atomic percent to 40 atomic percent. In some embodiments, the Al / Ti ratio of the diffusion barrier layer ranges from approximately 0.2 to approximately 1.2. In some examples, the Al / Ti ratio of the Ti-rich layer ranges from approximately 0.2 to approximately 1.2 when the diffusion barrier layer is a Ti-rich layer.In some embodiments, when the diffusion barrier layer is a Ta-doped layer, the Ta-doped layer contains more tantalum than the central region adjacent to the Ta-rich layer(s), and the amount of Ta is in the range of approximately 20 atomic percent to approximately 50 atomic percent, and in other embodiments in the range of approximately 30 atomic percent to 40 atomic percent. In some embodiments, the Al / Ta ratio of the Ta-doped layer is in the range of approximately 0.2 to approximately 1.2. The Ta-doped TiAl layer is formed by an ALD, PEALD, or CVD process with cyclically alternating pulses and purges of Ta precursor (e.g., PDMAT / TaCl5), Al precursor, and Ti precursor, and Ar / He inert gas purging. In some embodiments, the process temperature is in the range of approximately 300°C to approximately 550°C.In some embodiments, where the diffusion barrier layer is a silicon-doped layer, the silicon-doped layer contains more silicon than the central region adjacent to the silicon-rich layer(s), and the amount of silicon is in the range of approximately 20 atomic percent to approximately 50 atomic percent, and in other embodiments in the range of approximately 30 atomic percent to 40 atomic percent. In some embodiments, the aluminum / silicon ratio of the silicon-doped layer is in the range of approximately 0.2 to approximately 1.2. The silicon-doped TiAl layer is formed by an ALD, PEALD, or CVD process with cyclically alternating shocks and purges of silicon precursors (SiCl₄, SiHCl₃, etc.), aluminum precursors, and titanium precursors, and argon / helium inert gas purges. In some embodiments, the process temperature is in the range of approximately 300°C to approximately 550°C.
[0062] As in the Fig. 7A and Fig. As shown in Figure 7B, in some embodiments the WFM layer 100 has an upper diffusion barrier layer 140 and a lower diffusion barrier layer 145. Fig. Figure 7B shows a structure at the boundary MB of an n-FET and a p-FET similar to Fig. 5C. As in the Fig. 7C and Fig. As shown in Figure 7D, in some embodiments the WFM layer 100 has only a lower diffusion barrier layer 145. In these arrangements, at least the lower diffusion barrier layer 145 can suppress aluminum diffusion from the WFM layer 100 of one of the FETs into the WFM layers 89-1 and / or 89-2 of adjacent FETs. Without the diffusion barrier layers, aluminum diffusion occurs from the metal boundaries of a left FET and a right FET. Due to the loss of aluminum from the WFM 100 of the left FET, the effective work function of the left FET increases, and due to the increase of aluminum in the WFM layers 89-1 and / or 89-2 of the right FET, the overall work function of the right FET decreases. However, by using one or more diffusion barrier layers, such Al diffusion across metal boundaries of neighboring FETs can be suppressed, resulting in better isolation of neighboring devices.
[0063] As in the Fig. 7E and Fig. As shown in Figure 7F, in some embodiments the WFM layer 100 further comprises only an upper diffusion barrier layer 140. In such a case, the upper diffusion barrier layer 140 improves the quality of the TiAl or TiAlC mass layer (central region of the WFM layer 100), which in turn can also suppress aluminum diffusion from the WFM layer 100 into the WFM layers 89-1 and / or 89-2. Although the suppression mechanism of Al diffusion is not entirely clear, improvements in the electrical properties of the devices have been observed.
[0064] The Fig. 8A and Fig. Figure 8B shows Ta or Si or Ti profiles in the WFM layer 100 having the upper and lower diffusion barrier layers in accordance with some embodiments.
[0065] In some embodiments, such as in Fig. As shown in Figure 8A, the concentration of Ta, Si, or Ti in a substantially constant section S2 has a concentration of X1 atomic percent, and the concentration of Ta, Si, or Ti gradually increases to X2 atomic percent at the edges (lower and upper surfaces). In some embodiments, the Ta, Si, or Ti concentration increases linearly, and in other embodiments, the Ta, Si, or Ti concentration increases gradually (non-linearly). Fig. 8B shows a stepped profile for Ta, Si, or Ti concentration.
