Self-aligning contact and manufacturing process for it
The method of forming self-aligned contacts in MuGFETs addresses stray capacitance issues in MOSFETs with reduced gate lengths, improving performance by minimizing misalignment and contact resistance.
Patent Information
- Application Number
- DE102016125144
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2016-12-01
- Filing Date
- 2016-12-21
- Publication Date
- 2024-11-28
- Estimated Expiration
- 2036-12-21
AI Technical Summary
Short-channel effects (SCEs) in metal-oxide-semiconductor field-effect transistors (MOSFETs) with reduced gate lengths lead to high off-state leakage current and drain-induced turn-off, which are not effectively addressed by existing multi-gate field effect transistor (MuGFET) architectures.
A method for forming self-aligned contacts in MuGFETs, such as double-gate or triple-gate FinFETs, involves creating a sacrificial inter-gate film that is selectively etched to reduce stray capacitance between the metal gate and contacts, using a process that includes forming a sacrificial film and replacing it with a conductive material to ensure precise alignment and minimize misalignment issues.
This approach reduces stray capacitance and improves device performance by allowing for smaller contact areas and larger contact areas without misalignment, thereby enhancing current density and reducing contact resistance.
Smart Images

Figure 00000000_0000_ABST
Abstract
Description
BACKGROUND
[0001] Metal-oxide-semiconductor field-effect transistors (FETs or MOSFETs) are widely used in integrated circuits (ICs). To increase the density of MOSFETs in ICs, the physical dimensions, such as the gate length (LG), of MOSFETs are drastically reduced. MOSFETs with short LGs can experience undesirable short-channel effects (SCEs), such as high off-state leakage current and high drain-induced cutoff.
[0002] To suppress SCEs in transistors with short gate lengths LG, the multi-gate field-effect transistor (MuGFET) architecture can be used. A MuGFET has better electrostatic control of the channel potential by the gate electrode than a planar device structure. Examples of MuGFETs include the double-gate transistor and the tri-gate or triple-gate transistor. A double-gate transistor is also known as a double-gate FinFET. A tri-gate transistor may be referred to as a tri-gate FinFET or simply FinFET. Double-gate or tri-gate devices operate with a channel that resembles a fin. The on-state or saturation drive current IDsat flows in a fin to achieve a high current density per footprint or layout area.
[0003] Other MuGFETs include pi-gate, omega-gate, surround-gate (SG), or gate-all-around (GAA) structures, where electrostatic gate control is further enhanced. The SG transistor has a channel resembling a nanowire, which can be oriented horizontally or vertically. For a horizontal nanowire transistor, multiple horizontally oriented nanowire channels can be stacked vertically.
[0004] For the state of the art, reference is made to the documents US 2015 / 0 069 532 A1, US 2015 / 0 145 027 A1, US 2015 / 0 333 148 A1 and US 9 318 581 B1.
[0005] The invention provides a method according to claim 1, a method according to claim 8, and a semiconductor device according to claim 17. Embodiments are specified in the dependent claims. BRIEF DESCRIPTION OF THE DRAWINGS
[0006] Aspects of the present disclosure are best understood from the following detailed description when read in conjunction with the accompanying figures. It should be noted that, in accordance with standard industry practice, various features are not drawn to scale. Rather, the dimensions of the various features may be arbitrarily exaggerated or reduced to facilitate discussion. Fig. 1A-13D illustrate various intermediate process steps in a manufacturing process according to some embodiments. Fig. 14A-17B illustrate various profiles that may be used in combination with other disclosed embodiments. Fig. 18A-28D illustrate various intermediate process steps in a manufacturing process according to some embodiments. DETAILED DESCRIPTION
[0007] It should be understood that this disclosure provides many embodiments or examples for implementing this invention. Specific examples are described below, which are not intended to be limiting. The following disclosure provides many different embodiments or examples for implementing various features of the invention. Specific examples of components and arrangements are described below to simplify the present disclosure. These are, of course, merely examples, and are not intended to be limiting.For example, the formation of a first feature over or on top of a second feature in the following description may include embodiments where the first and second features are formed in direct contact, and may also include embodiments where additional features may be formed between the first and second features such that the first and second features may not be in direct contact. Furthermore, the present disclosure may repeat reference numbers and / or letters throughout the various examples. This repetition is for the purpose of simplicity and clarity and does not automatically imply a relationship between the various embodiments and / or configurations discussed.
[0008] Furthermore, spatially relative terms such as "beneath," "under," "lower," "above," "upper," and the like may be used herein for convenience of description to describe the relationship of one element or feature to one or more other elements or features, as illustrated in the figures. The spatially relative terms are intended to encompass various orientations of the device in use or operation besides the orientation shown in the figures. The device may also be oriented differently (rotated 90 degrees, or other orientations), and the spatially relative descriptors used herein may also be interpreted accordingly.
[0009] Disclosed embodiments relate to a method for forming a contact for multi-gate field-effect transistors, including a method for forming self-aligned contacts and reducing stray capacitance between the gate and the contact for such devices with metal gate electrodes. Embodiments such as those disclosed herein are generally applicable to dual-gate or triple-gate FinFETs, surround-gate or gate-all-around transistors, and / or nanowire transistors.
[0010] As discussed in more detail below, embodiments are disclosed that provide a process flow for integrating a metal gate electrode and a self-aligned contact for multi-gate field-effect transistors or MuGFETs (double-gate, triple-gate FinFETs, and horizontal gate-all-around transistors), but can also be applied to other structures. Disclosed embodiments can reduce the stray capacitance between the metal gate and the contact for multi-gate field-effect transistors or MuGFETs (double-gate, triple-gate FinFETs, and horizontal gate-all-around transistors).
[0011] For illustrative purposes, the figures and the following discussion refer to a single fin and three gates. However, it should be noted that other embodiments may use multiple fins and fewer or more gates. Furthermore, other structures may also be used. For example, epitaxial regions of adjacent fins may be connected to form a single, larger source / drain region.
[0012] Fig. 1A-13C are cross-sectional and perspective views of intermediate stages in the fabrication of a FinFET and the respective contacts according to some embodiments. Variations of the embodiments are discussed. Like reference numerals are used to denote like elements throughout the various views and illustrative embodiments. Fig. 1A-13C illustrate the “A” figures (for example Fig. 1A, Fig. 2A etc.) a perspective view, the “B” figures (for example Fig. 1B, Fig. 2B etc.) illustrate a cross-sectional view along the Y-cut line illustrated in the respective “A” figure, and the “C” figures (for example Fig. 1C, Fig. 2C, etc.) illustrate a cross-sectional view along the X-cut line illustrated in the respective “A” figure.
[0013] We first turn to the Fig. 1A-1C, where a substrate 102 is shown having one or more fins, with a single fin 104 being illustrated. It should be understood that while a single fin is shown for illustrative purposes, other embodiments may include any number of fins. The substrate 102 may include different doped regions (e.g., a p-type substrate or an n-type substrate) depending on the design requirements. In some embodiments, the doped regions may be doped with p-type or n-type dopants. For example, the doped regions may be doped with p-type dopants, such as boron or BF2; may be doped with n-type dopants, such as phosphorus or arsenic; and / or combinations thereof. The doped regions may be configured for an n-type FinFET, or alternatively, they may be configured for a p-type FinFET.
[0014] In some embodiments, the substrate 102 may be made of any other suitable elemental semiconductor, such as diamond or germanium; a suitable compound semiconductor, such as gallium arsenide, silicon carbide, indium arsenide, or indium phosphide; or a suitable alloy semiconductor, such as silicon germanium carbide, III-V compound semiconductors (e.g., gallium arsenide, indium gallium arsenide, InmGa1-mAs, indium arsenide, indium phosphide, indium antimonide, gallium arsenophosphide, or gallium indium phosphide), or the like. Furthermore, the substrate 102 may include an epitaxial (epi) layer, may be expanded for performance enhancement, and / or may include a silicon-on-insulator (SOI) structure.
