Fork seat with reduced coupling effect
The air gap isolated fork-sheet FET design addresses capacitive coupling issues by using an air gap and dielectric pillar to enhance transistor performance and stability.
Patent Information
- Application Number
- JP2024523425
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2021-12-03
- Filing Date
- 2022-11-04
- Publication Date
- 2024-11-15
AI Technical Summary
Fork-sheet FET devices face challenges due to significant capacitive coupling between NFET and PFET devices, leading to transistor malfunctions and stability issues.
Incorporation of an air gap isolated fork-sheet FET design with a dielectric pillar between NFET and PFET devices, utilizing an air gap to reduce capacitive coupling by employing materials with lower dielectric constants.
Significantly reduces capacitive coupling, enhancing transistor performance and stability by minimizing capacitive interference between NFET and PFET devices.
Smart Images

Figure 2024542366000001_ABST
Abstract
Description
[Technical field]
[0001] The present invention relates to fork-seat field effect transistor (FET) devices, and more particularly to air-gap isolated fork-seat FET devices for reduced capacitance coupling. [Background technology]
[0002] Fork-sheet field-effect transistor (FET) devices offer further scaling potential over traditional finFET and nanosheet architectures. In fork-sheet devices, the spacing between n-channel FET (NFET) and p-channel FET (PFET) devices is reduced, enabling further area scaling.
[0003] However, implementing a fork-sheet FET design can present some significant challenges: For example, in a fork-sheet FET design, the NFET source / drain regions can be very close to the PFET source / drain regions, and the NFET gate can be very close to the PFET gate.
[0004] Conventional approaches employ minimal insulation between the source / drain regions and gates of the NFET and PFET, so that capacitance coupling between the source / drain regions and / or gates of the NFET and PFET can cause transistor malfunction and / or affect read / write stability.
[0005] Therefore, an improved fork-sheet FET device design with reduced capacitance coupling would be desirable. Summary of the Invention
[0006] The present invention provides an air-gap isolated fork-sheet field effect transistor (FET) device for reducing capacitance coupling. In one aspect of the present invention, a fork-sheet FET device is provided. The fork-sheet FET device includes at least a first nanosheet FET (e.g., an n-channel FET (NFET)) and a second nanosheet FET (e.g., a p-channel FET (PFET)), and a dielectric pillar disposed directly between the first nanosheet FET and the second nanosheet FET, the dielectric pillar including an air gap. For example, the first nanosheet FET and the second nanosheet FET are in contact with opposite sides of the dielectric pillar.
[0007] In another aspect of the present invention, there is provided another fork-sheet FET device comprising at least a first nanosheet FET and a second nanosheet FET and a dielectric pillar disposed directly between the first nanosheet FET and the second nanosheet FET, the dielectric pillar comprising an air gap, the first nanosheet FET and the second nanosheet FET having nanosheets extending horizontally on either side of the dielectric pillar.
[0008] In yet another aspect of the invention, a method of forming a fork-sheet FET device is provided, comprising forming at least a first nanosheet FET and a second nanosheet FET with a sacrificial pillar disposed directly therebetween, and removing the sacrificial pillar to form an opening between the first nanosheet FET and the second nanosheet FET, a lower portion of the opening having a width W1. OPENING and the upper portion of the opening has a width W2 OPENING and W2 OPENING <W1 OPENING and depositing a dielectric liner in the opening to line the opening, the W2 OPENING <W1 OPENINGand depositing and lining the dielectric liner to close an upper portion of the opening before a lower portion of the opening is completely filled based on the above, thereby creating a dielectric pillar having an air gap between the first nanosheet FET and the second nanosheet FET.
[0009] A more complete understanding of the present invention, as well as further features and advantages of the present invention, will be obtained by reference to the following detailed description and drawings. [Brief description of the drawings]
[0010] [Figure 1] 1A-1C are top view diagrams of the overall layout of a fork-sheet field effect transistor (FET) device structure and various cut orientations depicted in the figures, according to one embodiment of the present invention. [Figure 2A] FIG. 2 is a cross-sectional view taken along line XX′ showing a stack of alternating sacrificial and active layers formed on a substrate, the stack being patterned, according to one embodiment of the present invention. [Figure 2B] 2A-2C are cross-sectional views taken along lines Y1-Y1' and Y2-Y2' illustrating a stack of alternating sacrificial and active layers formed on a substrate, the stack being patterned, according to one embodiment of the present invention. [Figure 3A] FIG. XX′ cross-sectional view showing an n-channel FET (NFET)-p-channel FET (PFET) space open in the stack dividing the stack into at least one stack 204N of NFET devices and at least one stack of PFET devices, according to one embodiment of the present invention. [Figure 3B] 1A-1C are cross-sectional views taken along the lines Y1-Y1' and Y2-Y2', showing an NFET-PFET space being opened in the stack, dividing the stack into at least one stack of NFET devices and at least one stack of PFET devices, according to one embodiment of the present invention. [Figure 4]1A-1C are cross-sectional views along Y1-Y1' and Y2-Y2' illustrating a sacrificial material being deposited in and filling the NFET-PFET space to form a sacrificial pillar separating a stack of NFET devices from a stack of PFET devices, and forming a shallow trench isolation (STI) region at the base of the stack of NFET / PFET devices, according to one embodiment of the present invention. [Figure 5A] FIG. XX′ cross-sectional view showing a sacrificial gate formed on a stack of NFET devices and a stack of PFET devices, and a first sacrificial layer removed to form a cavity under the stack of NFET devices and the stack of PFET devices, according to one embodiment of the present invention. [Figure 5B] FIG. 1 is a Y1-Y1′ cross-sectional view illustrating a sacrificial gate being formed on a stack of NFET devices and a stack of PFET devices, and a first sacrificial layer being removed to form a cavity below the stack of NFET devices and the stack of PFET devices, according to an embodiment of the present invention. [Figure 5C] FIG. 1 is a cross-sectional view taken along the line Y2-Y2′ illustrating a sacrificial gate being formed on a stack of NFET devices and a stack of PFET devices, and a first sacrificial layer being removed to form a cavity below the stack of NFET devices and the stack of PFET devices, according to an embodiment of the present invention. [Figure 6A] FIG. XX' cross-sectional view showing a dielectric spacer material being deposited over the sacrificial gate and also deposited into and filling the cavity, and the dielectric spacer material being patterned into a bottom dielectric isolation layer in the cavity and into a gate spacer aligned with the sacrificial gate, in accordance with one embodiment of the present invention. [Figure 6B] FIG. 1 is a Y1-Y1′ cross-sectional view showing a dielectric spacer material being deposited over the sacrificial gate and also deposited into and filling the cavity, and the dielectric spacer material being patterned into a bottom dielectric