Method for preparing a device for measuring mechanical loss of epitaxial single crystal thin film and method for measuring mechanical loss of epitaxial single crystal thin film at low temperature

Abstract

This invention provides a method for fabricating a device for measuring the mechanical loss of epitaxial single-crystal thin films and a method for measuring the mechanical loss of epitaxial single-crystal thin films at low temperature, belonging to the field of semiconductor micro-nano fabrication technology. The fabrication method includes: growing an epitaxial single-crystal thin film on the front side of an epitaxial substrate using molecular beam epitaxy to obtain an epitaxial substrate with the epitaxial single-crystal thin film on the front side; performing a coating process on the back side of the epitaxial substrate to compensate for stress on the warpage of the epitaxial single-crystal thin film; directly bonding the surface of the epitaxial single-crystal thin film away from the epitaxial substrate to a cantilever substrate to form a bonding pair, obtaining a composite material; removing the stress compensation film and the epitaxial substrate from the side of the epitaxial single-crystal thin film away from the cantilever substrate in the composite material to obtain the device; the cantilever substrate is made of single-crystal silicon, the epitaxial substrate is a group III-V semiconductor single-crystal substrate, and the epitaxial single-crystal thin film is a group III-V semiconductor single-crystal thin film.

Description

Technical Field

[0001] This invention relates to the field of semiconductor micro-nano fabrication technology, and in particular to a method for fabricating a device for measuring the mechanical loss of epitaxial single-crystal thin films and a method for measuring the mechanical loss of epitaxial single-crystal thin films at low temperature. Background Technology

[0002] The cantilever damped oscillation method is one of the commonly used methods for measuring the mechanical loss of thin films. Its working principle is to make the cantilever beam attached to the film vibrate, cut off the excitation source, and accurately measure the natural decay rate of the vibration amplitude. By comparing the change in decay rate before and after the film is attached, the mechanical loss of the film itself can be calculated.

[0003] Currently, the common method for fabricating cantilever beams involves creating a mask on a silicon wafer using photolithography, followed by wet etching to form the cantilever shape. These cantilever beams are then cut into discrete components that can be individually mounted onto test fixtures. Finally, a film is deposited on the silicon cantilever using physical vapor deposition. However, this method is difficult to apply to substrate-transferred crystal films, making it challenging to test the mechanical loss of the film using the cantilever damping oscillation method. Summary of the Invention

[0004] In view of this, in order to at least partially solve the aforementioned technical problems, the present invention provides a method for fabricating a device for measuring the mechanical loss of epitaxial single crystal thin films and a method for measuring the mechanical loss of epitaxial single crystal thin films at low temperature.

[0005] According to one aspect of the present invention, a method for fabricating a device for measuring the mechanical loss of an epitaxial single-crystal thin film is provided, comprising: growing an epitaxial single-crystal thin film on the front side of an epitaxial substrate by molecular beam epitaxy to obtain an epitaxial substrate having an epitaxial single-crystal thin film on the front side, the epitaxial substrate including a front side and a back side; performing a coating process on the back side of the epitaxial substrate to compensate for stress on the warpage of the epitaxial single-crystal thin film, to obtain an epitaxial substrate having an epitaxial single-crystal thin film on the front side and a stress compensation film on the back side; forming a bonding pair by directly bonding the surface of the epitaxial single-crystal thin film away from the epitaxial substrate to a cantilever substrate to obtain a composite material; removing the stress compensation film and the epitaxial substrate on the side of the epitaxial single-crystal thin film away from the cantilever substrate from the composite material to obtain the device; wherein the material of the cantilever substrate includes single-crystal silicon, the epitaxial substrate includes a group III-V semiconductor single-crystal substrate, and the epitaxial single-crystal thin film is a group III-V semiconductor single-crystal thin film.

[0006] In some embodiments, a sacrificial layer is also included between the epitaxial single crystal thin film and the epitaxial substrate.

[0007] In some embodiments, the epitaxial substrate comprises gallium arsenide.

[0008] In some embodiments, the epitaxial single-crystal thin film includes either a single-layer film or a multilayer film.

[0009] In some implementations, the stress-compensating film is silicon nitride.

[0010] In some embodiments, the coating process is performed on the back side of the epitaxial substrate by ion-assisted reactive magnetron sputtering.

[0011] In some embodiments, the deposition process on the back side of the epitaxial substrate by ion-assisted reactive magnetron sputtering includes: using argon as the working gas and nitrogen as the reactive gas, using a silicon target, to deposit a film on the back side of the epitaxial substrate by reactive magnetron sputtering, and during the deposition process, activating an RF bias to bombard the back side of the epitaxial substrate to deposit a stress compensation film on the back side of the epitaxial substrate.

