Selective etching for nanowires
By using mixed etching gases containing fluorine-containing gas and hydrogen under low temperature conditions, combined with remote plasma to generate high-energy neutral particles, the insufficient selectivity and material loss of selective etching of silicon germanium and silicon in semiconductor manufacturing is solved, and high selectivity etching and accurate formation of nanowire structures are achieved.
Patent Information
- Application Number
- CN201980048517.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2018-07-20
- Filing Date
- 2019-07-12
- Publication Date
- 2025-08-29
- Estimated Expiration
- 2039-07-12
AI Technical Summary
The prior art is difficult to efficiently selectively etch silicon germanium and silicon in semiconductor manufacturing, and when forming nanowire structures, there are problems such as insufficient selectivity and excessive material loss.
Using a mixed etching gas containing fluorine and hydrogen at low temperatures, combined with a remote plasma to generate high-energy neutral particles, introduced into the reactor chamber through a spray head for selective etching, avoiding direct exposure to the plasma, and controlling the reactor chamber pressure and RF power to improve selectivity.
High selective etching of silicon germanium and silicon is achieved, reducing material losses, improving etching ratio, and ensuring the integrity and accuracy of nanowire structure.
Smart Images

Figure CN112470258B_ABST
Abstract
Description
[0001] CROSS-REFERENCE TO RELATED APPLICATIONS
[0002] This application claims the benefit of priority to U.S. Patent Application No. 62 / 701,314, filed on July 20, 2018, which is incorporated herein by reference for all purposes. Technical Field
[0003] The present disclosure generally relates to methods of forming semiconductor devices on semiconductor wafers. More particularly, the present disclosure relates to selective etching of nanowires. Background Art
[0004] When forming a semiconductor device, nanowires can be formed by selectively etching silicon germanium (SiGe) relative to silicon (Si). Nanowires can also be formed by selectively etching Si relative to SiGe. Summary of the Invention
[0005] To achieve the foregoing and in accordance with the purposes of the present disclosure, a method is provided for selectively etching silicon germanium relative to silicon in a stack, the stack being positioned on a chuck in an etching chamber. The chuck is maintained at a temperature below 15° C. The stack is exposed to an etching gas containing a fluorine-containing gas to selectively etch the silicon germanium relative to silicon.
[0006] In another embodiment, a method is provided for selectively etching silicon relative to silicon germanium in a stack, the stack being positioned on a chuck in an etching chamber. The chuck is maintained at a temperature below 15° C. The stack is exposed to an etching gas comprising hydrogen (H 2 ) and a fluorine-containing gas to selectively etch the silicon relative to the silicon germanium.
[0007] These and other features of the present invention will be described in more detail below in the detailed description of the invention and in conjunction with the following figures. BRIEF DESCRIPTION OF THE DRAWINGS
[0008] The present disclosure is illustrated by way of example and not limitation in the figures of the accompanying drawings and in which like references indicate similar elements and in which:
[0009] Figure 1 is a high-level flow chart of an implementation scheme.
[0010] Figure 2 is a schematic top view of a processing tool that can be used in one embodiment.
[0011] Figure 3 is a schematic diagram of an etching chamber that can be used in one embodiment.
[0012] Figure 4 is a schematic diagram of a computer system that can be used to implement an embodiment.
[0013] Figure 5A -D is a schematic cross-sectional view of a stack processed according to one embodiment.
[0014] Figure 6 is a more detailed flow chart of the atomic layer deposition process.
[0015] Figure 7A -D is a schematic cross-sectional view of a stack processed according to another embodiment.
[0016] Figure 8A -C is a schematic cross-sectional view of a stack processed according to another embodiment. DETAILED DESCRIPTION
[0017] The present disclosure will now be described in detail with reference to several exemplary embodiments shown in the accompanying drawings. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present disclosure. However, it will be apparent to those skilled in the art that the present disclosure can be practiced without some or all of these specific details. In other cases, well-known process steps and / or structures have not been described in detail in order to avoid unnecessarily obscuring the present disclosure.
[0018] In one embodiment, to form nanowires, a stack of alternating Si and SiGe layers is provided. The Si and SiGe layers may extend between and be connected to electrical contacts or other structures. If the SiGe layer is selectively etched away, the Si layer remains extending between the electrical contacts. The Si layer can be used as a nanowire. Similarly, if the Si layer is selectively etched away, the SiGe layer remains extending between the electrical contacts. The SiGe layer can be used as a nanowire. The dimensions of the Si and SiGe layers are so small that etching should be highly selective. Other processing should be provided to minimize the removal of unetched layers.