isolation layer in the cavity and into a gate spacer aligned with the sacrificial gate, in accordance with one embodiment of the present invention. [Figure 6C]FIG. 1 is a Y2-Y2′ cross-sectional view showing a dielectric spacer material being deposited over the sacrificial gate and also deposited into and filling the cavity, and the dielectric spacer material being patterned into a bottom dielectric isolation layer in the cavity and into a gate spacer aligned with the sacrificial gate, in accordance with one embodiment of the present invention. [Figure 7A] FIG. XX′ cross-sectional view showing a sacrificial gate and a gate spacer being used as a mask to pattern a trench in a stack of NFET devices between the sacrificial gate and a stack of PFET devices according to one embodiment of the present invention. [Figure 7B] FIG. 13 is a Y1-Y1′ cross-sectional view illustrating a sacrificial gate and a gate spacer being used as a mask to pattern a trench in a stack of NFET devices between the sacrificial gate and a stack of PFET devices, according to one embodiment of the present invention. [Figure 8A] FIG. XX′ cross-sectional view showing an embodiment of the present invention in which an internal spacer is formed alongside a second sacrificial layer in a stack of NFET devices and a stack of PFET devices, and source / drain regions of the NFET and PFET are formed in trenches along the second sacrificial layer and active layers in the stack of NFET devices and the stack of PFET devices on opposite sides of the sacrificial gate. [Figure 8B] FIG. 1 is a Y1-Y1′ cross-sectional view illustrating an embodiment of the present invention in which an internal spacer is formed alongside a second sacrificial layer in a stack of NFET devices and a stack of PFET devices, and source / drain regions of the NFET and PFET are formed in trenches along the second sacrificial layer and active layers in the stack of NFET devices and the stack of PFET devices on opposite sides of the sacrificial gate. [Figure 9] FIG. 13 is a cross-sectional view taken along the line Y1-Y1′ illustrating the upper portion of the sacrificial pillar exposed above the source / drain regions of the NFET / PFET being trimmed according to one embodiment of the present invention. [Figure 10A]FIG. XX′ cross-sectional view showing a stack of NFET / PFET devices (and the sacrificial pillars between them), a sacrificial gate and gate spacers, and an interlayer dielectric (ILD) deposited over and filling the source / drain regions of the NFET / PFET, in accordance with one embodiment of the present invention. [Figure 10B] FIG. 1 is a Y1-Y1′ cross-sectional view showing a stack of NFET / PFET devices (and the sacrificial pillars between them), a sacrificial gate and gate spacers, and an ILD deposited over and filling the source / drain regions of the NFET / PFET, in accordance with one embodiment of the present invention. [Figure 10C] FIG. 1 is a Y2-Y2′ cross-sectional view showing a stack of NFET / PFET devices (and the sacrificial pillars between them), a sacrificial gate and gate spacers, and an ILD deposited over and filling the source / drain regions of the NFET / PFET, in accordance with one embodiment of the present invention. [Figure 11A] FIG. XX′ cross-sectional view showing the implementation of a partial recession of the sacrificial gate to expose the upper portion of the sacrificial pillar in the gate region of the device according to one embodiment of the present invention. [Figure 11B] A cross-sectional view taken along the Y2-Y2' line showing the implementation of a partial recession of the sacrificial gate to expose the upper portion of the sacrificial pillar within the gate region of the device according to one embodiment of the present invention. [Figure 12] FIG. 2 is a cross-sectional view taken along the line Y2-Y2′, showing the upper portion of the sacrificial pillar exposed above the trimmed NFET device stack and the PFET device stack, according to one embodiment of the present invention. [Figure 13A] FIG. XX′ cross-sectional view illustrating the selective removal of remaining portions of the sacrificial gate to form a gate trench in the ILD overlying the NFET device stack and the PFET device stack between the NFET / PFET source / drain regions, and the removal of the second sacrificial layer from the NFET device stack and the PFET device stack to form a gap between the active layers, in accordance with one embodiment of the present invention. [Figure 13B]FIG. 13 is a cross-sectional view taken along line Y2-Y2′ illustrating the selective removal of remaining portions of the sacrificial gate to form a gate trench in the ILD overlying the NFET device stack and the PFET device stack between the NFET / PFET source / drain regions, and the removal of the second sacrificial layer from the NFET device stack and the PFET device stack to form a gap between the active layers, in accordance with one embodiment of the present invention. [Figure 14A] FIG. XX′ cross-sectional view showing a replacement gate formed in the gate trench and gap according to one embodiment of the present invention. [Figure 14B] FIG. 2 is a Y2-Y2′ cross-sectional view showing a replacement gate formed in the gate trench and gap according to one embodiment of the present invention. [Figure 15A] FIG. 2 is a cross-sectional view taken along line XX' illustrating the replacement gate having been recessed and a dielectric cap formed over the (recessed) replacement gate in accordance with one embodiment of the present invention. [Figure 15B] FIG. 2 is a cross-sectional view taken along line Y1-Y1′, illustrating the replacement gate being recessed and a dielectric cap being formed over the (recessed) replacement gate, in accordance with one embodiment of the present invention. [Figure 15C] FIG. 2 is a cross-sectional view taken along line Y2-Y2′, illustrating the replacement gate being recessed and a dielectric cap being formed over the (recessed) replacement gate, in accordance with one embodiment of the present invention. [Figure 16A] FIG. 11 is a Y1-Y1′ cross-sectional view showing the sacrificial pillars being removed to form openings between the source / drain regions of the NFET and PFET in the source / drain region of the fork sheet FET device and between replacement gates on top of a stack of NFET and PFET devices in the gate region of the fork sheet FET device, according to one embodiment of the present invention. [Figure 16B]FIG. 11 is a cross-sectional view taken along the line Y2-Y2′ showing the sacrificial pillars being removed to form openings between the source / drain regions of the NFET and PFET in the source / drain region of the fork sheet FET device and between replacement gates on top of a stack of NFET and PFET devices in the gate region of the fork sheet FET device, in accordance with one embodiment of the present invention. [Figure 17A] FIG. 11 is a cross-sectional view taken along line Y1-Y1′, illustrating a conformal dielectric liner being deposited within and lining the opening to form a dielectric air gap pillar, in accordance with one embodiment of the present invention. [Figure 17B] FIG. 13 is a cross-sectional view taken along line Y2-Y2', illustrating a conformal dielectric liner being deposited within and lining the opening to form a dielectric air gap pillar, in accordance with one embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0011] As mentioned above, fork-sheet field-effect transistor (FET) designs can present some significant challenges: placing the NFET source / drain regions too close to the PFET source / drain regions and the NFET gate too close to the PFET gate increases the risk of capacitance coupling, which can cause transistor malfunctions and / or affect read / write stability.