[0012] In some embodiments, the vacuum degree of the coating process is less than 1E-5 Torr, the ratio of nitrogen flow rate to argon flow rate is 1:(1~10), the power supply of the silicon target is 250~350W, and the power of the radio frequency bias voltage is 80~120W.

[0013] In some implementations, the warpage of the stress-compensated epitaxial single-crystal film is less than 5 μm.

[0014] In some embodiments, the conditions for direct bonding include: bombarding the surface of the epitaxial single crystal film away from the epitaxial substrate and the surface of the cantilever substrate with oxygen ions, performing pressure bonding under a pressure of 5 to 20 N, and then performing annealing.

[0015] In some embodiments, the cantilever substrate is prepared by: depositing mask layers on two opposite surfaces of a single-crystal silicon substrate to obtain a single-crystal silicon substrate with mask layers on both surfaces; performing patterning, resist removal, and cleaning sequentially on the mask layer of either of the two surfaces to obtain a single-surface patterned single-crystal silicon substrate; performing wet etching on the patterned surface side of the single-surface patterned single-crystal silicon substrate, and cleaning to obtain the cantilever substrate.

[0016] In some embodiments, between depositing the mask layer and performing the patterning process, the method further includes: cleaning the monocrystalline silicon substrate with mask layers on both surfaces using a piranha solution, and immersing the cleaned monocrystalline silicon substrate with mask layers on both surfaces in a buffer oxide etching solution; in the piranha solution, the volume ratio of sulfuric acid to hydrogen peroxide is (3~10):1; in the buffer oxide etching solution, the volume ratio of ammonium fluoride solution to hydrofluoric acid is (3~10):1.

[0017] According to another aspect of the present invention, a method for measuring the mechanical loss of epitaxial single-crystal thin films at low temperature is provided, wherein the device is prepared by the fabrication method of the device for measuring the mechanical loss of epitaxial single-crystal thin films as described above.

[0018] Based on the above technical solutions, the method for fabricating a device for measuring the mechanical loss of epitaxial single-crystal thin films and the method for measuring the mechanical loss of epitaxial single-crystal thin films at low temperature provided by the present invention have at least the following beneficial effects:

[0019] High-quality epitaxial single-crystal thin films with lattice matching are first grown on homogeneous substrates. Due to the internal stress of the epitaxial single-crystal thin film, the wafer warps. By depositing a stress compensation film on the back side of the epitaxial substrate, the stress on the front side of the epitaxial substrate is balanced, flattening the warped wafer and ensuring the successful subsequent direct bonding. Through direct bonding, the front side of the epitaxial single-crystal thin film is completely transferred to the single-crystal silicon cantilever substrate. Then, excess epitaxial substrate and stress compensation film are removed from the back side. Based on the extremely low mechanical loss of the single-crystal silicon cantilever, it is used as a carrier so that the measured mechanical loss signal mainly originates from the epitaxial single-crystal thin film. This allows for the fabrication of an ideal device for measuring the mechanical loss of epitaxial single-crystal thin films using the cantilever decay oscillation method, thereby achieving accurate measurement of the intrinsic mechanical loss of epitaxial single-crystal thin films. Attached Figure Description

[0020] The above and other objects, features and advantages of the present invention will become clearer from the following description of embodiments of the invention with reference to the accompanying drawings.

[0021] Figure 1 A flowchart illustrating a method for fabricating a device for measuring the mechanical loss of an epitaxial single-crystal thin film according to an embodiment of the present invention is shown.

[0022] Figure 2 A process flow diagram of the fabrication of a device for measuring the mechanical loss of epitaxial single crystal thin films according to an embodiment of the present invention is shown;

[0023] Figure 3 A schematic diagram of the process for measuring the mechanical loss of an epitaxial single-crystal thin film according to an embodiment of the present invention is shown.

[0024] Figure 4 A schematic diagram of the structure of a device for measuring the mechanical loss of epitaxial single crystal thin films according to an embodiment of the present invention is shown;

[0025] Figure 5 The image shows a roughness measurement diagram of the etched surface of a silicon cantilever wafer after wet etching according to Embodiment 1 of the present invention.

[0026] Figure 6a An optical microscope image of a gallium arsenide substrate without pre-cleaning and directly coated according to Embodiment 1 of the present invention is shown.

[0027] Figure 6b An optical microscope image of a gallium arsenide substrate surface after pre-cleaning and then coating according to Embodiment 1 of the present invention is shown.

[0028] Figure 7a The diagram shows the warpage measurement of the 4-inch aluminum gallium arsenide / gallium arsenide Bragg reflector epitaxial wafer according to Embodiment 1 of the present invention;

[0029] Figure 7b The diagram shows a warpage measurement of a 4-inch gallium arsenide substrate used for confirming stress-compensated coating process parameters in Embodiment 1 of the present invention.