[0019] Figure 22 is a schematic top view of a processing tool 200 for one embodiment. Cassette 202 holds unprocessed wafers before they are processed and then holds processed wafers once all processing is complete in processing tool 200. Cassette 202 can hold many wafers, typically up to 25. An atmospheric transport module (ATM) 214 is used to transport wafers to and from cassette 202. Load lock station 205 represents at least one device for transferring wafers back and forth between the atmosphere of ATM 214 and the vacuum of vacuum transport module (VTM) 212. VTM 212 is part of the processing tool and connects to multiple chambers. Different types of chambers may be present. In this embodiment, there are two breakout chambers 216, one etch chamber 220, and two atomic layer deposition (ALD) chambers 224. A robotic system within vacuum transport module 212 uses robotic arms to move wafers with stacks between load lock station 205 and the various chambers 216, 220, and 224. The ATM 214 uses a robotic system to transfer wafers between the cassette 202 and the load lock station 205 which is in a vacuum environment.
[0020] Figure 3 is a more detailed schematic diagram of an etch chamber 220 that can be used in one embodiment. In one or more embodiments, the etch chamber 220 includes a showerhead 306 and a chuck 308 located within a reactor chamber 310 and surrounded by chamber walls 312. The showerhead provides a gas inlet. Within the reactor chamber 310, a stack 314 is mounted on the chuck 308. A gas source 316 is connected to a remote plasma generator 320. The remote plasma generator 320 is connected to the reactor chamber 310 via the showerhead 306. A radio frequency (RF) source 330 provides 13.56 MHz RF power to the remote plasma generator 320. In this embodiment, the RF source 330 provides power to a coil. This power generates an inductively coupled plasma in the remote plasma generator 320. A chuck temperature controller 340 controls a cooler 344. The cooler 344 cools a coolant 348. The coolant is provided to a chuck cooling system 350. Controller 335 is controllably connected to RF source 330 , exhaust pump 352 , chuck temperature controller 340 , and gas source 316 .
[0021] Figure 4335 is a high-level block diagram illustrating a computer system 400 suitable for implementing the controller 335 used in the embodiments. The computer system 400 can have many physical forms ranging from integrated circuits, printed circuit boards, and small handheld devices to large supercomputers. The computer system 400 includes one or more processors 402 and may also include an electronic display device 404 (for displaying graphics, text, and other data), a main memory 406 (e.g., random access memory (RAM)), a storage device 408 (e.g., a hard drive), a removable storage device 410 (e.g., an optical drive), a user interface device 412 (e.g., a keyboard, touch screen, keypad, mouse or other pointing device, etc.), and a communication interface 414 (e.g., a wireless network interface). The communication interface 414 allows software and data to be transmitted between the computer system 400 and external devices via a link. The system may also include a communication infrastructure 416 (e.g., a communication bus, a crossbar, or a network) connected to the above-mentioned devices / modules.
[0022] The information transmitted via the communication interface 414 can be in the form of signals, such as electronic signals, electromagnetic signals, optical signals, or other signals that can be received by the communication interface 414 via a communication link that carries the signals and can be implemented using wires or cables, optical fibers, telephone lines, cellular phone links, radio frequency links, and / or other communication channels. Utilizing such a communication interface 414, it is contemplated that the one or more processors 402 can receive information from a network or output information to a network while performing the above-described method steps. Furthermore, the method embodiments can be executed solely on a processor or can be executed in conjunction with a remote processor that shares a portion of the processing via a network such as the Internet.
[0023] The term "non-transitory computer-readable medium" is generally used to refer to media such as main memory, secondary memory, removable storage devices, and storage devices such as hard disks, flash memory, disk drive memory, CD-ROMs, and other forms of persistent memory, and should not be interpreted as encompassing transient subject matter such as carrier waves or signals. Examples of computer code include machine code (such as produced by a compiler) and files containing higher-level code that are executed by a computer using an interpreter. Computer-readable media can also be computer code transmitted by a computer data signal, a processor.
[0024] Figure 1 is a high level flow diagram of one embodiment. In one exemplary embodiment, the stack is placed in a breakthrough chamber 216 (step 104). The breakthrough chamber 216 is a chamber used to break through or remove layers to provide access to the material to be etched. Figure 5Ais a schematic cross-sectional view of a portion of a stack 314 on a wafer. In this embodiment, the stack 314 includes a first Si layer 508 adjacent to a SiGe layer 512. The SiGe layer 512 is adjacent to a second Si layer 516. The first Si layer 508 and the second Si layer 516 are located on opposite sides of the SiGe layer 512. The first Si layer 508, the SiGe layer 512, and the second Si layer 516 extend between two structures (not shown) and are connected to these structures. Because the stack 314 is exposed to air, a coating layer in the form of a native oxide layer 520 is formed on the sides of the first Si layer 508, the SiGe layer 512, and the second Si layer 516. Alternatively, the stack 314 can be coated with another type of coating layer (e.g., ALD silicon oxide (SiO2), silicon carbide (SiC), or silicon nitride (SiN)). The coating layer can be deposited to prevent the formation of a coating layer formed by native oxidation.