[0012] Advantageously, provided herein is an air-gap isolated fork sheet FET device in which capacitance coupling is significantly reduced as compared to conventional designs, where dielectric materials such as silicon oxide (SiOx) and / or silicon nitride (SiN) are used to separate the NFET and PFET devices in a fork sheet FET. Nevertheless, capacitance coupling still has a significant impact on device performance. On the other hand, air has a significantly lower dielectric constant than these conventional oxide and nitride dielectric materials. For example, and by way of example only, at room temperature (i.e., 25 degrees Celsius (°C)), air has a dielectric constant of 1.00059, whereas SiN has a dielectric constant of about 9.5. Thus, it is believed that implementing an air gap spacer between the NFET and PFET devices in a fork sheet FET will significantly reduce capacitance coupling.
[0013] An exemplary method for fabricating an air-gap isolated fork-sheet FET device according to the present technology will now be described with reference to FIGS. 1-17. In each of the following figures, a cross-sectional view through a portion of the fork-sheet FET device structure is provided. See, for example, FIG. 1, which shows a top view of the overall layout of the present fork-sheet FET device structure as it will be depicted in the figures, and the orientation of the various cuts depicted in the figures. As shown in FIG. 1, according to an exemplary embodiment, the present fork-sheet FET device includes at least one n-channel FET (NFET) device (labeled "NFET") and at least one p-channel (PFET) device (labeled "PFET"), separated by an air-gap isolated pillar (labeled "air-gap pillar"). For clarity, the terms "first" and "second" may also be used herein to refer to the NFET and PFET devices, respectively. As will be described in more detail below, each NFET / PFET device includes a stack of layers (e.g., nanosheets) that extend horizontally along a first direction (in this case, the X-direction) on either side of an air-gap pillar. As will be described in more detail below, the air gap pillar is a vertical structure disposed directly between the NFET stack and the PFET stack. Above the NFET stack and the PFET stack is the gate (labeled "gate") of the fork-sheet FET device. As shown in Figure 1, the gate extends along a second direction (in this case the Y direction) that is orthogonal to the first direction / X direction.
[0014] The pattern used for the gates in Figure 1 is representative of sacrificial gates that will be placed on top of the NFET device stack and the PFET device stack early in the process. That is, as described in more detail below, this example employs a replacement gate process in which these sacrificial gates act as placeholders during source / drain region formation and are later replaced by the final gates of the fork-sheet FET devices (also referred to herein as "replacement gates"). It should be noted, however, that the orientation of the gates relative to the NFET and PFET device stacks is the same for both the sacrificial and replacement gates.
[0015] As shown in FIG. 1, cross section XX′ provides a cut perpendicular to the gates through the stack of PFET devices. It should be noted that the process depicted with reference to cross section XX′ through the stack of PFET devices would be performed in exactly the same manner in a stack of NFET devices and would therefore look the same. Cross section Y1-Y1′ provides a cut perpendicular to the stack of NFET devices and the stack of PFET devices between two of the gates. Another cross section Y2-Y2′ provides a cut perpendicular to the stack of NFET devices and the stack of PFET devices through one of the gates.
[0016] The process begins with forming a stack 204 of alternating sacrificial and active layers on a substrate 202, followed by patterning of the stack 204. See Figure 2A (XX' cross section) and Figure 2B (Y1-Y1' and Y2-Y2' cross sections). That is, at this stage of the process, the Y1-Y1' and Y2-Y2' cross sections will look the same, i.e., as shown in Figure 2B.
[0017] According to an exemplary embodiment, the substrate 202 is a bulk semiconductor wafer, such as bulk silicon (Si), bulk germanium (Ge), bulk silicon germanium (SiGe), or bulk III-V semiconductor wafer, or a combination thereof. Alternatively, the substrate 202 can be a semiconductor-on-insulator (SOI) wafer. An SOI wafer includes an SOI layer separated from the underlying substrate by a buried insulator. When the buried insulator is an oxide, it is also referred to herein as a buried oxide or BOX. The SOI can include any suitable semiconductor, such as Si, Ge, SiGe, or III-V semiconductors, or a combination thereof. The substrate 202 may already have pre-built structures (not shown), such as transistors, diodes, capacitors, resistors, interconnects, wiring, etc.
[0018] The stack 204 is formed by depositing sacrificial and active layers on top of each other on the substrate 202. According to an exemplary embodiment, the sacrificial and active layers in the stack 204 are nanosheets. As used herein, the term "nanosheet" generally refers to a sheet or layer having nanoscale dimensions. Furthermore, the term "nanosheet" is intended to encompass other nanoscale structures, such as nanowires. For example, the term "nanosheet" can refer to a nanowire having a larger width, or the term "nanowire" can refer to a nanosheet having a smaller width, or vice versa.
[0019] In this example, stack 204 includes a (first) sacrificial layer 206 deposited directly on substrate 202, alternating (second) sacrificial layers 208, 208', 208'', 208''', etc. and active layers 210, 210', 210'', etc. deposited on first sacrificial layer 206. As used herein, the term "sacrificial" refers to structures (sacrificial layer 206 / sacrificial layers 208, 208', 208'', 208''', etc.) that are removed in whole or in part during fabrication of a fork-sheet FET device. In contrast, active layers 210, 210', 210'', etc. remain in place and will function as channels of the fork-sheet FET device, as described in more detail below. It should be noted that the number of sacrificial layers 208, 208', 208", 208'", etc. and active layers 210, 210', 210", etc. shown in the figures are provided merely as examples intended to illustrate the present technology, and embodiments are contemplated herein in which more or fewer sacrificial layers 208, 208', 208", 208'", etc. or more or fewer active layers 210, 210', 210", etc., or combinations thereof, are present in the stack 204 than are shown.
[0020] By way of example only, sacrificial layer 206 and each of sacrificial layers 208, 208', 208'', 208''', etc. and active layers 210, 210', 210'', etc. may be deposited on substrate 202 using an epitaxial growth process. According to an exemplary embodiment, sacrificial layer 206 and each of sacrificial layers 208, 208', 208'', 208''', etc. and active layers 210, 210', 210'', etc. have a thickness ranging from about 6 nanometers (nm) to about 25 nm and therebetween.
[0021] The materials employed for the sacrificial and active layers are such that the sacrificial layers 208, 208', 208", 208'", etc. can be selectively removed with respect to the active layers 210, 210', 210", etc. later in the process. Additionally, the materials employed for the first and second sacrificial layers are such that the sacrificial layer 206 can be selectively removed with respect to the sacrificial layers 208, 208', 208", 208'", etc. later in the process. This allows for the formation of a lower dielectric isolation layer. Advantageously, the lower dielectric isolation layer prevents leakage current of the source / drain regions through the substrate 202.
[0022] For example, according to an exemplary embodiment, sacrificial layer 206 and each of sacrificial layers 208, 208', 208'', 208''', etc. are formed from SiGe, while each of active layers 210, 210', 210'', etc. are formed from Si. Etchants such as hot wet SC1, vapor phase hydrogen chloride (HCI), vapor phase chlorine trifluoride (CIF3), and other reactive cleaning processes (RCPs) can then be employed for selective removal of the SiGe sacrificial layers relative to the Si active layers.