[0015] For example, the ridge 104 may be formed using a patterning process to form trenches 108 such that the ridge 104 is formed between adjacent ridges. In some embodiments, photolithography techniques are used to pattern a masking layer (not shown). Generally, a photoresist material (not shown) is deposited over the masking layer. The photoresist material is irradiated (exposed) with energy, for example, light, through a patterned reticle to induce a reaction in those portions of the photoresist material that were exposed to the energy. The photoresist material is developed to remove a portion of the photoresist material, with the remaining photoresist material protecting the underlying material from subsequent processing steps, such as etching.As discussed in more detail below, fin 104 is used to form a MuGFET. An example of the crystal orientation of a channel sidewall surface may be (110), and a channel top surface may be (001). The channel sidewall surface may have other crystal orientations, such as (551). The channel sidewall and the channel top may have different crystal orientations.
[0016] Isolation regions, such as shallow trench isolations (STIs) 106, are formed along sidewalls of fin 104 in trenches 108. Prior to forming STIs 106, one or more liners (collectively referred to as a liner 110) are formed over substrate 102 and the sidewalls of fins 104. In some embodiments, liner 110 has a single-layer structure with a thickness between approximately 1 nm (10 Å) and approximately 5 nm (50 Å). In other embodiments, liner 110 has a two-layer structure including a first liner sublayer and a second liner sublayer. In some embodiments, the first liner sublayer comprises silicon oxide and has a thickness between about 0.5 nm (5 Å) and about 2 nm (20 Å), and the second liner sublayer comprises silicon nitride and has a thickness between about 0.5 nm (5 Å) and about 3 nm (30 Å).The liner 110 may be deposited using one or more processes such as physical vapor deposition (PVD), chemical vapor deposition (CVD), or atomic layer deposition (ALD), although any acceptable process may be used. In some embodiments, the trenches 108 have a depth from a top of the ridge 104 of approximately 50 nm (500 Å) to approximately 300 nm (3000 Å). Other materials, dimensions, and / or processes may also be used.
[0017] The STIs 106 may be formed from suitable dielectric materials, such as silicon oxide, silicon nitride, silicon oxynitride, fluoride-doped silicate glass (FSG), low-k dielectrics such as carbon-doped oxides, ultra-low-k dielectrics such as porous carbon-doped silicon dioxide, a polymer such as polyimide, combinations thereof, or the like. In some embodiments, the STIs 106 are formed by a process such as CVD, flowable CVD (FCVD), or a spin-on-glass process, although any acceptable process may be used. Subsequently, portions of the STIs 106 that extend above the tops of the fins 104 and portions of the liner 110 above the tops of the fins 104 are removed, for example, using an etching process, chemical mechanical polishing (CMP), or the like.
[0018] In some embodiments, the STI 106 and the liner 110 are provided with recesses to expose side walls of the ribs 104, as shown in the Fig. 1A-1C. In some embodiments, the STI 106 and the liner 110 are recessed using one or more selective etch processes using the fins 104 as an etch mask. For example, the STI 106 and the liner 110 are recessed using a single etch process. In alternative embodiments, the STI 106 and the liner 110 are recessed using multiple etch processes. For example, the STI 106 is recessed using a first etch process using the fins 104 and the liner 110 as an etch mask, and then the liner 110 is recessed using a second etch process. In embodiments where the STI 106 comprises silicon oxide, the etch process may be, for example, a dry etch, a chemical etch, or a wet cleaning process. For example, chemical etching can use a fluorine-containing chemical, such as dilute hydrogen fluoride (dHF) acid.After the rib formation process, the rib height H. fin 30 nm or higher, such as 50 nm or higher. It is understood that the fin height can be modified through subsequent processing. Other materials, processes, and dimensions can also be used.
[0019] The Fig. 1A-1C further illustrate a dummy gate dielectric 112 and a dummy gate electrode 114 formed over the exposed fin 104, according to some embodiments. The dummy gate dielectric 112 and the dummy gate electrode 114 are subsequently used to define and form the source / drain regions. In some embodiments, the dummy gate dielectric 112 and the dummy gate electrode 114 are formed by depositing and patterning a dummy gate dielectric layer (not shown) formed over the exposed fin 104 and a dummy gate electrode layer (not shown) over the dummy gate dielectric layer. The dummy gate dielectric layer may be formed by thermal oxidation, CVD, sputtering, or other methods known to and used by those skilled in the art to form a dummy gate dielectric layer.In some embodiments, the dummy gate dielectric layer may be formed from a same material as the STIs 106. In other embodiments, the dummy gate dielectric 112 may be composed of one or more suitable dielectric materials, such as silicon oxide, silicon nitride, SiCN, SiON, Si3N4, and SiN. x H y , low-k dielectrics such as carbon-doped oxides, ultra-low-k dielectrics such as porous carbon-doped silicon dioxide, a polymer such as polyimide or the like, or a combination thereof. In some embodiments, the dummy gate dielectric layer contains dielectric materials with a high dielectric constant (k value), for example, greater than 3.9. The materials may be metal oxides such as HfO 2, HfZrO x , HfSiO x , HfTiO x , HfAlO xor the like or combinations and multilayers thereof.
[0020] Subsequently, the dummy gate electrode layer is formed over the dummy gate dielectric layer. In some embodiments, the dummy gate electrode layer is a conductive material and may be selected from a group including amorphous silicon, polysilicon, amorphous germanium, polygermanium, amorphous silicon germanium, polysilicon germanium, metallic nitrides, metallic silicides, metallic oxides, and metals. In one embodiment, the dummy gate electrode layer may be deposited by PVD, CVD, sputtering, or other techniques known and used by those skilled in the art for depositing conductive materials. Other materials, both conductive and non-conductive, may also be used. The top surface of the dummy gate electrode layer typically has a non-planar top surface and may be planarized after it is deposited.
[0021] A hard mask, such as a formed dummy gate mask 117, may be formed over the dummy gate electrode layer to assist in patterning. The dummy gate mask 117 includes one or more masking layers and is used to pattern the dummy gate dielectric layer and the dummy gate electrode layer to form the dummy gate dielectric 112 and the dummy gate electrode 114, as shown in FIGS. Fig. 1A-1C. The dummy gate mask 117 may include one or more patterning layers. In some embodiments, the dummy gate mask 117 may be made of SiO2, SiCN, SiON, Si3N4, Al2O3, SiN x H y or other suitable materials.
[0022] In some embodiments, the dummy gate mask 117 includes a first hard mask layer and a second hard mask layer. The first hard mask layer may be an oxide layer (e.g., silicon oxide), and the second hard mask layer may be a nitride (e.g., silicon nitride). The first hard mask layer and the second hard mask layer may be deposited by a process such as CVD or a spin-on-glass process, although any acceptable process may be used. The first hard mask layer may have a thickness of approximately 1 nm (10 Å) to approximately 25 nm (250 Å), and the second hard mask layer may have a thickness of approximately 15 nm (150 Å) to approximately 85 nm (850 Å). The dummy gate electrode 114 and the dummy gate dielectric 112 together form a dummy gate stack 116.
[0023] Sidewall spacers 118 are formed along sidewalls of the dummy gate stack 116 according to some embodiments. The sidewall spacers 118 may be formed by depositing and patterning a spacer layer (not shown) over the dummy gate stack 116, the fin 104, and the STIs 106. In some embodiments, the spacer layer is formed from silicon nitride and may have a single-layer structure. In alternative embodiments, the spacer layer may have a composite structure including multiple layers. For example, the spacer layer may include a silicon oxide layer and a silicon nitride layer over the silicon oxide layer. Other materials, such as SiO2, SiCN, SiON, Si3N4, SiN x H y , SiOCN, other low-k materials or combinations thereof can also be used.