[0066] In some embodiments, when the WFM layer 100 is made of TiAl or TiAlC, the Ti concentration ranges from approximately 5 atomic percent to approximately 15 atomic percent, and in other embodiments from approximately 8 atomic percent to approximately 12 atomic percent; the Al concentration ranges from approximately 10 atomic percent to approximately 30 atomic percent; and the remainder is carbon and / or oxygen. In some embodiments, the Al / Ti ratio ranges from approximately 1.5 to approximately 4.8. When the diffusion barrier layer is a Ti-rich layer, the Ti concentration of the Ti-rich layer ranges from approximately 20 atomic percent to approximately 50 atomic percent in some examples, and the Al / Ti ratio ranges from approximately 0.2 to approximately 1.2. If the diffusion barrier layer is a Ta-doped layer, the Ta concentration of the Ta-doped layer is in a range of approximately 20 atomic% to approximately 50 atomic%.If the diffusion barrier layer is a Si-doped layer, the Si concentration of the Si-doped layer is in a range of approximately 20 atomic% to approximately 50 atomic%.
[0067] If the WFM layer 100 is made of TaAl or TaAlC, the Ta concentration in some embodiments ranges from approximately 5 atomic percent to approximately 15 atomic percent, and in other embodiments from approximately 8 atomic percent to approximately 12 atomic percent; the Al concentration ranges from approximately 10 atomic percent to approximately 30 atomic percent; and the remainder is carbon and / or oxygen. In some embodiments, the Al / Ta ratio ranges from approximately 1.5 to 3.0. If the diffusion barrier layer is a Ta-rich layer, the Ta concentration of the Ta-rich layer in some embodiments ranges from approximately 20 atomic percent to approximately 50 atomic percent, and the Al / Ta ratio ranges from approximately 0.2 to approximately 1.2. If the diffusion barrier layer is a Ti-doped layer, the Ti concentration of the Ti-doped layer is in a range of approximately 20 atomic% to approximately 50 atomic%.If the diffusion barrier layer is a Si-doped layer, the Si concentration of the Si-doped layer is in a range of approximately 20 atomic% to approximately 50 atomic%.
[0068] In some embodiments, the atomic percentage X1 of Ti is in the range of approximately 5 to approximately 15 nm, and in other embodiments, it is in the range of approximately 8 nm to approximately 15 nm. In some embodiments, the atomic percentage X1 of Ti is in the range of approximately 0 to approximately 2 nm, wherein the WFM layer is 100% TaAl or TaAlC. In some embodiments, the atomic percentage X1 of Ti is in the range of approximately 0 to approximately 2 nm, wherein the WFM layer is 100% TaAl or TaAlC, and in other embodiments, it is in the range of approximately 5 to approximately 15 nm. In some embodiments, the atomic percentage X1 of Si in a Si-doped layer is in the range of approximately 0 to approximately 5 nm. In some embodiments, X2 is in the range of approximately 20 to approximately 50 nm, and in other embodiments, it is in the range of approximately 30 to approximately 40 nm.In some embodiments, a silicon layer that does not contain aluminum is used as the silicon-rich layer. In some embodiments, the thickness of the constant section S2 is at least approximately 20% of the total thickness of the WFM layer 100, and in other embodiments, it is at least approximately 40% of the total thickness of the WFM layer 100. In some embodiments, the thickness of the constant section S2 is at most approximately 90% of the total thickness of the WFM layer 100, and in other embodiments, it is at most approximately 60% of the total thickness of the WFM layer 100.
[0069] In some embodiments, the thickness of the diffusion barrier layers 140 / 145 is adjusted based on the Ta, Si, or Ti concentration of the diffusion barrier layer. If the Ta, Si, or Ti concentration in the diffusion barrier layer 140 / 145 is high, the thickness of the diffusion barrier layer 140 / 145 can be low. The thickness and / or the Ta / Si / Ti concentration of the diffusion barrier layers in the WFM layer 100 are adjusted in some embodiments based on the desired work function of the WFM layer 100.