[0023] Additionally, SiGe with a high germanium (Ge) content can be selectively removed relative to SiGe with a low Ge content using an etchant such as dry HCI. Thus, according to an exemplary embodiment, sacrificial layer 206 is formed from SiGe having a high Ge content, while each of sacrificial layers 208, 208', 208'', 208''', etc. is formed from SiGe having a low Ge content. By way of example only, high Ge content SiGe is considered herein to be SiGe having about 50% Ge to about 100% Ge (i.e., pure Ge), and ranges therebetween. For example, in one non-limiting example, sacrificial layer 206 is formed from SiGe60 (SiGe having about 60% Ge content). In contrast, low Ge content SiGe is considered herein to be SiGe having about 20% Ge to about 50% Ge, and ranges therebetween. For example, in one non-limiting example, each of the sacrificial layers 208, 208', 208'', 208''', etc. are formed from SiGe30 (SiGe with approximately 30% Ge content). This configuration allows the sacrificial layer 206 to be selectively removed relative to the sacrificial layers 208, 208', 208'', 208''', etc. during formation of the lower dielectric isolation layer (see below).
[0024] As highlighted above, the stack 204 is then patterned. Patterning the stack 204 allows for the formation of shallow trench isolation (STI) regions in the substrate 202 at the bottom of the NFET and PFET stacks. Standard lithography and etching techniques can be employed to pattern the stack 204. A lithography stack (not shown), e.g., photoresist / antireflective coating (ARC) / organic planarization layer (OPL), is used in a standard lithography and etching process (not shown) to pattern a hard mask 212 on the stack 204. Suitable hard mask materials include, but are not limited to, nitride hard mask materials such as silicon nitride (SiN), silicon oxynitride (SiON), silicon carbonitride (SiCN), and / or oxide hard mask materials such as silicon oxide (SiOX). Alternatively, the hard mask 212 can be formed by other suitable techniques, including, but not limited to, sidewall image transfer (SIT), self-aligned double patterning (SADP), self-aligned quadruple patterning (SAQP), and other self-aligned multiple patterning (SAMP).
[0025] Etching is then used to transfer the pattern from the hard mask 212 into the sacrificial and active layers of the stack 204, and partway through the underlying substrate plate 202. A directional (anisotropic) etching process such as reactive ion etching (RIE) can be employed to etch the stack 204 / substrate 202. At this point, a trench 214 exists in the substrate 202 at the base of the (patterned) stack 204, as shown in FIG. 2B.
[0026] The NFET-PFET spaces 304 are then opened in the stack 204, see FIG. 3A (XX′ cross section) and FIG. 3B (Y1-Y1′ and Y2-Y2′ cross sections). To this end, a patterned block mask 302 can first be formed on the stack 204. Etching can then be used to transfer the pattern from the block mask 302 to the hard mask 212 and the underlying stack 204 to form the NFET-PFET spaces 304 in the stack 204. Suitable materials for the block mask 302 include, but are not limited to, organic planarization layer (OPL) materials that can be deposited on the stack 204 using casting methods such as spin-on or spray coating, chemical vapor deposition (CVD), atomic layer deposition (ALD), or physical vapor deposition (PVD). Standard lithography and etching techniques (see above) can be employed to pattern the block mask 302.
[0027] Etching the NFET-PFET space 304 may employ a directional (anisotropic) etching process, such as RIE. The NFET-PFET space 304 divides the stack 204 into at least one stack 204N of NFET devices and at least one stack 204P of PFET devices. As shown in FIG. 3B, the stack 204N of NFET devices and the stack 204P of PFET devices are separated from each other by the NFET-PFET space 304. It should be noted that the process depicted in FIGS. 3A and 3B is merely one exemplary and non-limiting embodiment contemplated herein for forming the NFET-PFET space 304. That is, in accordance with the present technique, the NFET-PFET space 304 may be formed using any other suitable lithographic or similar patterning process.
[0028] A sacrificial material is then deposited and filled in the NFET-PFET space 304 to form a sacrificial pillar 402 separating the stack of NFET devices 204N and the stack of PFET devices 204P, after which the block mask 302 is removed and an STI region 404 is formed in the trench 214 at the base of the stack of NFET / PFET devices 204N / 204P. See FIG. 4 (Y1-Y1′ and Y2-Y2′ cross-sectional views). Suitable sacrificial materials for the sacrificial pillar 402 include, but are not limited to, aluminum oxide (ALOx), silicon carbide (SiC), or titanium oxide (TiOx), or combinations thereof, which can be deposited in the NFET-PFET space 304 using processes such as CVD, ALD, or PVD. After deposition, the sacrificial material can be etched back using a plasma dry etching process. As described above, the block mask 302 can be formed from an OPL material. In that case, the block mask 302 can be removed using an ashing process.
[0029] According to an exemplary embodiment, the STI region 404 is formed by filling the trench 214 with a dielectric material, such as an oxide material (also generally referred to herein as "STI oxide"), and then backing off the STI oxide. Although not explicitly shown, a liner (e.g., thermal oxide or silicon nitride (SiN)) may be deposited in the trench 214 prior to the STI oxide. Suitable STI oxides include, but are not limited to, silicon oxide (SiOx). Processes such as CVD, ALD, or PVD may be employed to deposit the STI oxide in the trench 214. Following deposition, an oxide-selective etch may then be used to back off the STI oxide.
[0030] The hard mask 212 is then removed, a sacrificial gate 504 is formed on the stack of NFET devices 204N and the stack of PFET devices 204P, and the sacrificial layer 206 is removed to form a cavity 506 below the stack of NFET devices 204N and the stack of PFET devices 204P, see Figure 5A (XX' cross section), Figure 5B (Y1-Y1' cross section), and Figure 5C (Y2-Y2' cross section).
[0031] To form the sacrificial gate 504, a sacrificial material is first blanket deposited on the NFET device stack 204N and the PFET device stack 204P. Suitable sacrificial materials include, but are not limited to, polysilicon (poly-Si) and / or amorphous silicon (a-Si), which may be deposited using processes such as CVD, ALD, or PVD. Although not shown, it is preferable to first form a thin (e.g., about 1 nanometer (nm) to about 5 nm) layer of silicon oxide (SiOx) on the NFET device stack 204N and the PFET device stack 204P, followed by deposition of poly-Si and / or a-Si.