[0024] The spacer layer may be patterned to form sidewall spacers 118, for example, using an anisotropic etching process to remove the spacer layer over horizontal portions of the device and along sidewalls of the dummy gate stack 116. Due to the difference in the thicknesses of the spacer layer over horizontal portions of the device and along sidewalls of the fin 104, the spacer layer remains along sidewalls of the dummy gate stack 116, while the fins 104 are exposed in the source / drain regions, as shown in the Fig. 1A-1C illustrates.
[0025] It should be noted that the figures show three gate electrodes for illustrative purposes. The gate electrode in the center may represent an active gate electrode, while the gate electrodes on either side are dummy gate electrodes (e.g., non-active ones) used for patterning purposes in some embodiments. In other embodiments, all gate electrodes may be active gate electrodes.
[0026] Fig. 1A-1C further illustrate a first source / drain region 120 and a second source / drain region 122 formed on exposed portions of the fins 104 along opposite sides of the dummy gate stack 116, according to some embodiments. In some embodiments, the fins 104 may be recessed, and the first source / drain region 120 and the second source / drain region 122 are epitaxially formed on the exposed portion of the recessed fin. The use of epitaxially grown materials in the source / drain regions allows the source / drain regions to exert mechanical stress in the channel region 124.The materials used for the first source / drain region 120 and the second source / drain region 122 can be varied for n-type and p-type FinFETs, such that a certain type of material is used for the n-type FinFETs to exert tensile strain in the channel region, and a different type of material is used for the p-type FinFETs to exert compressive stress. For example, SiP or SiC can be used to form n-type FinFETs, and SiGe or Ge can be used to form p-type FinFETs. Other materials can also be used.
[0027] In embodiments where different materials are used for the n-type and p-type devices, it may be desirable to mask one (e.g., the n-type fins) while the epitaxial material is formed on the other (e.g., the p-type fins), and repeat the process for the other. The first source / drain regions 120 and the second source / drain regions 122 may be doped either by an implantation process to implant appropriate dopants or by in-situ doping while the material is grown. For example, for a p-channel MuGFET where the channel is Si or Si 1-x Ge x can be the doped epitaxial film boron-doped Si 1-y Ge ywhere y is at least x, to induce a longitudinal compressive strain in the channel to improve hole mobility. For an n-channel MuGFET, where the channel may be Si, the doped epitaxial film may be, for example, phosphorus-doped silicon (Si:P) or silicon-carbon (Si 1-z C z :P). In the case where the channel is a compound semiconductor, such as In m Ga 1-m As, the doped epitaxial film can be, for example, In n Ga 1-n As, where n is not greater than m.
[0028] We now turn to the Fig. 2A-2C. A liner dielectric film 230 is deposited on the first source / drain region 120, the second source / drain region 122, the sidewall spacers 118, and the dummy gate stack 116. As will be described in more detail below, a fill material is formed over the liner dielectric film 230, which is subsequently patterned. The liner dielectric film 230 acts as an etch stop during patterning of the fill material. In some embodiments, the liner dielectric film 230 comprises SiO2, SiCN, SiON, Si3N4, and SiN x H y, but other suitable dielectric materials may also be used. The liner may further comprise multiple layers comprising combinations of the materials discussed above. The liner dielectric film 230 may be deposited by one or more processes such as PVD, CVD, or ALD, although any acceptable process may be used. Other materials and / or processes may also be used.
[0029] In some embodiments, the liner dielectric film 230 has a thickness t1 of 5 nm or less, such as 3 nm or less. Other thicknesses may also be used.
[0030] Fig. 3A-3C illustrate the formation of a sacrificial inter-gate film 332 over which liner dielectric film 230 is formed, according to some embodiments. In some embodiments, the sacrificial inter-gate film 332 comprises Si, Ge, SiGe, SiO2, SiCN, SiON, SiOx C y , SiO x H y , SiN x H y or other suitable semiconductor or dielectric materials. In some embodiments, the sacrificial inter-gate film 332 is formed by a process such as CVD, FCVD, or a spin-on-glass process, although any acceptable process may be used. Subsequently, portions of the liner dielectric film 230 and the sacrificial inter-gate film 332 that extend above the top of the fins 104 are removed, for example, using an etch process, CMP, or the like.
[0031] As noted above and explained in more detail below, the sacrificial inter-gate film 332 is subsequently removed from between the dummy gate stacks 116, and during the removal process, the underlying liner dielectric film 230 acts as an etch stop, protecting the underlying structures. Therefore, it may be desirable to select materials with high etch selectivity for the liner dielectric film 230 and the sacrificial inter-gate film 332, such that—for a selected etch process—the etch rate for the sacrificial inter-gate film 332 is greater than the etch rate for the liner dielectric film 230.
[0032] We now turn to the Fig. 4A-4C. The sacrificial inter-gate film 332 is recessed, and an inter-gate mask 434 is formed in the recesses according to some embodiments. The inter-gate mask 434 prevents or reduces the loss of sacrificial inter-gate film 332 during removal of the dummy gate electrode 114 and the dummy gate dielectric 112 in subsequent processes. In embodiments where the dummy gate mask 117 comprises silicon nitride and the sacrificial inter-gate film 332 comprises silicon oxide, the sacrificial inter-gate film 332 may be recessed using an isotropic dry or wet etching process, such as wet etching using dilute hydrofluoric acid.
[0033] In some embodiments, the inter-gate mask 434 is made of SiO2, SiCN, SiON, Si3N4, Al2O3, La2O3, SiN x H yor the like by a CVD process, which may be plasma-enhanced CVD (PECVD), molecular layer deposition (MLD), combinations thereof, or other suitable processes. A planarization process, such as a CMP process, may be performed to expose the dummy gate mask 117 (see Fig. 3A-3C). In some embodiments, the planarization process may be continued to remove the dummy gate mask 117 and to expose the dummy gate electrode 114, as shown in the Fig. 4A-4C illustrates.
[0034] The Fig. 5A-5C illustrate the removal of the dummy gate electrode 114 and the dummy gate dielectric 112 according to some embodiments. While the inter-gate mask 434 protects the sacrificial inter-gate film 332, the dummy gate stack 116 is removed. The removal process may include one or more etching processes. For example, in embodiments where the inter-gate mask 434 comprises silicon nitride, the dummy gate electrode 114 comprises polysilicon, and the dummy gate dielectric 112 comprises silicon oxide, the removal process may include selective etching using either dry or wet etching. In the case where dry etching is used, the process gas may include CF4, CHF3, NF3, SF6, Br2, HBr, Cl2, or combinations thereof. Diluent gases such as N2, O2, or Ar can optionally be used. In the case of wet etching, the chemicals may include NH4OH:H2O2:H2O (APM), NH2OH, KOH, HNO3:NH4F:H2O, and / or the like.The dummy gate dielectric 112 can be removed using a wet etching process, for example, using diluted HF acid. Other processes and materials can also be used.
[0035] Fig. 6A-6C illustrate the formation of a gate dielectric layer 642 formed over a channel region 124 of the fin 104, according to some embodiments. In one embodiment, the gate dielectric layer 642 includes one or more high-k dielectric layers (e.g., having a dielectric constant greater than 3.9). For example, the one or more gate dielectric layers may include one or more layers of a metal oxide or silicate of Hf, Al, Zr, combinations thereof, and multilayers thereof. Other suitable materials include La, Mg, Ba, Ti, Pb, and Zr in the form of metal oxides, metal-alloyed oxides, and combinations thereof. Example materials include MgO x , BaTi x O y , BaSr x Ti y O z , PbTi x O y , PbZr x Ti y O z , SiCN, SiON, Si3N4, Al2O3, La2O3, Ta2O3, Y2O3, HfO2, ZrO2, HfSiON, YGex O y , YSi x O y and LaAlO3, and the like. Formation methods of the gate dielectric layer 642 include molecular beam deposition (MBD), ALD, PVD, and the like. In one embodiment, the gate dielectric layer 642 may have a thickness of approximately 0.3 nm (3 Å) to approximately 3 nm (30 Å).