[0070] In some embodiments, the diffusion barrier layers 140 / 145 can be formed by an ALD process. In the ALD process, at least one of the parameters comprising the source gas (precursor) quantity, source gas (precursor) flow time, gas flow ratio, deposition temperature, and deposition pressure is varied to obtain the Ta / Si / Ti concentration profile described above.
[0071] Fig. Figure 9 shows a process flow, and the Fig. Figures 10A-10F show cross-sectional views of various stages in the fabrication of the semiconductor device in accordance with embodiments of the present disclosure. It is understood that in the sequential fabrication process, one or more additional operations are performed before, during, and after the stages described in the Fig. 9 and Fig. Stages shown in 10A - 10F can be provided, and some of the operations described below can be substituted or omitted for additional embodiments of the method. The sequence of operations / processes can be interchangeable. In the following embodiments, materials, arrangements, dimensions, processes, and / or operations which relate to embodiments of the Fig. 1-8B are described, and a detailed description of them can be omitted. Although the Fig. 10A - 10F which in relation to Fig. 7A and Fig. 8A explained WFM layer 100 show, can any in Fig. 7C, Fig. 7E and Fig. The structure shown in Figure 8B can be used as the WFM layer 100.
[0072] S901 of Fig. 9. Channel regions of the fin structures are exposed for a first n-FET N1, a second n-FET N2, a third n-FET N3, a first p-FET P1, a second p-FET P2, and a third p-FET P3, respectively. In some embodiments, the channel regions for the n-FETs are made of Si, and the channel regions for the p-FETs are made of SiGe.
[0073] S902 from Fig. 9. Interfacial layers 81N and 81P are formed on each of the channel regions 20 using chemical oxidation processes. If the channel is made of Si, the interfacial layer is, in some embodiments, a silicon oxide layer 81N, and if the channel is made of SiGe, the interfacial layer is a silicon-germanium oxide layer 81P. In S903 of Fig. 9 A dielectric gate layer (e.g., a dielectric gate layer with high k) 82 is formed on the interface layer 81N / 81P. In some embodiments, different dielectric gate layers are formed on different FET devices. In some embodiments, at S904 of Fig. 9 a cap layer 83 is formed on the dielectric gate layer 82, followed by an annealing process at S905, and at S906 by Fig. 9. A barrier layer 84 is formed on the cap layer 83. In other embodiments, neither a cap layer nor a barrier layer is formed. In other embodiments, a cap layer is formed, tempering is carried out, and then the cap layer is removed, and no barrier layer is formed.
[0074] S907 from Fig. 9 a third WFM layer 89-1 is formed on the dielectric gate layer 82, as in Fig. Figure 10A shows the third WFM layer 89-1 containing WN, WCN, W, Ru, TiN, or TiSiN, for example formed by CVD, ALD, PVD, or any other suitable film formation process. In some embodiments, the thickness of the third WFM layer 89-1 is in the range of approximately 0.5 nm to approximately 20 nm, and in other embodiments, it is in the range of approximately 1 nm to approximately 10 nm.
[0075] S908 from Fig. 9. A first structuring process is performed to remove the third WFM layer 89-1 from the areas of the first n-FET N1, the second n-FET N2, the second p-FET P2, and the third p-FET P3. In some embodiments, a lower antireflection coating layer 200 made of an organic material is formed on the third WFM layer 89-1, and a photoresist layer 205 is formed on the lower antireflection coating layer 200, as shown in Fig. Figure 10A shows that by using one or more lithography processes, the photoresist layer 205 is patterned to expose the lower antireflection coating layer 200 in the areas for the first and second n-FETs and the second and third p-FETs. The exposed lower antireflection coating layer 200 is then removed by one or more plasma etching processes to expose the third WFM layer 89-1 in the areas for the first and second n-FETs and the second and third p-FETs, as shown in Figure 10A. Fig. Figure 10A shows that the plasma etching process uses a gas containing N2 and H2, a gas containing O2 / Cl2, and / or an O2 gas. In some embodiments, no lower antireflective layer is used, and a photoresist layer made of organic materials is formed on the third WFM layer 89-1.