[0030] Figure 7c The diagram shows the warpage measurement of a 4-inch aluminum gallium arsenide / gallium arsenide Bragg reflector epitaxial wafer after stress compensation according to Embodiment 1 of the present invention. Detailed Implementation

[0031] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the invention. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of the invention for ease of explanation. However, it will be apparent that one or more embodiments may be practiced without these specific details. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concept of the invention.

[0032] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the invention. The term "comprising" as used herein indicates the presence of features, steps, or operations, but does not exclude the presence or addition of one or more other features.

[0033] Group III-V semiconductor single-crystal thin films are emerging ultra-high reflectivity thin-film materials with wide applications in quantum precision measurement fields such as atomic optical lattice clocks and laser interferometry gravitational wave measurement. In these applications, the mechanical loss of Group III-V semiconductor single-crystal thin films is a crucial parameter. Cutting-edge research in precision measurement requires understanding the mechanical loss level of Group III-V semiconductor single-crystal thin films at low temperatures, which can improve the performance of devices integrated from these single-crystal thin films.

[0034] Because single-crystal silicon cantilever substrates have lower mechanical losses, integrating the thin film whose mechanical loss needs to be measured onto a single-crystal silicon cantilever substrate is more conducive to accurate measurement of the film's mechanical loss. However, the lattice constant difference between group III-V semiconductor single-crystal thin films and single-crystal silicon is significant, making it difficult to directly grow group III-V semiconductor single-crystal thin films on single-crystal silicon substrates, and thus making it difficult to find a method for accurately measuring the mechanical loss of group III-V semiconductor single-crystal thin films.

[0035] In the process of realizing the concept of this invention, it was found that, for the sake of lattice adaptation, epitaxial single crystal films are grown on epitaxial substrates of the same family, which results in high quality epitaxial single crystal films. However, there is stress inside the grown epitaxial single crystal film, which causes wafer warping and makes it difficult to directly bond with cantilever substrates.

[0036] Furthermore, in order to solve the above-mentioned problems, a stress compensation film was deposited on the back side of the epitaxial substrate to balance the stress on the front side, ensuring that the subsequent epitaxial single crystal film could be directly bonded to the cantilever substrate, thereby preparing an epitaxial single crystal film bonded to the surface of single crystal silicon. This facilitates the subsequent more accurate measurement of the mechanical loss of the epitaxial single crystal film using the cantilever decay oscillation method.

[0037] Specifically, according to one embodiment of the present invention, a method for fabricating a device for measuring the mechanical loss of epitaxial single-crystal thin films is provided. Figure 1 A flowchart illustrating a method for fabricating a device for measuring the mechanical loss of an epitaxial single-crystal thin film according to an embodiment of the present invention is shown. Figure 1 As shown, the preparation method includes operations S101 to S104.

[0038] In operation S101, an epitaxial single-crystal thin film is grown on the front side of the epitaxial substrate by molecular beam epitaxy, resulting in an epitaxial substrate with an epitaxial single-crystal thin film on the front side. The epitaxial substrate includes a front side and a back side.

[0039] In operation S102, a coating process is performed on the back side of the epitaxial substrate to compensate for the stress of the warping of the epitaxial single crystal film, resulting in an epitaxial substrate with an epitaxial single crystal film on the front side and a stress compensation film on the back side.

[0040] In operation S103, the surface of the epitaxial single crystal thin film away from the epitaxial substrate is directly bonded to the cantilever substrate to form a bonding pair, thus obtaining a composite material.

[0041] In operation S104, the stress compensation film and epitaxial substrate on the side of the epitaxial single crystal film away from the cantilever substrate in the composite material are removed to obtain the device.

[0042] According to embodiments of the present invention, the cantilever substrate is made of single-crystal silicon, the epitaxial substrate is a group III-V semiconductor single-crystal substrate, and the epitaxial single-crystal thin film is a group III-V semiconductor single-crystal thin film. By growing the epitaxial single-crystal thin film on the epitaxial substrate and then transferring it to the cantilever substrate, the epitaxial single-crystal thin film can be successfully deposited on single-crystal silicon. For example, single-crystal gallium arsenide / aluminum gallium arsenide Bragg mirrors can be successfully deposited, thereby achieving ultra-low Brownian thermal noise of ultra-high reflectivity Bragg mirrors based on lower mechanical losses. Subsequently, direct bonding, such as surface-activated bonding or plasma-activated bonding, is used so that no additional intermediate layer is introduced during the bonding process. This is suitable for the construction of optical and electronic devices that are sensitive to the aforementioned intermediate layer, such as Bragg mirrors. Furthermore, the direct bonding temperature is relatively low (e.g., below 200°C), which is more favorable for compound semiconductors such as group III-V semiconductor single-crystal thin films that may decompose easily at high temperatures. However, direct bonding requires high wafer flatness. This invention reduces the warpage of the epitaxial single crystal film by compensating for the back stress of the epitaxial substrate, enabling the epitaxial single crystal film to be successfully directly bonded to the cantilever substrate. This helps to accurately detect the mechanical loss of the epitaxial single crystal film and apply it to quantum precision measurement fields such as atomic lattice clocks and laser interferometry gravitational wave measurement.