[0036] In some embodiments, an interface layer (not shown) may be formed over the channel region 124 prior to forming the gate dielectric layer 642, and the gate dielectric layer 642 is formed over the interface layer. The interface layer helps buffer the subsequently formed high-k dielectric layer from the underlying semiconductor material. In some embodiments, the interface layer is a chemical silicon oxide, which may be formed from chemical reactions. For example, a chemical oxide may be formed using deionized water + ozone (DIO3), NH4OH + H2O2 + H2O (APM), or other methods. Other embodiments may use a different material or different processes for the interface layer. In one embodiment, the interface layer 640 may have a thickness of approximately 0.3 nm (3 Å) to approximately 0.7 nm (7 Å).
[0037] The Fig. 6A-6C further illustrate a gate electrode 644 formed over the gate dielectric layer 642. The gate electrode 644 may be a metal selected from a group consisting of W, Cu, Ti, Ag, Al, TiAl, TiAlN, TaC, TaCN, TaSiN, Mn, Co, Pd, Ni, Re, Ir, Ru, Pt, and Zr. In some embodiments, the gate electrode 644 comprises a metal selected from a group consisting of TiN, WN, TaN, and Ru. Metal alloys such as Ti-Al, Ru-Ta, Ru-Zr, Pt-Ti, Co-Ni, and Ni-Ta may be used, and / or metal nitrides such as WN x , TiN x , MoN x , TaN x and TaSi x N y, may be used. In some embodiments, the gate electrode 644 has a thickness in the range of approximately 5 nm to approximately 100 nm. The gate electrode 644 may be formed using a suitable process such as ALD, CVD, PVD, plating, or combinations thereof. A planarization process such as CMP may be performed to remove excess materials.
[0038] We now turn to the Fig. 7A-7C, where the recessing of the gate electrode 644 and the formation of a gate electrode mask 746 over remaining portions of the gate electrode 644 are illustrated, according to some embodiments. In some embodiments where the inter-gate mask 434 comprises silicon nitride, the gate electrode may be recessed, for example, using a dry etch process using Cl2 / O2 / BCl3 in a temperature range of 24°C to 150°C and at an operating pressure of less than 133.32 Pa (1 Torr).
[0039] After recessing the gate electrode 644, the gate electrode mask 746 may be formed in the recess to protect the gate electrode 644 during subsequent processes. In some embodiments, the gate electrode mask 746 comprises SiO2, SiCN, SiON, Si3N4, Al2O3, La2O3, SiN x H y , a combination thereof, or the like, but any other suitable dielectric film may also be used. The gate electrode mask 746 may be formed, for example, by CVD, PVD, spin-on deposition, or the like. Other suitable process steps may also be used.
[0040] A planarization process, such as CMP, may be performed to remove excess materials. Furthermore, the planarization process may be continued to remove the inter-gate mask 434, as shown in FIGS. Fig. 7A-7C.
[0041] Fig. 8A-8C illustrate the removal of the sacrificial inter-gate film 332 according to some embodiments. As discussed in more detail below, the sacrificial inter-gate film 332 is replaced with a different metal (e.g., the sacrificial film 950) that has greater etch rate selectivity for the materials of the other structures, such as the silicon oxide, silicon nitride, etc., the gate electrode mask 746, the liner dielectric film 230, and the sidewall spacers 118, allowing for more controlled contact formation.
[0042] In embodiments where the sacrificial inter-gate film 332 is formed from silicon oxide and the gate electrode mask 746 is formed from silicon nitride, the sacrificial inter-gate film 332 may be removed using an isotropic dry or wet etching process. For example, a dry etching process may be used that uses a process gas of C4F6 or fluorine-based gas, or combinations thereof, and a carrier gas, such as N2, O2, or Ar. In the case where wet etching is used, the etchants may include diluted HF and / or the like. As described in the Fig. 8A-8C, after removing the sacrificial inter-gate film 332, the liner dielectric film 230 is exposed.
[0043] Fig. 9A-9C illustrate the formation of a sacrificial film 950 in the recesses formed by removing the sacrificial inter-gate film 332 (see Fig. 8A-8C), according to some embodiments. In some embodiments, the sacrificial film 950 comprises polysilicon, although other materials, such as SiO2, SiCN, SiON, Si3N4, SiN x H y , Si, Ge, SiGe, SiGeC, GeSn or III-V materials, other suitable semiconductors or dielectric films, or the like may be used. In some embodiments, the deposition temperature of the sacrificial film 950 is kept below 600°C, such as below 500°C or below 400°C, to reduce or minimize the thermal impact on the threshold voltage of the previously formed high-k metal gate and dielectric stack. The sacrificial film 950 may also have good gap filling capacity. For example, in one embodiment, a width of the opening ω c less than 20 nm, such as less than 10 nm or, for example, less than 7 nm.
[0044] For example, in some embodiments, the sacrificial film 950 is Si, Ge, SiGe, or SiGeC formed by a CVD process. For example, precursors such as silane (SiH4), disilane (Si2H6), trisilane (Si3H8), tetrasilane (Si4H 10 ) or other higher-order silanes. The CVD process temperature generally depends on the precursor used. For growth using silane or disilane, the growth temperature can be approximately 300°C or more. For growth using higher-order silanes, temperatures less than 300°C can be used. For the other example of Ge growth, the precursor can be germane (GeH4), digermane (Ge2H6), trigermane (Ge3H8), tetragermane (Ge4H 10) or other higher-order germanes. The growth temperature can be approximately 250°C to approximately 650°C. The operating pressure range can be 1333.22 Pa (10 Torr) to 101324.72 Pa (760 Torr).
[0045] As explained in more detail below, portions of the sacrificial film 950 are selectively removed, and therefore, it may be desirable to select a material for the sacrificial film 950 that has a high etch rate selectivity compared to the liner dielectric film 230, the sidewall spacers 118, and the gate electrode mask 746. For example, in embodiments where the sacrificial film 950 is formed from polysilicon and the liner dielectric film 230, the sidewall spacers 118, and the gate electrode mask 746 are formed from silicon nitride or silicon oxide, an etch process may be performed such that the sacrificial film 950 is etched at a much higher rate than the liner dielectric film 230, the sidewall spacers 118, and the gate electrode mask 746.
[0046] The Fig. 10A-10C illustrate the resulting structure after patterning the sacrificial film 950 according to some embodiments. As will be explained below, in subsequent steps, the remaining material of the sacrificial film 950 is replaced with a conductive material to form contacts to the source / drain regions. In some embodiments, photolithography techniques are used to pattern the sacrificial film 950. Generally, a photoresist material (not shown) is deposited over the sacrificial film 950. The photoresist material is irradiated (exposed) with energy and developed such that portions of the photoresist remain over regions of the sacrificial film 950 where contacts to the source / drain regions are desired. The protected portions of the sacrificial film 950 represent the positions of contacts to the source / drain regions. In some embodiments, a length of the contact 1c is less than 40 nm, such as less than 20 nm.As in the . Fig. 10A-10C, portions of the liner dielectric film 230, the sidewall spacers 118, and the gate electrode mask 746 may be removed during the etching process.
[0047] The removal of the sacrificial film 950 can be performed such that the etch rate of the sacrificial film is greater than the etch rate of exposed materials, for example, liner, spacer, gate mask, and insulating dielectric film. The etch selectivities S l , S s , S gm and S id may be the ratios of the etch rate of the sacrificial film to that of the liner dielectric film 230, the sidewall spacers 118, the gate electrode mask 746, or the STI 106. S l , S s , S gm and S idmay each be greater than 3 and in some embodiments may be greater than 10, such as greater than 20. During etching of the sacrificial film 950, the liner dielectric film 230 may be removed along a top surface of the source / drain when the sacrificial film is etched below the top surface of the source / drain. In some embodiments, S ı at least approximately H fin / t l . In some embodiments, H fin 60 nm, and t l is 3 nm, and S l can be at least 20.