[0032] Standard lithography and etching techniques (see above) are then used to pattern the sacrificial gate hard mask 502 on the sacrificial material to mark the footprint and location of each of the sacrificial gates 504. Suitable materials for the sacrificial gate hard mask 502 include, but are not limited to, nitride hard mask materials such as SiN, SiON, or silicon carbonitride (SiCN), or combinations thereof, or oxide hard mask materials such as SiOx, or both. Alternatively, the sacrificial gate hard mask 502 can be formed by other suitable techniques, including, but not limited to, SIT, SADP, SAQP, and other SAMPs. Etching is then used to transfer the pattern from the sacrificial gate hard mask 502 to the sacrificial material to form the individual sacrificial gates 504. A directional (anisotropic) etching process such as RIE can be employed to etch the sacrificial gates.
[0033] As highlighted above, the sacrificial gate 504 will be removed later in the process and replaced with a replacement gate that will function as the final gate of the fork-sheet FET device. This is called a "gate-last" approach because the replacement gate is formed last, near the end of the process. The use of a gate-last approach is advantageous because it prevents the metal gate stack material from being exposed to potentially damaging conditions during subsequent processing steps. For example, the high-κ dielectrics used for the replacement gate can be damaged by exposure to high temperatures during the formation of the source / drain regions. Therefore, in this scheme, the metal gate stack material is first laid down near the end of the process.
[0034] A lower dielectric isolation layer will be formed in the cavity 506 beneath the stack of NFET devices 204N and the stack of PFET devices 204P (see below). The lower dielectric isolation layer serves to prevent leakage current of the source / drain regions through the substrate 202. The cavity 506 is formed by selective removal of the sacrificial layer 206. As described above, the sacrificial layer 206 can be formed from SiGe with high Ge content (e.g., SiGe having about 50% Ge to about 100% Ge (i.e., pure Ge) and ranges therebetween, e.g., SiGe60). The sacrificial layer 206 can then be selectively removed using an etchant such as dry HCl to form the cavity 506.
[0035] A dielectric spacer material is deposited on the sacrificial gate 504 and in the cavity 506 to fill it, followed by a directional (anisotropic) etching process, such as RIE, which is used to pattern the dielectric spacer material in the cavity 506 as the lower dielectric isolation layer 602 and as the gate spacer 604 aligned with the sacrificial gate hard mask 502 / sacrificial gate 504. See FIG. 6A (XX′ cross section), FIG. 6B (Y1-Y1′ cross section), and FIG. 6C (Y2-Y2′ cross section). Suitable dielectric spacer materials include, but are not limited to, SiOx, SiC, silicon oxycarbide (SiCO) or SiN, or combinations thereof, which can be deposited using processes such as CVD, ALD, or PVD.
[0036] The sacrificial gate hard mask 502 / sacrificial gate 504 and gate spacer 604 are then used as a mask to pattern trenches 702 in the stack of NFET devices 204N and the stack of PFET devices 204P between the sacrificial gates 504. See FIG. 7A (XX′ cross section) and FIG. 7B (Y1-Y1′ cross section). A directional (anisotropic) etching process such as RIE can be employed to etch the trenches. As shown in FIG. 7A and FIG. 7B, the trenches 702 extend through each of the sacrificial layers 208, 208′, 208″, 208′″, etc. and the active layers 210, 210′, 210″, etc., stopping on the lower dielectric isolation layer 602.
[0037] Next, first, an inner spacer 802 is formed alongside the sacrificial layers 208, 208', 208'', 208''', etc., and then, on opposite sides of the sacrificial gate 504, an NFET source / drain region 804 and a PFET source / drain region 806 are formed in the trench 702 alongside the sacrificial layers 208, 208', 208'', 208''', etc. and active layers 210, 210', 210'', etc. of the NFET device stack 204N and the PFET device stack 204P, respectively. See FIG. 8A (XX' cross section) and FIG. 8B (Y1-Y1' cross section). To form the inner spacer 802, a selective lateral etch is performed to recess the sacrificial layers 208, 208', 208'', 208''', etc. exposed along the sidewalls of the trench 702. 8A, this setback etch forms pockets along the sidewalls of trench 702, which are then filled with a spacer material to form interior spacers 802 within the pockets. These interior spacers 802 serve to offset the replacement gates from the NFET / PFET source / drain regions 804 / 806 (see below).
[0038] As noted above, the sacrificial layers 208, 208′, 208″, 208′″, etc. may be formed from SiGe. The recess etching to form the pocket may then employ a SiGe-selective non-directional (isotropic) etching process, such as a wet chemical etch or a vapor phase etch. Suitable materials for the inner spacer 802 include, but are not limited to, SiN, SiOx, SiC, or SiCO, or combinations thereof, which may be deposited in the pocket using processes such as CVD, ALD, or PVD. Following deposition, excess inner spacer material may be removed from the trench 702 using an isotropic etching process, such as a wet etch or a selective dry etch.
[0039] According to an exemplary embodiment, NFET / PFET source / drain regions 804 / 806 are formed from epitaxial material, such as epitaxial Si, epitaxial SiGe, etc., that is in-situ doped (i.e., dopants are introduced during growth) or ex-situ doped (e.g., dopants are introduced via ion implantation). Suitable n-type dopants include, but are not limited to, phosphorus (P) and / or arsenic (AS). Suitable p-type dopants include, but are not limited to, boron (B). It is noted that when the inner spacer 802 is in place along the sidewall of the trench 702, the epitaxial growth of the NFET / PFET source / drain regions 804 / 806 is templated only from the ends of the active layers 210, 210′, 210″, etc. of the NFET / PFET device stack 204N / 204P that are exposed along the sidewall of the trench 702. As shown in FIGS. 8A and 8B, the NFET / PFET source / drain regions 804 / 806 are separated from the substrate 202 by a lower dielectric isolation layer 602.
[0040] 8B, the NFET / PFET source / drain regions 804 / 806 are disposed on opposite sides of the sacrificial pillar 402. However, the sacrificial pillar 402 is taller than the NFET / PFET source / drain regions 804 / 806, such that the NFET / PFET source / drain regions 804 / 806 are aligned with a lower portion 810 of the sacrificial pillar 402, while an upper portion 812 of the sacrificial pillar 402 above the NFET / PFET source / drain regions 804 / 806 is exposed. Alternatively, the NFET / PFET source / drain regions 804 / 806 are not aligned with the upper portion 812 of the sacrificial pillar 402.