[0025] After the stack 314 has been placed in the breakthrough chamber 216, a breakthrough process is provided to remove the coating layer, i.e., the native oxide layer 520 (step 108). In this example, the breakthrough process is a wet etching process. The breakthrough process is provided by exposing the stack 314 to a dilute hydrofluoric acid (49%) aqueous solution having a volume dilution ratio of 300:1 to 10:1 for 10-300 seconds at room temperature (RT). Figure 5B is a cross-sectional view of stack 314 after native oxide layer 520 has been removed.
[0026] The VTM 212 transports the stack 314 from the breakthrough chamber 216 to the etch chamber 220 in a vacuum environment (step 112). If the stack 314 is not kept in a vacuum environment, a new native oxide layer will grow on the stack 314. The stack 314 is placed on the chuck 308 located in the etch chamber 220.
[0027] The chuck is cooled (step 116). In this example, the chuck is cooled to a temperature of -15°C or lower. In this example, the cooler 344 cools the coolant 348 to a temperature of approximately -60°C. A remote plasma is generated from the etching gas (step 120). In this example, the etching gas is 45 sccm of carbon tetrafluoride (CF4), 1000 sccm of argon (Ar), and 1000 sccm of helium (He). In this example, the etching gas does not contain hydrogen or is substantially free of hydrogen. The etching gas flows from the gas source 316 to the remote plasma generator 320. The RF source 330 provides 200 watts of RF power at 13.56 megahertz (MHz) to the remote plasma generator 320. In the remote plasma generator 320, the etching gas is converted into plasma ions and high-energy neutral particles of the etching gas.
[0028] High-energy neutral particles flow from the remote plasma generator 320 through the showerhead 306 and into the reactor chamber 310. The high-energy neutral particles can be high-energy neutral atoms or high-energy neutral molecules of the etching gas. The high-energy neutral particles can include high-energy free radical molecules or atoms. In this example, plasma ions do not flow into the reactor chamber 310. The reactor chamber pressure is maintained at no less than 300 mTorr. The high-energy neutral particles selectively etch the stack 314 by selectively etching SiGe relative to Si (step 128). In this example, the step of selectively etching SiGe relative to Si has an etching ratio of SiGe to Si greater than 20:1. Figure 5C 1 is a cross-sectional view of stack 314 after SiGe layer 512 has been selectively etched (step 128). In this example, SiGe layer 512 has been completely etched away, leaving behind first Si layer 508 and second Si layer 516. First Si layer 508 and second Si layer 516 may extend between the two structures.
[0029] The VTM 212 transfers the stack 314 from the etch chamber 220 to the ALD chamber 224 in a vacuum environment (step 132). If the stack 314 is not kept in a vacuum environment, a new native oxide layer will grow on the stack 314. This new native oxide layer will consume some of the first Si layer 508 and the second Si layer 516.
[0030] An ALD layer is deposited on the stack 314 (step 136 ). Figure 6 6 is a more detailed flow chart of the atomic layer deposition process (step 136). The atomic layer deposition process (step 136) includes at least one cycle of providing a precursor (step 604) and converting the precursor (step 608). The precursor is provided to the stack 314 (step 604). In this embodiment, the liquid silicon-containing precursor is vaporized and delivered to the ALD chamber 224 in vapor form to saturate the stack 314. Thus, a precursor layer is formed on the stack 314. In this example, the precursor has a composition of the general formula C(x)H(y)N(z)O(a)Si(b). In certain embodiments, the precursor has one of the following compositions: N,N,N',N',N",N"-hexamethylsilanetriamine (C6H 19 N3Si,C8H 22 N2Si), (3-aminopropyl)triethoxysilane (C9H 23 NO3Si), and tetra(isopropoxy)silane (C 12 H 28O4Si). In this example, the precursor is provided plasma-free. The precursor has silicon functional groups. Since the precursor does not attach to another precursor, the precursor forms a single layer on the stack 314.
[0031] Once the stack 314 is dosed with the precursor, the delivery of the precursor vapor is stopped. A purge step is then provided to purge any excess precursor remaining in the ALD chamber 224. The precursor is then converted (step 608). In one embodiment, this is accomplished by subjecting the stack 314 to a flash process. The flash process comprises delivering 1000 sccm to 2000 sccm of oxygen (O2) flash gas to the ALD chamber 224. In this example, a power of 100 to 3000 watts at 13.56 MHz is delivered to convert the flash gas into a plasma. A pressure of 20 mTorr to 100 mTorr is provided. Because the power delivery time is relatively rapid, for example, between about 0.5 seconds and about 4 seconds, this flash process is referred to as an "O2 flash" operation. The O2 flash operation uses a monolayer of a silicon-containing precursor to form a monolayer of silicon oxide on the stack 314. Once the O2 flash operation is complete, the ALD chamber 224 is purged and the cycle can then be repeated. Figure 5D is a cross-sectional view of the stack 314 after the ALD layer 528 has been deposited around or encapsulates the entire perimeter of the Si layers 508, 516. The ALD layer 528 prevents the Si layers 508, 516 from being consumed by native oxide formation. Without the ALD layer 528, the native oxide may consume approximately Arrive at the appointment Si layers 508, 516.