[0048] For example, in some embodiments, the sacrificial film 950 is formed from polysilicon, and the liner dielectric film 230, the sidewall spacers 118, and the gate electrode mask 746 are formed from silicon nitride. In such an embodiment, a dry etch contact isolation etch process may be used to avoid lateral overetching that may occur in some wet etching processes. The process gas may include HBr / N2 / O2 or Cl2 / SF6, and the process is performed at a temperature of less than 200°C (e.g., less than 100°C), an RF power of less than 3 kW (e.g., less than 600 W), and a pressure of less than 1333.22 Pa (10 Torr) (e.g., less than 399.97 Pa (3 Torr)).
[0049] As another example, in some embodiments, the sacrificial film 950 is formed from spin-deposited SiOC, and the liner dielectric film 230, the sidewall spacers 118, and the gate electrode mask 746 are formed from silicon nitride. In these embodiments, an anisotropic etch process may be used to remove this sacrificial film 950 in the contact isolation regions. The process gas may include N2 and H2 or SO2 and O2, and the process is performed at a temperature of less than 200°C (e.g., 20-100°C), an RF power of more than 100 W (e.g., more than 300 W), and a pressure of less than 399.97 Pa (3 Torr) (e.g., less than 26.66 Pa (200 mTorr)).
[0050] As another example, in some embodiments, the sacrificial film 950 is formed from silicon oxide formed by flowable CVD, and the liner dielectric film 230, the sidewall spacers 118, and the gate electrode mask 746 are formed from silicon nitride. Removal of the sacrificial film 950 may be performed using a plasma etching process, where the process gas may include C4F6 or a fluorine-based gas, at a temperature of less than 200°C (e.g., less than 150°C), an RF power of more than 50 W (e.g., more than 100 W), and a pressure of less than 399.97 Pa (3 Torr) (e.g., less than 26.66 Pa (200 mTorr)).
[0051] As another example, in some embodiments, sacrificial film 950 is formed from germanium, and liner dielectric film 230, sidewall spacers 118, and gate electrode mask 746 are formed from silicon nitride. In these embodiments, a plasma etch process may be used to remove 950 by anisotropic etching. The process gas may include CF4 chemicals (e.g., CF2Cl2, CF3Br, or the like, or HBr, Cl2, or other halogen gases), and the process may be performed at a temperature of less than 200°C (e.g., less than 60°C), an RF power of greater than 2,000 kW (e.g., from approximately 50 W to approximately 300 W), and a pressure of less than 1333.22 Pa (10 Torr) (e.g., less than 66.66 Pa (500 mTorr)).
[0052] We now turn to the Fig. 11A-11C, where an interlayer dielectric (ILD) 1150 is formed according to one embodiment. The ILD 1150 may comprise a single layer or multiple layers. For example, in some embodiments, an ILD liner 1152 is deposited on the first source / drain region 120, the second source / drain region 122, the sidewall spacers 118, and the dummy gate stack 116, and an ILD fill material 1154 is deposited over the ILD liner 1152. (The ILD liner 1152 and the ILD fill material 1154 are collectively referred to as the ILD 1150.) In some embodiments, the ILD liner 1152 comprises SiO2, SiCN, SiON, Si3N4, and SiN x H y, but other suitable dielectric materials may also be used. The ILD liner 1152 may further comprise multiple layers comprising combinations of the above-mentioned materials. The ILD liner 1152 may be deposited by one or more processes, such as PVD, CVD, or ALD, although any acceptable process may be used. Other materials and / or processes may also be used.
[0053] In some embodiments, the ILD fill material 1154 comprises SiO2, SiCN, SiOC, SiON, Si3N4 and SiN x H y, but other suitable dielectric films may also be used. The insulating dielectric may be cured or treated after deposition. For example, the curing may involve irradiation with ultraviolet radiation, and the treatment may involve annealing in an ambient environment of N2, O2, or H2O at temperatures in the range above 200°C. After curing or treatment, the insulating dielectric may have a relative transmittance of less than 6, such as less than 5, and for example less than 4. For example, the insulating dielectric may be SiO2 formed by a CVD, PECVD, or ALD deposition process, an FCVD, or a spin-on-glass process. A planarization process, such as a CMP process, may be performed to remove excess materials and expose the sacrificial film 950.
[0054] The Fig. 12A-12C illustrate the removal of the sacrificial film 950 (see Fig. 11A-11C) according to some embodiments. As discussed in more detail below, the sacrificial film 950 is replaced by a conductive material that forms the contacts to the source / drain regions. In some embodiments, the sacrificial film 950 may be removed using a selective etch process. Therefore, it may be desirable to select materials for the sacrificial film 950, the ILD 1150, the liner dielectric film 230, the sidewall spacers 118, and the gate electrode mask 746 such that there is a high etch rate selectivity between the material of the sacrificial film 950 and the materials of the ILD 1150, the liner dielectric film 230, the sidewall spacers 118, and the gate electrode mask 746.
[0055] For example, in some embodiments, the sacrificial film 950 is formed from polysilicon, and the ILD liner 1152, the ILD fill material 1154, the liner dielectric film 230, the sidewall spacers 118, and the gate electrode mask 746 are formed from silicon nitride and / or silicon oxide. In these embodiments, a ratio of an etch rate of the polysilicon to an etch rate of the silicon nitride and silicon oxide may be approximately 30 to approximately 50 using an isotropic etch process. In the case where plasma etching is used, the process gas may comprise Cl2 / NF3 / He or SF6 or NF3 or CF4 or another suitable halogen-based etching gas, and the process is carried out at a temperature of less than 200°C (e.g., less than 100°C), an RF power of less than 3 kW (e.g., less than 600 W), and a pressure of less than 1333.22 Pa (10 Torr) (e.g., less than 399.97 Pa (3 Torr)).In some embodiments, the wet etching process is used to remove the sacrificial film 950. The etching material of the wet etching process may be NH4OH or TMAH, or another material may remove the Si film. As another example, a dry chemical etching process may be performed using fluorine-based gas / NH3 at a temperature of less than 200°C (e.g., less than 100°C).
[0056] As another example, in some embodiments, the sacrificial film 950 is formed from SiOC formed by spin-on deposition, the ILD fill material 1154 is SiO2 formed by flowable CVD, the ILD liner 1152 may be an ALD silicon nitride, and the gate electrode mask 746 may be silicon nitride. In these embodiments, a ratio of an etch rate of the SiOC to an etch rate of the silicon nitride and the silicon oxide may be greater than 50 using a plasma etching process, wherein the process gas may include N2 and H2 or SO2 and O2, and the process may be carried out at a temperature of less than 200°C (e.g., 20-100°C), an RF power of more than 100 W (e.g., more than 300 W), and a pressure of less than 399.97 Pa (3 Torr) (e.g., less than 26.66 Pa (200 mTorr)).
[0057] As another example, in some embodiments, the sacrificial film 950 is formed from silicon oxide formed by flowable CVD, the ILD fill material 1154 is SiO2 formed by flowable CVD, the ILD liner 1152 may be silicon nitride, and the gate electrode mask 746 may be silicon nitride. Removal of the sacrificial film 950 may be performed using a plasma etching process. The process gas may include C4F6 or a fluorine-based gas, and the process may be performed at a temperature of less than 200°C (e.g., less than 150°C), an RF power of more than 50 W (e.g., more than 100 W), and a pressure of less than 399.97 Pa (3 Torr) (e.g., less than 26.66 Pa (200 mTorr)).