[0043] In some embodiments, a sacrificial layer is also included between the epitaxial single-crystal film and the epitaxial substrate. It is understood that after directly bonding the epitaxial single-crystal film to the cantilever substrate, it is necessary to remove the epitaxial substrate and the stress compensation film. Since the epitaxial substrate and the epitaxial single-crystal film are often of the same type of material or have similar chemical properties, directly removing the epitaxial substrate using methods such as physical thinning or wet etching, especially wet etching, can easily damage the high-quality epitaxial single-crystal film. Therefore, a sacrificial layer of a certain thickness (e.g., 100-1000 nm) can be grown before growing the epitaxial single-crystal film. When the direct bonding is completed and the epitaxial substrate needs to be removed, most of the substrate can be removed first by physical thinning or chemical etching. Then, selective etching can be performed using a chemical solution that can only etch the sacrificial layer, without etching the epitaxial single-crystal film and the cantilever substrate, thus completely preserving the high-quality epitaxial single-crystal film on the cantilever substrate.

[0044] In some embodiments, the epitaxial substrate includes gallium arsenide (GaAs). The epitaxial substrate and the grown Group III-V semiconductor single-crystal thin film (e.g., indium gallium arsenide, aluminum gallium arsenide) belong to the same group of materials and have the same or similar lattice constants. This facilitates obtaining a single-crystal thin film with low defect density and high crystal integrity during molecular beam epitaxy. When using GaAs as the epitaxial substrate, aluminum arsenide or aluminum gallium arsenide can be grown as a sacrificial layer, thereby enabling precise and non-destructive separation of the epitaxial substrate from the epitaxial single-crystal thin film.

[0045] In some embodiments, the epitaxial single-crystal thin film includes either a single-layer film or a multilayer film. A single-layer film can serve as a standard sample to measure the fundamental intrinsic mechanical loss of the material, facilitating the study of the defect relaxation behavior of a single material. Multilayer films are often used in the aforementioned Bragg reflectors and other thin films, allowing the calculation of the mechanical loss of one material from the intrinsic mechanical loss of another. Furthermore, as the actual functional structure between quantum precision measuring devices, the direct and accurate measurement of the mechanical loss of multilayer films is more conducive to improving the performance of precision devices, thus leading to their widespread application in quantum precision measurement fields such as atomic optical lattice clocks and laser interferometry gravitational wave detection.

[0046] Optionally, the epitaxial single crystal thin film can be, for example, a gallium arsenide single-layer film, or a stacked film of gallium arsenide and aluminum gallium arsenide. The present invention does not particularly limit this, as long as the lattice between the epitaxial film and the epitaxial substrate is compatible.

[0047] It is understood that a sacrificial layer may or may not be present between the epitaxial single-crystal thin film and the epitaxial substrate. For example, when the chemical composition of the epitaxial substrate is completely identical to that of the epitaxial single-crystal thin film, a sacrificial layer is preferably included to avoid damage to the epitaxial single-crystal thin film during chemical etching. When the epitaxial single-crystal thin film is a multilayer film and the composition of the film layers adjacent to the epitaxial substrate is different from that of the epitaxial substrate, or when the chemical compositions of the epitaxial single-crystal thin film and the epitaxial substrate are inconsistent, a sacrificial layer may not be present between the epitaxial single-crystal thin film and the epitaxial substrate.

[0048] In some embodiments, the stress compensation film is silicon nitride. As mentioned above, direct bonding requires high flatness of both wafers, while the inherent stress of silicon nitride can be precisely controlled within a wide range by adjusting the fabrication process, adapting to both high tensile stress and high compressive stress. The tensile or compressive stress provided by silicon nitride flattens the epitaxial single-crystal film, reducing its warpage and providing a flatness basis for successful subsequent direct bonding.