[0032] The resulting coated Si layers 508, 516 can extend between the two structures. Thus, the first Si layer 508 and the second Si layer 516 are used as horizontal nanowires of an n-type metal oxide semiconductor (NMOS) device. This embodiment provides an etch selectivity greater than 20:1 for etching SiGe relative to Si. In addition, less than 1% of the SiGe is lost due to oxidation or etching during this process. This embodiment selectively etches SiGe relative to silicon oxide (SiO2) and silicon nitride (SiN) with a selectivity greater than 100:1. Because this embodiment selectively etches SiGe relative to SiO2 and SiN with a selectivity greater than 100:1, this embodiment has separate breakthrough steps to etch SiO2 or SiN. In one embodiment, the selective etching of SiGe can be performed in less than 60 seconds.
[0033] It has been unexpectedly discovered that maintaining a low temperature while selectively etching SiGe layer 512 relative to Si layers 508, 516 increases selectivity. In one exemplary embodiment, chuck 308 is cooled to a temperature of less than 15°C. In another exemplary embodiment, chuck 308 is cooled to a temperature below 0°C. In another exemplary embodiment, chuck 308 is cooled to a temperature of less than or equal to -15°C. In another exemplary embodiment, chuck 308 is cooled to a temperature of less than or equal to -40°C. In some embodiments, liquid nitrogen is used as the coolant that flows through chuck 308 to provide cooling. In other embodiments, liquid Vertel Siner manufactured by DuPont Corporation of Wilmington, DE TM Can be used as a coolant. This coolant can be cooled to -60°C.
[0034] In various embodiments, the etching gas comprises a fluorine-containing component. In various embodiments, the fluorine-containing component is a fluorocarbon (e.g., CF4, hexafluoro-2-butyne (C4F6), or octafluorocyclobutane (C4F8)), or a hydrofluorocarbon (e.g., fluoroform (CHF3) or difluoromethane (CH2F2)). In various embodiments, the etching gas has a total flow rate per etch gas molecule. The fluorine-containing component comprises fluorine. The fluorine-containing component has a fluorine flow rate per fluorine atom. The ratio of the total flow rate of the etching gas per etch gas molecule to the fluorine flow rate per fluorine atom is from 1000:1 to 3:1. For example, in the above embodiments, the etching gas is 45 sccm of CF4, 1000 sccm of Ar, and 1000 sccm of He. In this example, since there are 4 fluorine atoms in CF4, the flow rate of the fluorine atoms is 180 sccm. The total flow rate of all molecules of the etching gas is 2045 sccm. Therefore, the ratio of the total flow rate of the etching gas to the flow rate of the fluorine atoms is 2045:180. 2045:180 is approximately 11:1. To reduce the percentage of fluorine atoms compared to the total etching gas flow rate, one or more inert gases (e.g., noble gases) are added to the etching gas. To provide a more controllable process, the noble gas may be used as a diluent.
[0035] During etching, the energetic neutral molecules of the etching gas are used to selectively etch the SiGe layer 512 relative to the Si layers 508 and 516. This etching is achieved by generating the energetic neutral molecules via a remote plasma, rather than exposing the stack 314 to the plasma. In the plasma, the stack 314 is bombarded by more ions than the energetic neutral molecules. Reducing the number of ions bombarding the stack 314 helps reduce etching of the Si layers 508 and 516. To further reduce etching of the Si layers 508 and 516, low RF power is used to generate the remote plasma. In various embodiments, the RF power provided to generate the remote plasma is less than 300 watts. Furthermore, in various embodiments, a bias voltage of less than 50 volts is provided. In an exemplary embodiment, no bias voltage is provided. In the absence of a bias voltage, ions are not accelerated toward the stack 314. In an exemplary embodiment, no RF power is provided to the reactor chamber 310 in which the stack 314 is located, and RF power is provided only to the remote plasma generator 320. The remote plasma generator 320 is disposed outside the reactor chamber 310. The remote plasma generator 320 is separated from the reactor chamber 310 by the showerhead 306. A large gap of at least 50 mm exists between the showerhead 306 and the top of the stack 314 to further reduce the number of ions reaching the stack 314. In some embodiments, since the plasma is formed outside the reactor chamber 310, the processing in the reactor chamber 310 is plasma-free.