[0058] As another example, in some embodiments, the sacrificial film 950 is formed of germanium, the ILD fill material 1154 is SiO2, the ILD liner 1152 may be an ALD silicon nitride, and the gate electrode mask 746 may be silicon nitride. In these embodiments, a ratio of an etch rate of the germanium to an etch rate of the silicon nitride and the silicon oxide may be greater than 15 using a plasma etching process, wherein the process gas may include CF4 chemicals (e.g., CF2Cl2, CF3Br, or the like, or HBr, Cl2, or other halogen gases), and the process may be performed at a temperature of less than 200°C (e.g., less than 60°C), an RF power of more than 2,000 kW (e.g., from about 50 W to about 300 W), and a pressure of less than 1333.22 Pa (10 Torr) (e.g., less than 66.67 Pa (500 mTorr)).
[0059] After removing the sacrificial film 950, exposed portions of the liner dielectric film 230 may be removed to expose the underlying first source / drain region 120 and the second source / drain region 122. In embodiments where the liner dielectric film 230 is formed from silicon nitride, the ILD liner 1152 is formed from silicon nitride, and the ILD fill material 1154 is formed from silicon oxide, the liner dielectric film 230 may be removed using a dry etch process with a fluorine-based etch gas. Due to the different materials, the liner dielectric film 230 may be removed without removing or limiting the removal of the ILD liner 1152 and the ILD fill material 1154.
[0060] As illustrated above, by using the sacrificial film 950 and selecting a material that can be selectively etched compared to other materials with high etch rate selectivity, contact openings can be formed using a self-aligned process. Therefore, embodiments such as those disclosed herein allow the formation of smaller contact areas as devices continue to shrink to meet market demands, with less risk of misalignment that can lead to short-circuit and device failure problems. Furthermore, the full space between the sidewall spacers can be utilized, enabling larger contact areas, which can reduce contact resistance and increase device performance.
[0061] We now turn to the Fig. 13A-13D, where the formation of contacts 1358 is shown according to some embodiments, wherein Fig. 13D an enlarged view of a section of Fig. 13B illustrates. The contacts 1358 may comprise a single-layer or a multi-layer structure. For example, in some embodiments, the contacts 1358 include a liner, such as a diffusion barrier layer, an adhesion layer, or the like, and a contact filler formed over the contact liner in the openings. The contact liner may include Ti, TiN, Ta, TaN, or the like formed by ALD, CVD, or the like. The contact filler may be formed by depositing a conductive material, such as one or more layers of Ni, Ta, TaN, W, Co, Ti, TiN, Al, Cu, Au, alloys thereof, combinations thereof, or the like; but other suitable materials may also be used. A planarization process, such as CMP, may be performed to remove excess material from a surface of the ILD fill material 1154.
[0062] As in Fig. 13D, the contacts 1358 may have a tapered contact sidewall profile, and the angle between the upper contact plane and the contact sidewall is greater than 90°. As shown in Fig. 13D, a width of the contacts 1358 along a top of the contacts 1358C is w1 less than a width of the contacts 1358 along a bottom of the contacts 1358 C w2 .
[0063] The Fig. 14A-15B illustrate various etch profiles that may be obtained in a contact isolation region during the removal of the sacrificial film 950, according to some embodiments. The process described in FIGS. Fig. 14A-15B, assumes that the processes described above with respect to the Fig. 1A-10C are carried out beforehand, and that the processes described above with respect to the Fig. 11A-13, are carried out according to the processes described below with reference to the Fig. 14A-15B. Accordingly, after performing the processes described above with reference to the Fig. 1A-10C, an additional etch process may be performed to laterally widen an upper portion of the opening above the contact isolation region, while protecting portions of the sacrificial film 950 over the source / drain regions, as discussed above with reference to the Fig. 10A-10C. As discussed in the Fig. As illustrated in Figures 14A-14B, the second etch may laterally remove upper portions of the sidewall spacers 118 and portions of the remaining gate electrode mask 746, forming a funnel-shaped opening profile. The funnel-shaped opening profile allows a portion of the sidewall spacers 118 and the gate electrode mask 746 to be replaced with a lower dielectric constant material, thereby reducing stray capacitance between the gate electrode 644 and the contact 1358 and increasing device performance.
[0064] In some embodiments where the sidewall spacers 118 and the gate electrode mask 746 are formed from silicon nitride, a lateral dry etch plasma process may be performed using a process gas of CF4 chemicals (e.g., CF2Cl2, CF3Br, or the like) at a temperature of less than 250°C (e.g., less than 150°C), an RF power of more than 3 kW (e.g., less than 2 kW), and a pressure of less than 666.61 Pa (5 Torr) (e.g., less than 333.31 Pa (2.5 Torr)). In some embodiments, the height (H A ) and the width (W A ) larger than 4 nm or larger than 2 nm.
[0065] After that, processes such as those described above with reference to the Fig. 11A-11C, to fill the funnel-shaped opening with the ILD material (for example, the ILD liner 1152 and the ILD filler material 1154), thereby forming the structure shown in the Fig. 15A-15B. Processes such as those described above with reference to the Fig. 11A-13D can be performed to complete the formation of the contacts.
[0066] The Fig. 16A-17B illustrate various etch profiles that may be obtained in the contact regions during the removal of the sacrificial film 950 in the contact regions, according to some embodiments. As discussed above, the Fig. 14A-15B show various profiles that can be obtained for the openings in which the ILD 1150 is formed. Fig. 16A-17B illustrate that similar etch profiles are used to form the openings in which the contacts 1358 are formed.
[0067] The processes that take place in the Fig. 16A-17B, assume that the processes described above with respect to the Fig. 1A-1C, to the process of removing the sacrificial film 950, as discussed in relation to the Fig. 12A-12C, before the processes discussed in the Fig. 16A-17B, wherein an additional etch process may be performed to laterally widen an upper portion of the opening above the contact region of the source / drain regions. As shown in the Fig. 16A-16B, the second etch may laterally remove upper portions of sidewall spacers 118 and portions of the gate electrode mask 746, thereby forming a funnel-shaped opening profile, as shown in FIGS. Fig. 16A-16B. The ILD liner 1152 may be removed from above the source / drain regions as described above with reference to the Fig. 12-12C was discussed.
[0068] In some embodiments where the ILD liner 1152 and the gate electrode mask 746 are formed from silicon nitride, a lateral dry etch plasma process may be used using a process gas of CF4 chemicals (e.g., CF2Cl2, CF3Br, or the like) at a temperature of less than 250°C (e.g., less than 150°C), an RF power of more than 3 kW (e.g., less than 2 kW), and a pressure of less than 666.61 Pa (5 Torr) (e.g., less than 333.31 Pa (2.5 Torr)). In some embodiments, the height (H B ) and the width (W B ) less than 7 nm or less than 3 nm, respectively. In some embodiments, H B and W B greater than H A or W AThe funnel-shaped opening can then be filled with a conductive material, as described above with reference to the Fig. 13A-13D, which leads to a contact that is in the Fig. 17A-17B.
[0069] Other processes may also be performed. For example, in some embodiments, a metal silicide may be formed over the first source / drain regions 930 and the second source / drain regions 932. In some embodiments, a metal silicide formation process is performed after the inter-gate film removal process described above with reference to the Fig. 8A-8C, or after the contact hole formation discussed above with reference to the Fig. 12A-12C. The metal silicide formation process may form a metal silicide on the doped source / drain region to reduce the R cbetween the doped source / drain regions and the subsequently formed contact metal. The metal silicide formation process includes a metal film deposition on the source / drain region, a heat treatment to form a metal silicide at the interface between the source / drain regions, and an etching process to remove the excess unreacted metal. The metal silicide comprises TiSi x , NiSi x , CoSi x , NiCoSi x and TaSi x , but other suitable silicide materials may also be used. In some embodiments, silicide formation may be performed by ILD removal.
[0070] It should be noted that other embodiments may use various other steps or sequences of steps. For example, Fig. 18-28 show various cross-sectional and perspective views of intermediate stages in the fabrication of a FinFET according to some embodiments. Fig. 18A-28C illustrate the “A” figures (for example Fig. 18A, Fig. 19A etc.) a perspective view, the “B” figures (for example Fig. 18B, Fig. 19B etc.) illustrate a cross-sectional view along the Y-cut line illustrated in the respective “A” figure, and the “C” figures (for example Fig. 18C, Fig. 19C, etc.) illustrate a cross-sectional view along the X-cut line illustrated in the respective “A” figure.