[0049] In some embodiments, the coating process is performed on the back side of the epitaxial substrate using ion-assisted reactive magnetron sputtering. During reactive magnetron sputtering, by adding an auxiliary ion source, such as an RF bias, to bombard the back side of the epitaxial substrate, the film materials (epitaxygen substrate and stress compensation film) can be smoothed and compacted simultaneously with the coating process. This results in a high-stress film layer, which is more conducive to smoothing out the warpage of the epitaxial single-crystal thin film through stress compensation by the stress compensation film. Furthermore, the introduction of the auxiliary ion source allows for pre-cleaning of the back side of the epitaxial substrate before coating, improving the adhesion of the stress compensation film to the back side of the epitaxial substrate and avoiding the risk of the high-stress stress compensation film detaching from low-adhesion substrates such as gallium arsenide. Additionally, reactive magnetron sputtering can deposit high-stress silicon nitride at relatively low temperatures (≤100°C), avoiding the risk of decomposition of the epitaxial single-crystal thin film (e.g., gallium arsenide) above 400°C, making the process more process-friendly.

[0050] In some further embodiments, the deposition process on the back side of the epitaxial substrate using ion-assisted reactive magnetron sputtering includes: using argon as the working gas and nitrogen as the reactive gas, and employing a silicon target, to deposit a stress-compensating film on the back side of the epitaxial substrate via reactive magnetron sputtering. During the deposition process, an radio frequency bias is activated to bombard the back side of the epitaxial substrate, thereby depositing a stress-compensating film on the back side. By introducing an radio frequency bias, the continuous bombardment of the growing silicon nitride film surface with high-energy argon ions can make the deposited silicon nitride film denser, thereby altering the microstructure of the silicon nitride film and allowing for continuous and precise adjustment of the stress of the silicon nitride film over a wide range. The temperature of the epitaxial substrate is maintained at a relatively low range during reactive magnetron sputtering, and combined with the energy provided by the radio frequency bias, high-quality, high-stress silicon nitride deposition on the back side can be achieved without damaging the epitaxial single-crystal film on the front side.

[0051] Figure 2 A process flow diagram illustrating the fabrication of a device for measuring the mechanical loss of epitaxial single-crystal thin films, according to an embodiment of the present invention, is shown. Figure 2 As shown on the right, the power of the RF bias and the power of the silicon target can be adjusted based on the warpage of the epitaxial single-crystal film to be tested. For example, the warpage of the epitaxial substrate with the epitaxial single-crystal film on the front side can be detected first. Then, the coating process and thickness of the stress compensation film are designed according to the warpage of the epitaxial single-crystal film to be bonded. Then, ion-assisted reactive magnetron sputtering is performed according to the designed process, and the thickness of the stress compensation film is gradually increased so that the warpage of the epitaxial substrate with the epitaxial single-crystal film on the front side is optimized to a level that allows direct bonding.

[0052] It should be noted that warpage detection methods include, but are not limited to, using 3D wafer surface scanning measurement equipment, laser interferometers, white light interferometers, etc.

[0053] It is understandable that direct surface activation bonding or plasma activation bonding of 4-inch wafers requires a total thickness variation (TTV) of less than 3 μm and a warpage (BOW) of less than 5 μm. Directly growing epitaxial single-crystal thin films on epitaxial substrates increases the warpage of the substrate. Continuing with the example of a gallium arsenide (GaAs) and aluminum gallium arsenide (AlGaAs) stacked film as a Bragg mirror, the AlGaAs / GaAs Bragg mirror structure grown on the GaAs substrate introduces compressive stress during the substrate transfer single-crystal film process, leading to increased wafer warpage and bonding surface bulges. The BOW on a 4-inch wafer can reach 10 μm to 70 μm. Heat treatment is a common process for addressing film stress, but its effectiveness is limited on molecular beam epitaxy (MBE) films. This is because MBE growth temperatures can reach 700℃~900℃, but heat treatment equipment cannot reach such high temperatures, otherwise gallium arsenide (GaAs) and aluminum gallium arsenide (AlGaAs) would be damaged. Furthermore, AlGaAs and GaAs themselves have lattice mismatches, and the stress generated by these mismatches cannot be eliminated by heat treatment. For bonding highly warped wafers, traditional solutions typically involve applying significant pressure to the wafer during bonding to flatten the warped wafer and achieve bonding. However, for fragile epitaxial substrates like gallium arsenide, excessive pressure can cause the epitaxial substrate to fracture. Additionally, if the wafer contains suspended structures, such as silicon cantilever arms with a thickness of 60μm~80μm used for mechanical loss measurement, excessive pressure can also cause the device structure to fracture.

[0054] Taking a gallium arsenide and aluminum gallium arsenide stacked film as a Bragg reflector as an example, the vacuum degree of the coating process can be lower than 1E-5 Torr, preferably 5E-6 Torr. The ratio of nitrogen flow rate to argon flow rate is 1:(1~10), for example, it can be 1:1, 1:2, 1:4, 1:6, 1:8 or 1:10, or any range between the above two values. The power supply of the silicon target is 250~350W, for example, 250W, 300W or 350W; the power of the RF bias voltage is 80~120W, for example, 80W, 100W or 120W. This setting allows the formed epitaxial single crystal film to be successfully bonded directly to the cantilever substrate after stress compensation.