[0036] The pressure in the reactor chamber 310 is maintained at no less than 300 mTorr. In an exemplary embodiment, the pressure is at least 500 mTorr. Higher pressures promote selective etching of the SiGe layer 512 relative to the Si layers 508, 516. Pressures of less than 100 mTorr will be used for etching processes that rely on ion etching. Conversely, since various embodiments use high-energy neutral particles to perform etching, the pressure is maintained at no less than 300 mTorr. Higher pressures promote etching that uses high-energy neutral particles instead of ions. In addition, higher pressures help reduce the presence of ions and reduce undesirable species. It is believed that various embodiments can provide an etching selectivity ratio of SiGe to Si of at least 50:1.
[0037] The native oxide breakthrough step (step 108) removes the native oxide layer to enable subsequent etching of the SiGe layer 512. Transferring the stack 314 from the breakthrough chamber 216 to the etch chamber 220 under vacuum (step 112) prevents the native oxide layer from regenerating during the transfer. Transferring the stack 314 from the etch chamber 220 to the ALD chamber 224 under vacuum (step 132) prevents native oxide from forming on the Si layers 508, 516. The formation of such a native oxide layer consumes some of the Si layers 508, 516. The consumption of some of the Si layers 508, 516 due to the formation of the native oxide results in some of the Si layers 508, 516 being removed. Deposition of the ALD layer 528 (step 136) forms a layer on the Si layers 508, 516 without consuming silicon from the Si layers 508, 516.
[0038] In another embodiment, the Si layer is selectively etched relative to the SiGe layer. Figure 7A is a schematic cross-sectional view of stack 314. In this embodiment, stack 314 includes a first SiGe layer 708 adjacent to a Si layer 712. Si layer 712 is adjacent to a second SiGe layer 716. First SiGe layer 708 and second SiGe layer 716 are located on opposite sides of Si layer 712. Because stack 314 is exposed to air, a native oxide layer 720 forms on the sides of first SiGe layer 708, Si layer 712, and second SiGe layer 716.
[0039] The stack 314 is placed into the breakthrough chamber 216. A breakthrough treatment is provided to remove the native oxide layer 720 (step 108). In this example, the breakthrough treatment is provided by exposing the stack 314 to a dilute hydrofluoric acid (49%) aqueous solution having a volume dilution ratio of 300:1 to 10:1 for 10 to 300 seconds at room temperature (RT). Figure 7B is a cross-sectional view of stack 314 after native oxide layer 720 has been removed.
[0040] The VTM 212 transfers the stack 314 from the breakthrough chamber 216 to the etch chamber 220 in a vacuum environment (step 112 ). The stack 314 is placed on the chuck 308 located in the etch chamber 220 .
[0041] The chuck 308 is cooled (step 116). In this example, the chuck 308 is cooled to a temperature of no greater than -15°C. In this example, the cooler 344 cools the coolant 348 to a temperature of approximately -60°C. A remote plasma is generated from the etching gas (step 120). In this example, the etching gas is 10 sccm of CF4, 100 sccm of H2, 1000 sccm of Ar, and 1000 sccm of He. A small flow of sulfur hexafluoride (SF6) or hydrogen sulfide (H2S) (0-100 sccm) may be added to improve selectivity. The etching gas flows from the gas source 316 to the remote plasma generator 320. The RF source 330 provides 200 watts of RF power at 13.56 MHz to the remote plasma generator 320. In the remote plasma generator 320, the etching gas is converted into a plasma.
[0042] The high-energy neutral particles of the etching gas flow from the remote plasma generator 320 through the shower head 306 and enter the reactor chamber 310. The high-energy neutral particles can be high-energy neutral atoms or high-energy neutral molecules of the etching gas. In this example, the plasma ions do not flow into the reactor chamber 310. The reactor chamber pressure is maintained at not less than 300mTorr. The high-energy neutral particles selectively etch Si relative to SiGe to selectively etch the stack 314 (step 128). In this example, the step of selectively etching Si relative to SiGe is to have an etching ratio of Si to SiGe greater than 20:1. Figure 7C is a cross-sectional view of the stack 314 after the Si layer 712 has been selectively etched (step 128). In this example, the Si layer 712 has been completely etched away, leaving behind the SiGe layers 708, 716.
[0043] The VTM 212 transfers the stack 314 from the etch chamber 220 to the ALD chamber 224 in a vacuum environment (step 132 ). If the stack 314 is not kept in a vacuum environment, a new native oxide layer will grow on the stack 314 .