[0076] Subsequently, the third WFM layer 89-1 in the first and second n-FETs and in the second and third p-FETs is removed by a suitable etching process, as shown in Fig. Figure 10B shows that in some embodiments, the etching process includes a wet etching process. The etching solution (the etchant) contains an aqueous solution of HCl and H₂O₂, an aqueous solution of the combination of NH₄OH and H₂O₂, an aqueous solution of the combination of HCl, NH₄OH and H₂O₂, an aqueous solution of HF, NH₄OH and H₂O₂ and / or an aqueous solution of H₃PO₄ and H₂O₂.
[0077] S909 from Fig. 9 The second WFM layer 89-2 is formed on the dielectric gate layer 82 in the areas for the first and second n-FETs and the second and third p-FETs, as well as on the third WFM layer 89-1 in the areas for the third n-FET and the first p-FET, as shown in Fig. Figure 10C shows the second WFM layer 89-2, which contains WN, WCN, W, Ru, TiN, or TiSiN, for example, formed by CVD, ALD, PVD, or any other suitable film formation process. In some embodiments, the thickness of the second WFM layer 89-2 is in the range of approximately 0.5 nm to approximately 20 nm, and in other embodiments, it is in the range of approximately 1 nm to approximately 10 nm. In some embodiments, the second and third WFM layers are made of the same material.
[0078] S910 from Fig. 9. A second structuring operation is performed to remove the second WFM layer 89-2 from the areas for the first n-FET N1 and the third p-FET P3. The second structuring operation is essentially the same as, or similar to, the first structuring operation. Fig. Figure 10D shows the structure after removal of the second WFM layer 89-2 from the areas for the first n-FET N1 and the third p-FET P3.
[0079] S911 from Fig. 9 The first WFM layer 100 is formed on the dielectric gate layer 82 in the areas for the first n-FET and the third p-FET, as well as on the second WFM layer 89-2 in the areas for the second and the third n-FET and the first and the second p-FET, as shown in Fig. Figure 10E shows that the first WFM layer 100 is formed, for example, by ALD in some embodiments. In some embodiments, the thickness of the first WFM layer 100 ranges from approximately 0.6 nm to approximately 40 nm, and in other embodiments, it ranges from approximately 1 nm to approximately 20 nm. In some embodiments, the thickness of the first WFM layer 100 is greater than that of both the second and the third WFM layer.
[0080] S912 of Fig. 9. A cap layer 110 is formed on the first WFM layer 100. In some embodiments, the cap layer 110 has one or more metal nitride layers, such as TiN, TaN, TaTiN, WN, TiSiN, WCN, and MoN. In other embodiments, no cap layer is formed.
[0081] S913 of Fig. 9 an adhesive layer 87 is formed before at S914 of Fig. 9 a body-gate metal layer 88 is formed, as in Fig. Figure 10E shows that in some embodiments the adhesive layer 87 is made of TiN, Ti and / or Co. In some embodiments the body metal layer 88 is made of W, Al, Co or any other suitable metallic material.
[0082] As explained above, the Ti-rich layer of the first WFM layer suppresses Al diffusion from WFM layer 100 to the underlying WFM layers 89 and the dielectric gate layer 82 and / or the adjacent FETs, as shown in Fig. 11 shown.
[0083] Furthermore, the use of the diffusion barrier layer can also reduce the thickness of the interface layer 81 by removing oxygen from the interface layer 81. Fig. 12A and Fig. Figure 12B shows thickness reduction effects of an interface layer in accordance with embodiments of the present disclosure.
[0084] As in Fig. 12A and Fig.As shown in Figure 12B, with increasing thickness of the diffusion barrier layer and / or increasing Ti / Al ratio, Ta amount, or Si amount in the diffusion barrier layer, the thickness of the interface layer decreases, which improves the operating speed of the device. In some embodiments, the thickness of the interface layer can be reduced to approximately 0.6 nm or less (but more than zero).