[0055] This application achieves high adhesion deposition of a back-side stress-compensating film using ion-assisted reactive magnetron sputtering at a lower temperature, resulting in a warpage of less than 5 μm in the stress-compensated epitaxial single-crystal film. Furthermore, it avoids damage to the epitaxial single-crystal film caused by high-temperature processes while meeting the requirements for direct bonding.

[0056] In some embodiments, the direct bonding conditions include: bombarding the surface of the epitaxial single-crystal film away from the epitaxial substrate and the surface of the cantilever substrate with oxygen ions, performing pressure bonding under a pressure of 5-20 N, and then performing annealing. This setup allows for the formation of a bond pair between the cantilever substrate and the epitaxial single-crystal film through intermolecular forces and covalent bonds by applying a relatively low bonding pressure. This enables direct bonding of the high-warpage substrate under mild pressure, thereby avoiding bonding failure and the fragmentation of the epitaxial substrate or the epitaxial single-crystal film.

[0057] Optionally, the pressure can be, for example, 5N, 10N, 15N, or 20N.

[0058] Figure 3 A schematic diagram of the fabrication process for a device used to measure the mechanical loss of an epitaxial single-crystal thin film, according to an embodiment of the present invention, is shown. Figure 2 and Figure 3 As shown, the cantilever substrate is prepared as follows: Mask layers are deposited on two opposite surfaces of a single-crystal silicon substrate, resulting in single-crystal silicon substrates with mask layers on both surfaces. Patterning, resist removal, and cleaning are sequentially performed on the mask layer of either surface to obtain a single-surface patterned single-crystal silicon substrate. Wet etching is performed on the patterned surface side of the single-surface patterned single-crystal silicon substrate, followed by cleaning to obtain the cantilever substrate. The mask layers deposited on both surfaces of the single-crystal silicon substrate provide protection. One surface of the single-crystal silicon substrate is patterned using photolithography and resist removal, exposing the area to be etched. Wet etching can be performed using, for example, potassium hydroxide or tetramethylammonium hydroxide solution.

[0059] In some implementations, such as Figure 2 As shown on the left, between depositing the mask layer and performing patterning, the process includes: cleaning the single-crystal silicon substrate with mask layers on both surfaces using a piranha solution, and then immersing the cleaned single-crystal silicon substrate with mask layers on both surfaces in a buffer oxide etching solution. This cleaning method prevents the formation of the etching mask, and the circulating filtration of the liquid during wet etching achieves superior surface roughness of the cantilever substrate, thereby enabling more accurate measurement of thin film mechanical losses.

[0060] In some specific embodiments, the volume ratio of sulfuric acid to hydrogen peroxide in the piranha solution is (3~10):1; and the volume ratio of ammonium fluoride solution to hydrofluoric acid in the buffer oxide etching solution (BOE) is (3~10):1.

[0061] More specifically, sulfuric acid can be a sulfuric acid solution with a mass percentage of 98%, hydrogen peroxide can be a mass percentage of 30%, ammonium fluoride solution can be a mass percentage of 40%, and hydrofluoric acid can be a hydrofluoric acid solution with a mass percentage of 40%.

[0062] Understandable, such as Figures 2-3 As shown, this invention combines the fabrication process of cantilever substrates with the process of epitaxial single-crystal thin films, thereby fabricating devices suitable for low-temperature measurements. It overcomes the limitations of traditional methods, such as the excessive intrinsic mechanical losses of cantilevers leading to poor signal-to-noise ratio in the measurement of mechanical losses in Group III-V semiconductor single-crystal thin films, the physicochemical properties of Group III-V semiconductor single-crystal thin films making semiconductor processes with temperatures exceeding 400°C unusable (which would cause thermal decomposition of the Group III-V semiconductor single-crystal thin films), difficulties such as low-pressure chemical vapor deposition on gallium arsenide epitaxial substrates, the difficulty of direct bonding due to warping of epitaxial wafers (i.e., epitaxial single-crystal thin films grown on epitaxial substrates), and the difficulty of using direct bonding methods for thin, flexible, and fragile cantilevers.

[0063] The following is based on Figures 2-3 The fabrication process of the device will be explained in detail.

[0064] Silicon nitride is deposited on both surfaces of a double-polished intrinsic single-crystal silicon substrate using methods including but not limited to low-pressure chemical vapor deposition (LPCVD) or ion-assisted reactive magnetron sputtering, as a mask layer for subsequent wet etching.

[0065] Patterning is defined on a mask layer on one of the surfaces using photolithography and reactive ion etching, followed by resist removal and cleaning.

[0066] Use piranha solution to completely remove residual organic matter and photoresist, and finally soak in BOE solution to remove the oxide layer produced by the piranha solution for 3 to 10 minutes.