[0044] An ALD layer is deposited on the stack 314 (step 136 ). Figure 66 is a more detailed flow chart of the ALD layer deposition process (step 136). The atomic layer deposition process (step 136) includes at least one cycle of providing a precursor (step 604) and converting the precursor (step 608). The precursor is provided to the stack 314 (step 604). In this embodiment, the liquid silicon-containing precursor is vaporized and delivered to the ALD chamber 224 in vapor form to saturate the stack 314. Thus, a precursor layer is formed on the stack 314. In this example, the precursor has a composition of the general formula C(x)H(y)N(z)O(a)Si(b). In certain embodiments, the liquid precursor has one of the following compositions: C6H 19 N3Si、C8H 22 N2Si,C9H 23 NO3Si and C 12 H 28 O4Si. In this example, the precursor is provided without plasma. The precursor has silicon functional groups. Since the precursor does not attach to another precursor, the precursor forms a single layer on the stack 314.
[0045] Once the stack 314 is dosed with the precursor, the delivery of the precursor vapor is stopped. A purge step is then provided to purge any excess precursor remaining in the ALD chamber 224. The precursor is then converted (step 608). In one embodiment, the conversion step is accomplished by subjecting the stack 314 to a flash vaporization process. The flash vaporization process comprises delivering 1000 sccm to 2000 sccm of oxygen (O2) flash gas to the ALD chamber 224. In this example, a power of 100 to 3000 watts at 13.56 MHz is delivered to form the flash gas into a plasma. A pressure of 20 mTorr to 100 mTorr is provided for between about 0.5 seconds and about 4 seconds. Using a monolayer of the silicon-containing precursor, a monolayer of silicon oxide is formed on the stack 314. The ALD chamber 224 is then purged. The cycle can then be repeated. Figure 7D is a cross-sectional view of stack 314 after ALD layer 728 has been deposited.
[0046] The resulting SiGe layers 708, 716 can be used as horizontal nanowires for p-type metal oxide semiconductor (PMOS) devices. This embodiment provides an etch selectivity greater than 20:1 for etching Si relative to SiGe. In addition, because oxidation or etching during this process results in a loss of less than This embodiment selectively etches Si relative to silicon oxide (SiO2) and silicon nitride (SiN) with a selectivity greater than 100:1. This embodiment can provide this selective etching in less than 60 seconds.
[0047] During etching, energetic neutral molecules are used to selectively etch Si layer 712 relative to SiGe layers 708 and 716. This etching is achieved by generating energetic neutral molecules via a remote plasma, rather than exposing stack 314 to a plasma where it would be bombarded by more ions than energetic neutral molecules. Reducing the number of ions bombarding stack 314 helps reduce etching of SiGe layers 708 and 716. To further reduce etching of SiGe layers 708 and 716, low RF power is used to generate the remote plasma. In various embodiments, the RF power provided to generate the remote plasma is less than 300 watts. Furthermore, in various embodiments, a bias voltage of less than 50 volts is provided. One exemplary embodiment does not provide a bias voltage. Therefore, ions are not accelerated toward stack 314. One exemplary embodiment does not provide any RF power to reactor chamber 310 where stack 314 is located, but rather provides RF power only to remote plasma generator 320. A large gap of at least 50 mm exists between the showerhead 306 and the top of the stack 314. This gap further reduces the number of ions reaching the stack 314. In certain embodiments, since the plasma is formed outside of the reactor chamber 310, the processing in the reactor chamber 310 is plasma-free.
[0048] The pressure in the reactor chamber 310 is maintained at not less than 300 mTorr. In an exemplary embodiment, the pressure is at least 500 mTorr. Higher pressures promote selective etching of the Si layer 712 relative to the SiGe layers 708, 716. Pressures of less than 100 mTorr can be used for etching processes that rely on ion etching. Conversely, because various embodiments use high-energy neutral particles to perform etching, the pressure is maintained at not less than 300 mTorr. Higher pressures promote etching using high-energy neutral particles instead of ions. In addition, higher pressures help reduce the presence of ions and reduce undesirable species. It is believed that various embodiments can provide an etching selectivity ratio of Si to SiGe of at least 50:1. In the specification and claims, the phrase "high-energy neutral particles" includes reactive neutral molecules or atoms.
[0049] In other embodiments, the breakthrough process (step 108) may use a vapor etch or dry etch process. In one example of a breakthrough process (step 108) using vapor, hydrogen fluoride (HF) vapor may be used to provide the breakthrough process (step 108). An example of a dry breakthrough process (step 108) may provide a plasma formed from CF4 and a bias voltage of 25 to 50 volts.