[0071] As above with reference to the Fig. 1-15, a sacrificial inter-gate film 332 was temporarily deposited, as described in the Fig. 3A-3C, and then removed as shown in the Fig. 8A-8C, and replaced by a sacrificial film 950, as shown in the Fig. 9A-9C. In the embodiments shown in the Fig. 18-28, the sacrificial intermediate gate 332 has been omitted. Instead, as will be discussed in more detail below, the sacrificial film 950 is formed earlier in the process. The process is described in the following paragraphs with reference to the Fig. 18-28, where like reference numerals indicate like elements.
[0072] We first turn to the Fig. 18A-19C where processes and materials similar to those described above with reference to the Fig. 1A-2C, and where like reference numerals indicate like elements, repetition is omitted.
[0073] We now turn to the Fig. 20A-20C, where the sacrificial film 950 is formed, for example, using materials and processes similar to those described above with reference to the Fig. 9A-9C were discussed.
[0074] The Fig. 21A-21C illustrate recessing of the sacrificial film 950 according to some embodiments. As discussed in more detail below, the sacrificial film 950 is recessed, and a mask is formed over the recessed sacrificial film 950 to protect the sacrificial film 950 during subsequent processing. In some embodiments, the sacrificial film 950 is recessed from 10 nm (100 Å) to approximately 60 nm (600 Å) to allow sufficient thickness for the subsequently formed mask layer.
[0075] In embodiments where the sacrificial film 950 is formed from polysilicon and the dummy gate mask 116, the liner dielectric film 230, and the sidewall spacer 118 are formed from silicon nitride / silicon oxide, the sacrificial film 950 may be recessed using a time-controlled etch, for example, using a plasma etch with a process gas of NF3, HBr / N2 / O2, or Cl2 / NF3 / He, at a temperature of less than 200°C (e.g., less than 100°C), an RF power of less than 3 kW (e.g., less than 600 W), and a pressure of less than 1333.22 Pa (10 Torr) (e.g., less than 399.97 Pa (3 Torr)).
[0076] As another example, in some embodiments, the sacrificial film 950 is formed from SiOC formed by spin-on deposition, and the dummy gate mask 116, the liner dielectric film 230, and the sidewall spacers 118 are formed from silicon nitride / silicon oxide, and the sacrificial film 950 may be recessed using a time-controlled etch, for example, using plasma etching with a process gas containing N2 and H2 or SO2 and O2, at a temperature of less than 200°C (e.g., 20-100°C), an RF power of more than 100 W (e.g., more than 300 W), and a pressure of less than 399.97 Pa (3 Torr) (e.g., less than 26.66 Pa (200 mTorr)).
[0077] As another example, in some embodiments, the sacrificial film 950 is formed from silicon oxide formed by flowable CVD, and the dummy gate mask 116, the liner dielectric film 230, and the sidewall spacers 118 are formed from silicon nitride, and the sacrificial film 950 may be recessed using a time-triggered etch, for example, using a plasma etch process with a process gas containing C4F6 or a fluorine-based gas, and at a temperature of less than 200°C (e.g., less than 150°C), an RF power of more than 50 W (e.g., more than 100 W), and a pressure of less than 399.97 Pa (3 Torr) (e.g., less than 26.66 Pa (200 mTorr)).
[0078] As another example, in some embodiments, the sacrificial film 950 is formed from germanium, and the dummy gate mask 116, the liner dielectric film 230, and the sidewall spacers 118 are formed from silicon nitride / silicon oxide, and the sacrificial film 950 may be recessed using a time-triggered etch, for example, using a plasma etch process with a process gas containing CF4 chemicals (e.g., CF2Cl2, CF3Br, or the like, or HBr, Cl2, or other halogen gases), and at a temperature of less than 200°C (e.g., less than 60°C), an RF power of more than 2,000 kW (e.g., from about 50 W to about 300 W), and a pressure of less than 1333.22 Pa (10 Torr) (e.g., less than 66.66 Pa (500 mTorr)).
[0079] A planarization process, such as a CMP process, may be performed to expose the dummy gate mask 117 (see Fig. 19A-19C). In some embodiments, the planarization process may be continued to remove the dummy gate mask 117 and to expose the dummy gate electrode 114, as shown in FIGS. Fig. 21A-21C.
[0080] Thereafter, processing similar to that described above with reference to the Fig. 5A-7C, as described in the respective Fig. 22A-24C. Similar processes and materials may be used as those described above, so repetition is omitted.
[0081] We now turn to the Fig. 25A-25C, where the sacrificial film 950 is patterned according to some embodiments. The sacrificial film 950 may be patterned as described above with reference to Fig. 10A-10C, and therefore a repetition is omitted. It should be noted that in embodiments such as these, the processes described above with reference to Fig. 3 and Fig. 8 (forming and then removing the sacrificial inter-gate film 332 to form the sacrificial film 950) may be omitted.
[0082] Afterwards, the Fig. 26A-28D. The processes may be similar to those described above with reference to the Fig. 11A-13D were discussed, which is why a repetition is not carried out.
[0083] Processes described above with reference to the Fig. 14A-15B to form a funnel-shaped opening in the contact isolation region, and / or which were discussed with reference to the Fig. 16A-17B to form a funnel-shaped opening in the contact region can be integrated into the process described above with reference to the Fig. 18A-28C. For example, after performing the processes described above with reference to Fig. 18A-25C, the processes described above with reference to the Fig. 14A-15B. Accordingly, after performing the processes discussed above with reference to the Fig. 18A-25C, an additional etch process may be performed to laterally widen an upper portion of the opening above the contact isolation region, while protecting portions of the sacrificial film 950 over the source / drain regions, as discussed above with reference to the Fig. 25A-25C. As discussed in the Fig. As illustrated in Figures 14A-14B, the second etch may laterally remove upper portions of the sidewall spacers 118 and portions of the remaining gate electrode mask 746, forming a funnel-shaped opening profile. The funnel-shaped opening profile allows a portion of the sidewall spacers 118 and the gate electrode mask 746 to be replaced with a material having a lower dielectric constant, thereby reducing the stray capacitance between the gate electrode 644 and the contact 1358 and improving device performance.
[0084] In some embodiments where the sidewall spacers 118 and the gate electrode mask 746 are formed from silicon nitride, a lateral dry etch plasma process may be performed using a process gas of CF4 chemicals (e.g., CF2Cl2, CF3Br, or the like) at a temperature of less than 250°C (e.g., less than 150°C), an RF power of more than 3 kW (e.g., less than 2 kW), and a pressure of less than 666.61 Pa (5 Torr) (e.g., less than 333.31 Pa (2.5 Torr)). In some embodiments, the height (H A ) and the width (W A ) larger than 4 nm or larger than 2 nm.
[0085] After that, processes such as those described above with reference to the Fig. 26A-26C, to fill the funnel-shaped opening with the ILD material (for example, the ILD liner 1152 and the ILD filler material 1154), thereby obtaining the structure described in the Fig. 15A-15B. Processes such as those described above with reference to the Fig. 26A-28D can be performed to complete formation of the contacts.
[0086] The processes described above with reference to the Fig. 16A-17B can also be integrated into the process described above with reference to Fig. 18A-28C to form a funnel-shaped contact. For example, after performing the processes described above with reference to the Fig. 18A-27C, but prior to removing the liner dielectric film 230, an additional etch process may be performed to laterally widen an upper portion of the opening above the contact region of the source / drain regions. As shown in Fig. 16A-16B, the second etch may laterally remove upper portions of sidewall spacers 118 and portions of the gate electrode mask 746, forming a funnel-shaped opening profile as shown in FIGS. Fig. 16A-16B. The ILD liner 1152 may be removed over the source / drain regions as described above with reference to Fig. 27A-27C was discussed.