[0041] Next, the upper portion 812 of the sacrificial pillar 402 exposed above the NFET / PFET source / drain regions 804 / 806 is trimmed. See FIG. 9 (Y1-Y1′ cross-section). The (trimmed) upper portion of the sacrificial pillar 402 is now given the reference numeral 812a. To trim the upper portion 812a of the sacrificial pillar 402, a non-directional (i.e. isotropic) etching process such as wet chemical etching or gas phase etching can be employed. The upper portion 812a of the sacrificial pillar 402 is now given the reference numeral 812a. 9, the lower portion 810 of the sacrificial pillar 402 has a width W1 SACRIFICIAL PILLAR and the (trimmed) upper portion 812a of the sacrificial pillar 402 has a width W2 SACRIFICIAL PILLAR At this time, W2 SACRIFICIAL PILLAR is W1 SACRIFICIAL PILLAR is smaller than W2 SACRIFICIAL PILLAR <W1 SACRIFCIAL PILLAR Advantageously, trimming the sacrificial pillar 402 in this manner allows a replacement spacer to be formed in its place, including an air gap, as will be described in more detail below. More specifically, the trimmed upper portion 812a of the sacrificial pillar 402 serves to create a bottle neck that is closed before the underlying space can be completely filled, thereby creating the air gap. Note that during trimming, a slight reduction in the height of the upper portion 812a of the sacrificial pillar 402 may occur, although this effect is negligible. Additionally, the trimming of the sacrificial pillar 402 is now performed between the NFET / PFET source / drain regions 804 / 806, thereby allowing the present air gap pillar to be formed between the NFET / PFET source / drain regions 804 / 806. As will be described in more detail below, the sacrificial pillar 402 (now below the sacrificial gate 504) is also trimmed so that the present air gap pillar can also be formed between the NFET / PFET gates.
[0042] An interlayer dielectric (ILD) 1002 is then deposited over and fills the NFET / PFET device stack 204N / 204P (and the sacrificial pillar 402 between them), the sacrificial gate 504 and gate spacer 604, and the NFET / PFET source / drain regions 804 / 806. See FIG. 10A (XX′ cross section), FIG. 10B (Y1-Y1′ cross section), and FIG. 10C (Y2-Y2′ cross section). Suitable ILD 1002 materials include, but are not limited to, oxide low-κ materials such as SiOx and / or oxide ultra-low-κ interlayer dielectric (ULK-ILD) materials, e.g., having a dielectric constant κ less than 2.7. In comparison, silicon dioxide (SiO2) has a dielectric constant κ value of 3.9. Suitable ULK-ILD materials include, but are not limited to, porous organosilicate glass (pSiCOH). The ILD 1002 may be deposited using a process such as CVD, ALD, or PVD. Following deposition, the ILD 1002 may be polished using a process such as chemical mechanical polishing (CMP). According to an exemplary embodiment, the ILD 1002 is polished down to the sacrificial gate 504 and the sacrificial gate hard mask 502 is removed. In this way, the sacrificial gate 504 may be removed selectively to the ILD 1002 and replaced with a gate (see below).
[0043] As highlighted above, the trimming of the sacrificial pillar 402 between the NFET / PFET gates is also performed. Since this portion of the sacrificial pillar 402 is now under the sacrificial gate 504, a partial recession of the sacrificial gate 504 is first realized to expose the upper portion 812 of the sacrificial pillar 402 in this region. See FIG. 11A (XX′ cross section) and FIG. 11B (Y2-Y2′ cross section). A selective etching process such as RIE can be employed to etch the recession of the sacrificial gate 504. As shown in FIGS. 11A and 11B, by recessing the sacrificial gate 504 only partially, the underlying stack of NFET devices 204N and stack of PFET devices 204P remain covered and therefore protected during the trimming of the sacrificial pillar 402.
[0044] 11B, the stack of NFET devices 204N and the stack of PFET devices 204P are disposed on opposite sides of the sacrificial pillar 402. However, the sacrificial pillar 402 is taller than the stack of NFET devices 204N and the stack of PFET devices 204P, such that the stack of NFET devices 204N and the stack of PFET devices 204P are aligned with the lower portion 810 of the sacrificial pillar 402, while the upper portion 812 of the sacrificial pillar 402 above the stack of NFET devices 204N and the stack of PFET devices 204P is exposed. In other words, the stack of NFET devices 204N and the stack of PFET devices 204P are not aligned with the upper portion 812 of the sacrificial pillar 402.
[0045] Next, the upper portion 812 of the sacrificial pillar 402 exposed above the stack of NFET devices 204N and the stack of PFET devices 204P is trimmed. See FIG. 12 (Y2-Y2′ cross section). As mentioned above, the (trimmed) upper portion of the sacrificial pillar 402 is given the reference 812a. To trim the upper portion 812a of the sacrificial pillar 402, a non-directional (i.e. isotropic) etching process such as wet chemical etching or gas phase etching can be employed. In the same manner as above, trimming the upper portion 812a of the sacrificial pillar 402 reduces its width, i.e., with respect to the lower portion 810 of the sacrificial pillar 402. That is, as shown in FIG. 12, the lower portion 810 of the sacrificial pillar 402 has a width W1 SACRIFICIAL PILLAR and the (trimmed) upper portion 812 of the sacrificial pillar 402 has a width W2 SACRIFICIAL PILLAR Then, W2 SACRIFICIAL PILLAR is W1 SACRIFICIAL PILLAR is smaller than W2 SACRIFICIAL PILLAR <W1 SACRIFICIAL PILLARAs will be described in more detail below, trimming the sacrificial pillar 402 in this manner advantageously creates a bottle neck that is blocked before the underlying space can be completely filled, thereby allowing a replacement spacer, including an air gap, to be formed in its place. Note that during trimming, a slight reduction in the height of the upper portion 812a of the sacrificial pillar 402 may occur, although this effect is negligible. The upper portion 812a of the sacrificial pillar 402 has now been trimmed both at the source / drain regions of the fork sheet FET devices (i.e., between the NFET / PFET source / drain regions 804 / 806) and at the gate regions of the fork sheet FET devices (i.e., between the stack 204N of NFET devices and the stack 204P of PFET devices), which advantageously allows for the formation of this air gap pillar between the NFET / PFET source / drain regions 804 / 806 and between the NFET / PFET gates, as highlighted above.
[0046] The remaining portions of the sacrificial gate 504 are then selectively removed to form gate trenches 1302 in the ILD 1002 between the NFET / PFET source / drain regions 804 / 806 and over the stack of NFET devices 204N and the stack of PFET devices 204P, respectively. See FIG. 13A (XX′ cross section) and FIG. 13B (Y2-Y2′ cross section). As shown in FIG. 13A and FIG. 13B, the sacrificial layers 208, 208′, 208″, 208′″, etc. that are now accessible through the gate trenches 1302 are also removed. The removal of the sacrificial layers 208, 208′, 208″, 208′″, etc. releases the active layers 210, 210′, 210″, etc. from the stack of NFET devices 204N and the stack of PFET devices 204P. Gaps 1304 now exist between the active layers 210, 210', 210'', etc. in the stack of NFET devices 204N and the stack of PFET devices 204P. The active layers 210, 210', 210'', etc. function as channels for the fork-sheet FET devices. Releasing the active layers 210, 210', 210'', etc. from the stack of NFET devices 204N and the stack of PFET devices 204P allows for the formation of a replacement gate that surrounds at least a portion of the channel (i.e., the active layers 210, 210', 210'', etc.) in a gate-all-around or GAA configuration.