[0067] Wet etching is performed using a potassium hydroxide solution with a mass concentration of 30-40% and a temperature of 60-80°C. The liquid is circulated and filtered during wet etching. After etching the unprotected patterned areas to 60-80 μm, the etching mask is removed using hydrofluoric acid, and cleaning is completed.

[0068] Warpage of epitaxial wafers with grown epitaxial single-crystal thin films is determined by warpage testing.

[0069] Based on the warpage, determine the coating process and coating thickness of the stress compensation film used for epitaxial single crystal thin film warpage compensation.

[0070] In order to enable the silicon nitride stress compensation film to adhere firmly to epitaxial substrates with low adhesion such as gallium arsenide, the following coating process was developed based on ion-assisted reactive magnetron sputtering.

[0071] After placing the epitaxial substrate into the coating chamber, the chamber is evacuated to a vacuum level below 1E-5 Torr. Argon gas is then introduced into the vacuum chamber at a flow rate of 30-50 sccm. The radio frequency bias is then activated with a bias power of 120W. The surface of the epitaxial substrate is then pre-cleaned for 60 seconds.

[0072] Nitrogen and argon are introduced into the vacuum chamber, with a nitrogen flow rate of 5-30 sccm and an argon flow rate of 30-50 sccm. Sputtering is started on the silicon target, which uses a pulsed DC power supply with a power of 300W and a duty cycle of 80%. During the coating process, an RF bias is used to bombard the surface of the epitaxial substrate simultaneously with an RF power of 30-150W.

[0073] The coating thickness of the stress compensation film should be designed based on the warpage of the wafer to be bonded. Feasible methods include, but are not limited to, the following:

[0074] Take a bare substrate from the same batch as the wafer to be bonded, and perform ion-assisted reactive magnetron sputtering on the process surface of the bare substrate according to the aforementioned process flow. The final coating thickness should cause the bare substrate to produce the same warpage as the wafer to be bonded. This coating thickness meets the requirements for strain compensation and is set as the coating thickness parameter.

[0075] Based on the previously obtained coating thickness parameters, ion-assisted reactive magnetron sputtering is used to deposit a coating on the reverse side of the epitaxial wafer on which the epitaxial single-crystal thin film has been grown, following the aforementioned coating process. After coating, warpage is tested to observe whether the warpage has been optimized to an acceptable bonding level. If the warpage is still unacceptable, the steps described above, after confirming the coating thickness parameters, to deposit a stress-compensating film on the reverse side of the epitaxial wafer on which the epitaxial single-crystal thin film has been grown, are repeated. If the warpage is acceptable, proceed to the next step.

[0076] A composite material is obtained by directly bonding the surface of an epitaxial single-crystal thin film away from the epitaxial substrate to a cantilever substrate (including but not limited to single-crystal silicon) using methods such as surface activation bonding and plasma activation bonding.

[0077] The stress compensation film and epitaxial substrate on the side of the epitaxial single crystal film away from the cantilever substrate are removed by substrate thinning processes (including but not limited to physical thinning, chemical mechanical polishing, and wet etching). It should be noted that this step can completely remove the epitaxial substrate, or a certain thickness of epitaxial substrate can be retained for the purpose of protecting the epitaxial single crystal film.

[0078] After wafer dicing, a device for measuring the low-frequency mechanical loss of epitaxial single-crystal thin films in the liquid helium temperature range is obtained. If a certain thickness of epitaxial substrate was retained in the previous step, the remaining epitaxial substrate needs to be removed using a selective wet etching process after wafer dicing.

[0079] According to embodiments of the present invention, the above-described preparation method can fabricate cantilever devices for low-temperature measurement of mechanical losses in epitaxial single-crystal thin films. During the fabrication of the cantilever substrate, piranha solution and BOE cleaning are added multiple times to prevent the formation of the etching mask, and the liquid is circulated and filtered during wet etching, achieving better cantilever surface roughness, thereby enabling more accurate measurement of thin film mechanical losses. By using ion-assisted reactive magnetron sputtering to grow the stress compensation film, a stress compensation film with sufficiently high stress can be deposited at a lower temperature, avoiding the high-temperature damage to the epitaxial substrate and epitaxial single-crystal thin film caused by chemical vapor deposition. Improving the adhesion of the stress compensation film grown by ion-assisted reactive magnetron sputtering to substrates with poor surface adhesion, such as gallium arsenide, allows the stress compensation film layer to generate higher stress, thereby achieving stress compensation with a greater degree of warpage. For epitaxial single-crystal thin films with high warpage, this method can achieve bonding under lower bonding pressure, helping to protect the fragile epitaxial substrate and the cantilever structure from damage.