[0087] In some embodiments where the ILD liner 1152 and the gate electrode mask 746 are formed from silicon nitride, a lateral dry etch plasma process may be used using a process gas of CF4 chemicals (e.g., CF2Cl2, CF3Br, or the like) at a temperature of less than 250°C (e.g., less than 150°C), an RF power of more than 3 kW (e.g., less than 2 kW), and a pressure of less than 666.61 Pa (5 Torr) (e.g., less than 333.31 Pa (2.5 Torr)). In some embodiments, the height (H B ) and the width (W B ) less than 7 nm or less than 3 nm, respectively. In some embodiments, H B and W B greater than H A or W A The funnel-shaped opening can then be filled with a conductive material, as described above with reference to the Fig. 28A-28D, resulting in contact as described in the Fig. 17A-17B.
[0088] Other processes may also be performed. For example, in some embodiments, a metal silicide may be formed over the first source / drain regions 930 and the second source / drain regions 932. In some embodiments, a metal silicide formation process is performed after the inter-gate film removal process described above with reference to the Fig. 8A-8C, or after the contact hole formation discussed above with reference to the Fig. 12A-12C. The metal silicide formation process may form a metal silicide on the doped source / drain region to form R cbetween the doped source / drain regions and the subsequently formed contact metal. The metal silicide formation process includes a metal film deposition on the source / drain region, a heat treatment to form a metal silicide at the interface between the source / drain regions, and an etching process to remove the excess unreacted metal. The metal silicide comprises TiSi x , NiSi x , CoSi x , NiCoSi x and TaSi x , but other suitable silicide materials may also be used. In some embodiments, silicide formation may be performed by removing the ILD.
Claims
[1] A method of forming a semiconductor device, the method comprising: Forming a sacrificial film (950) over a source / drain region (120) of a fin (104) and an insulating region (106) adjacent to the fin (104); Removing a first portion of the sacrificial film (950) over the isolation region (106) to form a first recess, leaving a second portion of the sacrificial film (950) over the source / drain region (120); Forming a dielectric layer (1150) in the first recess; Removing the second portion of the sacrificial film (950) to form a second recess; and Forming a conductive layer (1358) in the second recess, wherein the sacrificial film (950) is formed between a first gate electrode and a second gate electrode, wherein a first gate mask is arranged over the first gate electrode and a second gate mask is arranged over the second gate electrode, wherein the conductive layer (1358) has a tapered profile along an axis parallel to the first gate electrode and the second gate electrode such that an outermost width of the conductive layer (1358) decreases as the conductive layer (1358) extends away from the source / drain region (120). [2] The method of claim 1, wherein the dielectric layer (1150) extends over a top surface of the first gate mask and the second gate mask. [3] The method of claim 1 or 2, wherein a first sidewall spacer (118) extends along a sidewall of the first gate electrode and the first gate mask, and a second sidewall spacer (118) extends along a sidewall of the second gate electrode and the second gate mask. [4] The method of claim 3, further comprising, after removing the first portion of the sacrificial film (950), removing an upper portion of the first gate mask and the second gate mask, wherein, after forming the dielectric layer (1150) in the first recess, the dielectric layer (1150) contacts the first gate mask. [5] The method of claim 4, wherein removing the upper portion of the first gate mask recesses the first gate mask by a distance of more than 4 nm. [6] The method of any one of claims 3 to 5, further comprising, after removing the second portion of the sacrificial film (950), removing an upper portion of the first gate mask and the second gate mask, wherein, after forming the conductive layer (1358), the conductive layer (1358) contacts the first gate mask. [7] The method of claim 6, wherein removing the upper portion of the first gate mask recesses the first gate mask by a distance of more than 7 nm. [8] A method of forming a semiconductor device, the method comprising: Forming a first gate electrode and a second gate electrode over a semiconductor structure; Forming a first sacrificial film (950) between the first gate electrode and the second gate electrode; Structuring the first sacrificial film (950) such that a remaining portion of the first sacrificial film (950) remains above the semiconductor structure, and such that first recesses are formed between the first gate electrode and the second gate electrode on opposite sides of the remaining portion of the first sacrificial film (950); Forming an interlayer dielectric, ILD, (1150) in the first recesses; Removing the remaining portion of the first sacrificial film (950) to form a second recess; and Forming a conductive structural element (1358) in the second recess, wherein the conductive structure element (1358) has a tapered profile along an axis parallel to the first gate electrode and second gate electrode, such that an outermost width of the conductive structure element (1358) decreases as the conductive structure element (1358) extends away from the semiconductor structure. [9] The method of claim 8, wherein the first sacrificial film (950) is silicon. [10] A method according to claim 8 or 9, further comprising: before forming the first gate electrode and the second gate electrode, forming a first dummy gate electrode and a second dummy gate electrode; before forming the first sacrificial film (950), forming a second sacrificial film between a first dummy gate electrode and a second dummy gate electrode; before forming the first sacrificial film (59), replacing the first dummy gate electrode and the second dummy gate electrode with the first gate electrode and the second gate electrode; and after replacing the first dummy gate electrode and the second dummy gate electrode, removing the second sacrificial film. [11] The method of any one of claims 8 to 10, wherein the semiconductor structure comprises a semiconductor fin (104), a plurality of semiconductor fins, a semiconductor nanowire, or a plurality of semiconductor nanowires. [12] The method of any one of claims 8 to 11, wherein the conductive structural element (1358) has a tapered profile such that an angle between the top surface and a sidewall surface is greater than approximately 90°. [13] The method of any one of claims 8 to 12, further comprising forming a mask layer over the first gate electrode and the second gate electrode, wherein removing the remaining portion of the first sacrificial film (950) is performed by an etching process having a ratio of an etch rate of a material of the first sacrificial film (950) to an etch rate of a material of the mask layer of greater than 15. [14] The method of any one of claims 8 to 13, further comprising forming a mask layer over the first gate electrode and the second gate electrode, wherein removing the remaining portion of the first sacrificial film (950) is performed by an etching process having a ratio of an etch rate of a material of the first sacrificial film (950) to an etch rate of a material of the mask layer of greater than 30. [15] The method according to any one of claims 8 to 14, wherein a width of the second recess is less than 20 nm. [16] The method of any one of claims 8 to 15, wherein a width of the conductive structure element (1358) extending along an axis between the first gate electrode and the second gate electrode is wider along a top side than along a bottom side. [17] A semiconductor device comprising: a gate electrode (644) over a semiconductor structure, the semiconductor structure having a first source / drain region (120), a second source / drain region (122), and a channel region (124) disposed between the first source / drain region (120) and the second source / drain region (122), the gate electrode (644) being disposed over the channel region (124); a gate mask (746) over the gate electrode (644); a sidewall spacer (118) along a sidewall of the gate electrode (644) and the gate mask (746), wherein a top surface of the sidewall spacer (118) is recessed from a top surface of the gate mask (746); and a contact (1358) adjacent to the sidewall spacer (118), the contact (1358) being electrically coupled to the first source / drain region (120), the contact (1358) contacting a sidewall of a top portion of the gate mask (746), the contact (1358) having a tapered profile along an axis parallel to the gate electrode (644) such that an outermost width of the contact (1358) decreases as the contact (1358) extends away from the first source / drain region (120). [18] The semiconductor device of claim 17, further comprising an ILD layer (1150) adjacent to the contact, wherein the ILD layer (1150) extends over at least a portion of the gate mask (746).
Citation Information
Patent Citations
Methods of forming finfet semiconductor devices with self-aligned contact elements using a replacement gate process and the resulting devices
US20150069532A1
Method for fabricating a semiconductor device
US20150145027A1
Semiconductor devices and methods of manufacturing the same
US20150333148A1
Forming wrap-around silicide contact on finFET
US9318581B1