[0047] Then, a replacement gate 1402 is formed in the gate trench 1302 and the gap 1304. See FIG. 14A (XX′ cross-sectional view) and FIG. 14B (Y2-Y2′ cross-sectional view). As shown in the close-up view 1404, according to an exemplary embodiment, each of the replacement gates 1402 includes a gate dielectric 1408 and a gate conductor 1410 disposed on the gate dielectric 1408. Although not explicitly shown in the figures, a thin (e.g., about 0.3 nm to about 5 nm) interfacial oxide (e.g., silicon oxide (SiOx) that may include other chemical elements therein, such as nitrogen (N), germanium (Ge)) can first be formed on exposed surfaces of the active layers 210, 210′, 210″, etc., and then the gate dielectric 1408 can be deposited on the interfacial oxide using a process such as CVD, ALD, or PVD.
[0048] Suitable materials for the gate dielectric 1408 include, but are not limited to, silicon oxide (SiOx), SiN, silicon oxynitride (SiOxNy), high-κ materials, or any combination thereof. As used herein, the term "high-κ" refers to a material that has a relative dielectric constant κ much higher than silicon dioxide (e.g., the dielectric constant κ is about 25 for hafnium oxide (HfO2) rather than 3.9 for SiO2). Suitable high-κ materials include, but are not limited to, metal oxides such as HfO2, hafnium silicon oxide (HfSiO), hafnium silicon oxide nitride (HfSiO), lanthanum oxide (La2O3), lanthanum aluminum oxide (LaAlO3), zirconium oxide (ZrO2), zirconium silicon oxide (ZrSiO4), zirconium silicon oxide nitride (ZrSiOxNy), tantalum oxide (TaOx), titanium oxide (TiO), barium strontium titanium oxide (BaO6SrTi2), barium titanium oxide (BaTiO3), strontium titanium oxide (SrTiO3), yttrium oxide (Y2O3), aluminum oxide (Al2O3), lead scandium tantalum oxide (Pb(Sc,Ta)O3), or lead zinc niobate (Pb(Zn,Nb)O), or combinations thereof. The high-κ material may further include a dopant, such as lanthanum (La), aluminum (Al), or magnesium (Mg), or a combination thereof. The gate dielectric 1408 may be deposited using a process, such as, but not limited to, thermal oxidation, chemical oxidation, thermal nitridation, plasma oxidation, plasma nitridation, CVD, ALD, or a combination of these processes. According to an exemplary embodiment, the gate dielectric 1408 has a thickness ranging from about 1 nm to about 5 nm and therebetween.
[0049] Suitable materials for the gate conductor 1410 include, but are not limited to, doped polysilicon and / or at least one work function setting metal. Suitable work function setting metals include, but are not limited to, titanium nitride (TiN), titanium aluminum nitride (TiAlN), hafnium nitride (HfN), hafnium silicon nitride (HfSiN), tantalum nitride (TaN), tantalum silicon nitride (TaSiN), tungsten nitride (WN), molybdenum nitride (MoN), niobium nitride (NbN), titanium carbide (TiC), titanium aluminum carbide (TiAlC), tantalum carbide (TaC), or hafnium carbide (HfC), or combinations thereof. The gate conductor 1410 can be deposited using processes such as, but not limited to, CVD, ALD, PVD, sputtering, plating, evaporation, ion beam evaporation, electron beam evaporation, laser assisted evaporation, chemical solution deposition, or combinations of these processes. According to an exemplary embodiment, the gate conductor 1410 has a thickness ranging from about 5 nm to about 15 nm and therebetween. In the exemplary embodiment depicted and illustrated in Figures 14A and 14B, the gate conductor 1410 includes at least one layer 1410a of the work function setting metal described above, and a (low resistivity) fill metal 1410b disposed on the layer 1410a of the work function setting metal so as to fill any remaining space in the replacement gate 1402. Suitable low resistivity fill metals include, but are not limited to, tungsten (W) and / or aluminum (Al), which may be deposited using processes including, but not limited to, CVD, ALD, PVD, sputtering, plating, evaporation, ion beam evaporation, electron beam evaporation, laser assisted evaporation, chemical solution deposition, and the like, or a combination of these processes. As shown in FIGS. 14A and 14B, the replacement gate 1402, i.e., the gate dielectric 1408 and the gate conductor 1410, completely surrounds at least a portion of each of the active layers 210, 210', 210'', etc. in a gate-all-around configuration.
[0050] The replacement gate 1402, i.e., the gate dielectric 1408 and the gate conductor 1410, are then recessed to form a dielectric cap 1502 on the (recessed) replacement gate 1402. See FIG. 15A (XX′ cross section), FIG. 15B (Y1-Y1′ cross section), and FIG. 15C (Y2-Y2′ cross section). Suitable dielectric cap materials include, but are not limited to, SiOx and / or SiN, which can be deposited using processes such as CVD, ALD, or PVD. After deposition, the dielectric cap material can be planarized using a process such as CMP. As shown in FIG. 15B, during the polishing of the dielectric cap 1502, the ILD 1002 is also recessed, thereby exposing the tops of the sacrificial pillars 402 in the source / drain regions of the fork sheet FET device. As shown in FIG. 15C, the polishing of the dielectric cap 1502 also exposes the tops of the sacrificial pillars 402 in the gate region of the fork sheet FET device. By exposing the tops of the sacrificial pillars 402, it is possible to selectively remove the sacrificial pillars 402 and replace them with (replacement) air gap pillars.
[0051] That is, the sacrificial pillar 402 is then selectively removed to form an opening 1602 between the NFET source / drain region 804 and the PFET source / drain region 806 in the source / drain region of the fork sheet FET device, and between the replacement gate 1402 on the stack of NFET devices 204N and the stack of PFET devices 204P in the gate region of the fork sheet FET device. See FIG. 16A (Y1-Y1′ cross section) and FIG. 16B (Y2-Y2′ cross section). Based on the above process, the opening 1602 formed by removing the sacrificial pillar 402 has a unique shape. That is, by trimming the upper portion of the sacrificial pillar 402 as described above, the lower portion 1604 of the opening 1602 has a width W1 OPENING and the upper portion 1606 of the opening 1602 has a width W2 OPENING At this time, W2 OPENING is W1 OPENING is smaller than W2 OPENING <W1OPENING This configuration, having a narrow top utilized to fill the opening 1602, serves to create a bottle neck for the filling process such that the lower portion 1604 of the opening 1602 becomes blocked before it can be completely filled, thus creating an air gap.