[0050] Instead of initially coating the stack 314 with a native oxide layer 520, a coating layer can be formed to prevent silicon consumption during the formation of the native silicon oxide. The coating layer can be SiO2 deposited by ALD, or it can be SiN or SiC. Different breakthrough processes (step 108) can be used for different coating layers. In other embodiments, instead of depositing SiO2 during the deposition of the ALD layer on the stack 314 (step 136), SiN or SiC can be deposited during the deposition of the ALD layer on the stack 314 (step 136). Layers containing SiO2, SiN, or SiC can be deposited in various embodiments.
[0051] In other embodiments, an inert gas can be used to provide inert conditions during the transfer of the stack 314 to the etch chamber 220 (step 112) or the transfer of the stack 314 to the ALD chamber 224 (step 124), rather than using inert conditions provided by a vacuum. Such an inert gas can be Ar, He, or N2. In other embodiments, the etch chamber 220 is equipped with precursors and fast response valves. In such an etch chamber 220, the deposition of the ALD layer 528 on the stack 314 (step 136) can be performed in the etch chamber 220. In such an embodiment, the stack 314 is not transferred to the ALD chamber 224.
[0052] In another embodiment of selectively etching Si relative to SiGe, an etching gas comprising SF6 and H2 can be used. The fluorine from SF6 can combine with hydrogen to form HF, and SF can be used to passivate Ge in the form of Ge-F. The formation of Ge-F helps to further passivate SiGe. In other embodiments, the etching gas may contain fluorocarbons, SF6 and H2S or sulfur-containing gases together with H2. In certain embodiments, CF4 and H2 are used, and the ratio of CF4 to H2 is between 1:1 and 1:1000. The above embodiments have a ratio of SF6 to etching gas of less than 1. The above embodiments have a ratio of H2S to etching gas of less than 1. In certain embodiments, CF4 can be replaced by another hydrofluorocarbon.
[0053] In another embodiment for selectively etching SiGe relative to Si, a wet breakthrough process is provided. Figure 8Ais a schematic cross-sectional view of a portion of stack 314 on a wafer. In this embodiment, stack 314 includes a first Si layer 808 adjacent to a SiGe layer 812. SiGe layer 812 is adjacent to a second Si layer 816. First Si layer 808 and second Si layer 816 are located on opposite sides of SiGe layer 812. As stack 314 is exposed to air, a capping layer in the form of a native oxide layer 820 forms on the sides of first Si layer 808, SiGe layer 812, and second Si layer 816. This breakthrough process is provided, in part, by exposing stack 314 to an aqueous solution of dilute hydrofluoric acid (49%) having a volume dilution ratio of 300:1 to 10:1 at room temperature (RT) for a sufficient time to remove a portion of native oxide layer 820, but not all of the native oxide layer. Figure 8B is a cross-sectional view of stack 314 after native oxide layer 820 has been partially removed.
[0054] A dry etch process using a fluorocarbon breakthrough gas is used to remove the remaining portion of the native oxide layer 820. In this example, the fluorocarbon breakthrough gas comprises CF4. The breakthrough gas is formed into a plasma. The plasma from the breakthrough gas removes the remaining native oxide layer 820 and selectively deposits amorphous carbon relative to the SiGe layer 812 to coat the first Si layer 808 and the second Si layer 816. Figure 8C FIG3 is a cross-sectional view of the stack 314 after an amorphous carbon layer 824 has been selectively deposited on the first Si layer 808 and the second Si layer 816. Etching and ALD processes used in other embodiments can be used to selectively etch the SiGe layer 812 relative to the first Si layer 808 and the second Si layer 816. The amorphous carbon layer 824 prevents oxidation of the first Si layer 808 and the second Si layer 816 and reduces etching of the first Si layer 808 and the second Si layer 816. By partially etching the native oxide layer 820 using wet etching, the time required for dry etching is reduced. Reducing the dry etching time reduces the time the substrate is exposed to ions, thereby reducing etching of the first Si layer 808 and the second Si layer 816 by dry etching.
[0055] Although the present invention has been described in terms of several exemplary embodiments, there are changes, modifications, permutations, and various alternative equivalents that fall within the scope of the present invention. It should also be noted that there are many alternative ways to implement the methods and apparatus of the present disclosure. Therefore, the following appended claims are intended to be interpreted as including all such changes, modifications, permutations, and various alternative equivalents that fall within the true spirit and scope of the present disclosure.
Claims
1. A method of forming silicon nanowires by selectively etching silicon germanium relative to silicon in a stack on a chuck in an etch chamber, wherein the stack comprises a first silicon layer, a second silicon layer, and a silicon germanium layer separating the first silicon layer and the second silicon layer, the method comprising: maintaining the chuck at a temperature of 0° C. or less to increase the selectivity of etching the silicon germanium relative to the silicon; exposing the stack to an etching gas including a fluorine-containing gas to selectively etch silicon germanium relative to silicon, wherein the silicon germanium layer is removed from between the first silicon layer and the second silicon layer to form a gap between the first silicon layer and the second silicon layer to form the first silicon layer and the second silicon layer into silicon nanowires; as well as After exposing the stack to the etching gas in the etching chamber, a layer containing SiO 2 , SiN, or SiC is deposited on the stack using atomic layer deposition, wherein the layer containing SiO 2 , SiN, or SiC surrounds the silicon nanowires.