[0085] It is understood that not all advantages have necessarily been discussed herein, that no particular advantage is required for all embodiments or examples, and that other embodiments or examples may offer different advantages. The invention is specified in the independent claims. Specific embodiments are described in the dependent claims.
[0086] In one or more of the preceding and following embodiments, the formation of the Ti-doped layer comprises the gradual modification of a source gas, at least for either aluminum and / or titanium, during TiAl deposition, or by introducing a Ti source precursor during TaAl or TaAlC deposition. In one or more of the preceding and following embodiments, the formation of the Ta-rich layer or the Ta-doped layer comprises the gradual modification of a source gas by introducing a Ta source precursor during TiAl or TiAlC deposition. In one or more of the following and preceding embodiments, the formation of the Si-doped layer comprises the introduction of a third Si source precursor during the cyclic ALD deposition of TiAl, such as by performing a cyclic pulse and purge process with Al-based precursors, Si-based precursors, and Ti- (or Ta-) based precursors.
Claims
[1] A semiconductor device comprising: a gate structure arranged over a channel region; and a source / drain region, where: the gate structure has: a gate dielectric layer (82) over the channel region; one or more work function matching material layers (100) over the gate dielectric layer (82); and a metal gate electrode layer (88) over the one or more work function matching material layers (100); wherein one of the one or more work function matching material layers (100) is an aluminum-containing layer made of a material selected from the group comprising TaAl, TaAlC, TiAl and TiAlC, and wherein a diffusion barrier layer (140, 145) is arranged in the aluminum-containing layer and at least at either a lower portion and / or an upper portion of the aluminum-containing layer, and the diffusion barrier layer (140, 145) is one or more selected from the group comprising: a) a Ti-doped layer if the material is TaAl or TaAlC, wherein a central portion of the aluminum-containing layer has a lower Ti concentration than the diffusion barrier layer, b) a Ta-rich layer having a higher Ta concentration than the central portion of the aluminum-containing layer when the material is TaAl or TaAlC, c) a Ta-doped layer if the material is TiAl or TiAlC, wherein the central portion of the aluminum-containing layer has a lower Ta concentration than the diffusion barrier layer, and d) a Si-doped layer if the material is TaAl or TaAlC or TiAl or TiAlC, wherein the central portion of the aluminum-containing layer has a lower Si concentration than the diffusion barrier layer (140, 145); and the Ti, Ta or Si concentration of the diffusion barrier layer (140, 145) is in a range of 20 atomic% to 50 atomic%. [2] The semiconductor device of claim 1, wherein the aluminum-containing layer has a work function of less than 4.4 eV. [3] A semiconductor device according to claim 1, wherein the Ti, Ta or Si concentration in the diffusion barrier layer (140, 145) gradually increases from a side at the central portion to a surface of the aluminum-containing layer. [4] A semiconductor device according to claim 1, wherein the Ti, Ta or Si concentration in the diffusion barrier layer (140, 145) is constant. [5] A semiconductor device according to any one of claims 1 to 4, wherein a Ti or Ta concentration in the central portion is constant. [6] A semiconductor device according to any one of claims 1 to 5, wherein an aluminum concentration of the center portion is in a range of 10 atomic% to 30 atomic%. [7] A semiconductor device according to any one of claims 1 to 6, wherein a thickness of the central portion is in a range of 20% to 60% of a total thickness of the aluminum-containing layer. [8] A semiconductor device according to any preceding claim, wherein the plurality of work function matching material layers (100) further comprise a work function material layer (89-1) having a work function of more than 4.4 eV and at least one of WCN, WN, W, Ru, TiN, TiNC or TiSiN disposed between the aluminum-containing layer and the gate dielectric layer (82). [9] Complementary metal oxide semiconductor (CMOS) device, comprising: a first field-effect transistor (FET), which is a semiconductor device according to any one of the preceding claims, wherein the gate structure is a first gate structure, the channel region is a first channel region, and the aluminum-containing layer is a first work function matching material layer; and a second FET having a second gate structure disposed over a second channel region, wherein: the first FET is