[0052] For example, a conformal dielectric liner 1702 is then deposited in the opening 1602 to line the opening 1602. See FIG. 17A (Y1-Y1′ cross section) and FIG. 17B (Y2-Y2′ cross section). Due to the narrowed entrance at the upper portion 1606 of the opening 1602 utilized while depositing the conformal dielectric liner 1702, the upper portion 1606 of the opening 1602 will be blocked before the lower portion 1604 of the opening 1602 is completely filled by the dielectric liner 1702. As shown in FIG. 17A and FIG. 17B, the dielectric liner 1702 deposited in this manner lines the lower portion 1604 of the opening 1602 and completely surrounds the air gap 1704 in the center of the lower portion 1604 of the opening 1602, which is the dielectric air gap pillar 1706 of the present invention. Suitable materials for the dielectric liner 1702 include, but are not limited to, SiOx, SiN, SiC, SiON, or SiCN, or combinations thereof, which may be deposited in the opening 1602 using a process such as CVD, ALD, or PVD. Following deposition, excess material may be removed using a process such as CMP.
[0053] As shown in FIGS. 17A and 17B, based on the unique shape of the opening 1602, the lower portion 1708 of the dielectric air gap pillar 1706 has a width W1 AIRGAP PILLAR At this time, W2 AIRGAP PILLAR is W1 AIRGAP PILLAR is smaller than W2 AIRGAP PILLAR <W1 AIRGAP PILLARThe dielectric air gap pillar 1706 is disposed directly between the NFET and PFET nanosheet devices, which in this particular example are shown to the right and left of the dielectric air gap pillar 1706, respectively. The nanosheets (i.e., active layers 210, 210′, 210″, etc.) of the NFET and PFET nanosheet devices extend horizontally on either side of the dielectric air gap pillar 1706.
[0054] In particular, as shown in Figure 17A, the dielectric air gap pillar 1706 directly contacts and separates the NFET source / drain region 804 and the PFET source / drain region 806 of the fork sheet FET device. As shown in Figure 17B, the dielectric air gap pillar 1706 directly contacts and separates the replacement gate 1402 on top of the stack of NFET devices 204N and the stack of PFET devices 204P (also referred to herein as the "NFET gate and PFET gate," respectively) in the gate region of the fork sheet FET device. As discussed above, by implementing the dielectric air gap pillar 1706 in this manner between the NFET and PFET devices, capacitance coupling is significantly reduced in the present fork sheet FET design.
[0055] Although exemplary embodiments of the present invention have been described herein, it should be understood that the invention is not limited to these precise embodiments, and that various other changes and modifications may be made by those skilled in the art without departing from the scope of the present invention.
Claims
1. 1. A fork seat field effect transistor (FET) device, comprising: At least a first nanosheet FET and a second nanosheet FET; a dielectric pillar disposed directly between the first nanosheet FET and the second nanosheet FET, the dielectric pillar comprising an air gap.
2. The fork-sheet FET of claim 1 , wherein the first nanosheet FET and the second nanosheet FET comprise nanosheets extending horizontally on either side of the dielectric pillar.
3. 3. The fork sheet FET of claim 2, wherein the dielectric pillar comprises a liner completely surrounding the air gap, the liner comprising a material selected from the group consisting of silicon oxide (SiOx), silicon nitride (SiN), silicon oxynitride (SiON), silicon carbonitride (SiCN), and combinations thereof.
4. The fork-sheet FET of claim 1 , wherein the first nanosheet FET comprises an n-channel FET (NFET) and the second FET comprises a p-channel FET (PFET).
5. The fork-sheet FET of claim 1 , wherein the first nanosheet FET and the second nanosheet FET contact opposite sides of the dielectric pillar.
6. The fork-sheet FET of claim 1 , wherein said dielectric pillar comprises a liner completely surrounding said air gap.
7. 10. The fork-seat FET of claim 1, wherein said liner comprises a material selected from the group consisting of silicon oxide (SiOx), silicon nitride (SiN), silicon oxynitride (SiON), silicon carbonitride (SiCN), and combinations thereof.
8. The fork-sheet FET of claim 1 or 2, wherein the dielectric pillar separates a gate of the first nanosheet FET from a gate of the second nanosheet FET.
9. The fork-sheet FET of claim 1 or 2, wherein the dielectric pillar separates the source / drain regions of the first nanosheet FET from the source / drain regions of the second nanosheet FET.
10. The lower portion of the dielectric pillar has a width W1 AIRGAP PILLAR The upper portion of the opening has a width W2 AIRGAP PILLAR and W2 AIRGAP PILLAR <W1 AIRGAP PILLAR 2. The fork seat FET of claim 1, wherein:
11. 1. A method of forming a fork sheet field effect transistor (FET) device, comprising: forming at least a first nanosheet FET and a second nanosheet FET having a sacrificial pillar disposed directly therebetween; removing the sacrificial pillar to form an opening between the first nanosheet FET and the second nanosheet FET, a lower portion of the opening having a width W1 OPENING and the upper portion of the opening has a width W2 OPENING and W2 OPENING <W1 OPENING and depositing a dielectric liner in the opening to line the opening, and during the deposition W2 OPENING <W1 OPENING and the dielectric liner blocks the upper portion of the opening before the lower portion of the opening is completely filled based on the above formula (1), thereby creating a dielectric pillar having an air gap between the first nanosheet FET and the second nanosheet FET.
12. The lower portion of the sacrificial pillar has a width W1 SACRIFICIAL PILLAR and the upper portion of the sacrificial pillar has a width W2 SACRIFICIAL PILLAR and W2 SACRIFICIAL PILLAR <W1 SACRIFICIAL PILLAR The method of claim 11 , further comprising trimming the upper portion of the sacrificial pillar so that
13. 12. The method of claim 11, wherein the sacrificial pillars comprise a material selected from the group consisting of aluminum oxide (AlOx), silicon carbide (SiC), titanium oxide (TiOX), and combinations thereof.
14. 12. The method of claim 11 , wherein the first nanosheet FET comprises an n-channel FET (NFET) and the second FET comprises a p-channel FET (PFET).
15. The method of claim 11 , wherein the dielectric liner completely surrounds the air gap.
16. 12. The method of claim 11, wherein the dielectric liner comprises a material selected from the group consisting of silicon oxide (SiOx), silicon nitride (SiN), silicon oxynitride (SiON), silicon carbonitride (SiCN), and combinations thereof.
17. The method of claim 11 , wherein the dielectric pillar separates a gate of the first nanosheet FET from a gate of the second nanosheet FET.
18. The method of claim 11 , wherein the dielectric pillar separates a source / drain region of the first nanosheet FET from a source / drain region of the second nanosheet FET.
Citation Information
Patent Citations
Semiconductor element having a pair of fins formed to define void, and method of manufacturing the same
JP2008124478A
Nanosheet field effect transistor with spacers between sheets
US20200035786A1
Gate-all-around devices with reduced parasitic capacitance
US20210233911A1
Semiconductor device and method for manufacturing same
WO2021229740A1