2. The method according to claim 1 , wherein the etching gas has a total flow rate calculated as etching gas molecules, and wherein the fluorine-containing gas comprises fluorine, wherein the fluorine has a fluorine flow rate calculated as fluorine atoms, and wherein a ratio of the total flow rate calculated as etching gas molecules to the fluorine flow rate calculated as fluorine atoms is between 1000:1 and 3:
1.
3. The method according to claim 1, further comprising: In a remote plasma generator, forming the etching gas into a plasma having ions of the etching gas and high-energy neutral particles; and The energetic neutral particles are flowed from the remote plasma generator into the etch chamber. 4 . The method of claim 3 , wherein the etching gas is not maintained as a plasma in the etching chamber to expose the stack to the energetic neutral particles. The method according to claim 1 , wherein the fluorine-containing gas comprises a hydrofluorocarbon or a fluorocarbon. The method according to claim 5 , wherein the etching gas further comprises a rare gas. The method of claim 1 , wherein the fluorine-containing gas comprises CF 4 .
8. The method of claim 1, further comprising maintaining an etch chamber pressure of at least 300 mTorr.
9. The method of claim 3, wherein flowing the energetic neutral particles from the remote plasma generator into the etch chamber to form silicon nanowires comprises: The stack is transferred from the etching chamber to an atomic layer deposition chamber under inert conditions.
10. The method of claim 1 , wherein the stack further comprises an oxide capping layer, the method further comprising: The cladding layer is removed from the stack using a breakthrough process. The method of claim 10 , wherein the breakthrough process comprises providing a wet etch of the cladding layer. 12 . The method of claim 11 , wherein the breakthrough process further comprises performing dry etching after the wet etching.
13. The method of claim 12, wherein the dry etch of the breakthrough process deposits a carbon-containing layer selectively on silicon relative to silicon germanium.
14. The method of claim 13, wherein the dry etching of the breakthrough process comprises: providing CF4-containing gas; and The CF 4 -containing gas is formed into a plasma, wherein the plasma provides a breakthrough etch and deposits the carbon-containing layer selectively on silicon relative to silicon germanium.
15. The method of claim 10, wherein the removing comprises providing a vapor etch or a dry etch of the cladding layer.
16. The method of claim 1, wherein the maintaining the chuck at a temperature of 0°C or less is maintaining the chuck at a temperature of -15°C or less.
17. A method of producing silicon germanium nanowires by selectively etching silicon relative to silicon germanium in a stack on a chuck in an etch chamber, wherein the stack comprises a first silicon germanium layer, a second silicon germanium layer, and a silicon layer separating the first and second silicon germanium layers, the method comprising: maintaining the chuck at a temperature of 0° C. or less to increase the selectivity of etching the silicon relative to the silicon germanium; exposing the stack to an etching gas comprising H2 and a fluorine-containing gas to selectively etch silicon relative to silicon germanium to form silicon germanium nanowires; as well as After exposing the stack to the etching gas in the etching chamber, a layer containing SiO 2 , SiN, or SiC is deposited on the stack using atomic layer deposition, wherein the layer containing SiO 2 , SiN, or SiC surrounds the silicon germanium nanowires.
18. The method according to claim 17, further comprising: In a remote plasma generator, forming the etching gas into a plasma having ions of the etching gas and high-energy neutral particles; and The energetic neutral particles are flowed from the remote plasma generator into the etch chamber.
19. The method of claim 18, wherein the etching gas is not maintained as a plasma in the etching chamber to expose the stack to the energetic neutral particles.
20. The method of claim 17, further comprising maintaining an etch chamber pressure of at least 300 mTorr.
21. The method of claim 17, wherein the etching gas further comprises at least one of H2S and SF6.
22. The method of claim 17, wherein the stack further comprises an oxide capping layer, the method further comprising: A breakthrough process is provided to remove the oxide capping layer from the stack.
23. The method of claim 22, wherein providing a breakthrough process comprises providing a wet etch, a vapor etch, or a dry etch of the oxide capping layer.
24. The method of claim 17, wherein the fluorine-containing gas comprises CF4.
25. The method of claim 17, wherein said maintaining the chuck at a temperature of 0°C or less is maintaining the chuck at a temperature of -15°C or less.
Citation Information
Patent Citations
Dry etching method
JP2013225604A
Silicon germanium mask for deep silicon etching
US20120225557A1
Silicon germanium processing
US20150126040A1