an n-FET (NFET) and the second FET is a p-FET (PFET), and the second gate structure has: a dielectric gate layer (82), a second work function matching material layer (89-1,89-2) over the gate dielectric layer (82), the first work function matching material layer (100) disposed over the second work function matching material layer (89-1, 89-2); and the metal gate electrode layer (88) disposed over the first work function matching material layer (100); the metal gate electrode layer (88) and the first work function adjustment material layer (100) are arranged continuously between the first FET and the second FET, such that the metal gate electrode (88) is arranged over the dielectric gate layer (82) of the first FET, over the second work function adjustment material layer (89-1, 89-2) of the second FET, and over a sidewall of the second work function adjustment material layer (89-1, 89-2) at a boundary (MB) of the first FET and the second FET, wherein the first work function adjustment material layer (100) covers a top surface and a side surface of the second work function adjustment material layer (89-1, 89-2) at the boundary (MB) of the first FET and the second FET. [10] The CMOS device of claim 9, wherein the second work function matching material layer (89-1, 89-2) contains at least one of WN, WCN, W, Ru, TiN, and TiSiN. [11] A CMOS device according to claim 9, wherein: the second work function matching material layer (89-1, 89-2) contains at least one of the materials WN, WCN, W, Ru, TiN and TiSiN; and the second work function matching material layer (89-1, 89-2) is interrupted at the boundary (MB) between the first FET and the second FET. [12] The CMOS device of any one of claims 9 to 11, wherein the diffusion barrier layer (140, 145) is disposed at the lower portion of the first work function matching material layer (100). [13] A CMOS device according to any one of claims 9 to 12, wherein the second work function matching material layer (89-1, 89-2) in the second FET does not contain Al. [14] The CMOS device of any one of claims 9 to 13, wherein the first work function matching material layer (100) is disposed directly on the gate dielectric layer (82). [15] The CMOS device according to any one of claims 9 to 14, wherein a thickness of the central portion is in a range of 20% to 90% of a total thickness of the first work function matching material layer (100). [16] A CMOS device according to any one of claims 9 to 15, wherein the first gate structure further comprises: an interface layer (81, 81N, 81P), and wherein the gate dielectric layer (82) of the first FET is arranged on the interface layer (81, 81N, 81P). [17] A method of manufacturing a semiconductor device, comprising: Forming (S903) a gate dielectric layer (82) over a channel region made of a semiconductor material; Forming (S911) a first work function matching material layer (100) over the gate dielectric layer (82); and Forming (S914) a metal gate electrode layer (88) over the first work function matching material layer (100), wherein: the first work function adjustment material layer is an aluminum-containing layer made of a material selected from the group comprising TaAl, TaAlC, TiAl and TiAlC, forming (S911) the first work function matching material layer (100) comprises forming a diffusion barrier layer (140, 145) at least in a lower portion and / or in an upper portion of the first work function matching material layer (100), wherein the diffusion barrier layer (140, 145) is one or more from the group comprising: a) a Ti-doped layer if the material is TaAl or TaAlC, wherein a central portion of the aluminum-containing layer has a lower Ti concentration than the diffusion barrier layer, b) a Ta-rich layer having a higher Ta concentration than the central portion of the aluminum-containing layer when the material is TaAl or TaAlC, c) a Ta-doped layer if the material is TiAl or TiAlC, wherein the central portion of the aluminum-containing layer has a lower Ta concentration than the diffusion barrier layer, and d) a Si-doped layer if the material is TaAl or TaAlC or TiAl or TiAlC, wherein the central portion of the aluminium-containing layer has a lower Si concentration than the diffusion barrier layer (140, 145); and wherein: the Ti, Ta or Si concentration of the diffusion barrier layer (140, 145) is in a range of 20 atomic% to 50 atomic%.
Citation Information
Patent Citations
SRAM (Static Random Access Memory) device and manufacturing method thereof
CN108122912A
Metal gate, semiconductor device and manufacturing method thereof
CN109786448A
Interconnection structure for n / p metal gates
US20120012937A1
N Metal for FinFET
US20140167187A1
Semiconductor device and method for fabricating the same
US20150014780A1