[0080] According to another aspect of the present invention, a method for measuring the mechanical loss of epitaxial single-crystal thin films at low temperature is provided, wherein the device is prepared by the fabrication method of the device for measuring the mechanical loss of epitaxial single-crystal thin films as described above.

[0081] According to embodiments of the present invention, group III-V semiconductor single-crystal thin films have wide applications in quantum precision measurement fields such as atomic optical lattice clocks. Since these quantum precision measurement methods are mostly performed at low temperatures, with devices typically operating at 4K, 17K, and 124K, it is necessary to understand the mechanical loss level of group III-V semiconductor single-crystal thin films in a temperature range, such as 4K to room temperature, to improve the performance of precision devices. However, due to limitations in measurement devices and device fabrication processes, the mechanical loss measurement of group III-V compound epitaxial crystal thin films in current related technologies is usually performed at room temperature. To perform mechanical loss measurement of semiconductor crystal films in a temperature range, such as 4K to room temperature, it is necessary to fabricate devices coated with epitaxial single-crystal thin films. Furthermore, due to the small size of the devices and the low cost of cooling equipment, the temperature can be lowered to a relatively low level.

[0082] Figure 4 A schematic diagram of a device for measuring the mechanical loss of epitaxial single-crystal thin films according to an embodiment of the present invention is shown, wherein, Figure 4 Image (a) shows a side view of the device. Figure 4(b) shows a top view of the device. (e.g.) Figure 4 As shown, the device includes a first part (such as...) Figure 4 (a) the thinner part on the left) and the second part (as in Figure 4 (b) The thicker part on the right), the second part is thicker, for example, 500 μm; the first part of the device is thinner, for example, 60~80 μm, the length of the first part of the device is for example 30~35 mm, and the width is for example 5 mm, the length of the second part of the device is for example 8 mm, and the width is for example 5 mm, the epitaxial single crystal film covers the flat surface of the device, the flat surface can be understood as the ungrooved surface of the cantilever substrate, that is Figure 4 The bottom surface of (a).

[0083] In the process of measuring the mechanical loss of epitaxial single-crystal thin films, the second part of the device is held by a clamping assembly located in the vacuum chamber of the mechanical loss measuring device. The first part of the device extends from the clamping assembly and includes the epitaxial single-crystal thin film. By driving the assembly to drive the first part of the device to vibrate, incident laser light is incident on the surface of the epitaxial single-crystal thin film side of the device. After being reflected by the epitaxial single-crystal thin film, reflected laser light carrying vibration information of the epitaxial single-crystal thin film and the cantilever substrate is obtained. The mechanical loss of the coated device is determined by using the reflected laser light. Of course, it is also possible to measure the mechanical loss of a simple single-crystal silicon cantilever substrate. .

[0084] According to the relevant theories of mechanical loss, the mechanical loss of a simple film layer... It can be obtained from the following formula (1):

[0085] (1).

[0086] In the above formula, It is the Young's modulus of a single-crystal silicon cantilever substrate. It is the thickness of the first part of the single-crystal silicon cantilever substrate. It is the Young's modulus of an epitaxial single-crystal thin film. It is the thickness of the epitaxial single crystal thin film.

[0087] It should be noted that all intrinsic frequencies of the clamping component must avoid all intrinsic frequencies of the device in order to prevent resonant coupling between the clamping component and the device.

[0088] The present invention will be further illustrated below through embodiments and related test experiments and results. In the following detailed description, numerous specific details are set forth for ease of explanation to provide a comprehensive understanding of the embodiments of the present invention. However, it will be apparent that one or more embodiments may be practiced without these specific details. Moreover, the details in the following embodiments can be arbitrarily combined to form other feasible embodiments without conflict.

[0089] It should be noted that the specific embodiments described below are merely illustrative examples, and the scope of protection of this invention is not limited thereto. The chemicals and raw materials used in the following embodiments are all commercially available or prepared using recognized processing methods.

[0090] Example 1:

[0091] In this embodiment, the raw material used to prepare the cantilever substrate is a double-sided polished intrinsic single-crystal silicon wafer with a thickness of 500 μm. Intrinsic single-crystal silicon wafers are used to ensure stable etching rate and surface roughness. The surface roughness R of each side of the single-crystal silicon wafer is... q The thickness is <0.2nm, and the total thickness variation (TTV) is <2μm, to accommodate subsequent bonding processes. High-density silicon nitride is deposited on both sides of a single-crystal silicon wafer using low-pressure chemical vapor deposition, and positive ultraviolet photoresist is used to define the pattern on the surface. Reactive ion etching is used to dry-etch the surface silicon nitride to expose the single-crystal silicon wafer, completing the pattern transfer on the silicon nitride mask. Subsequently, the surface photoresist is removed to complete the fabrication of the silicon nitride hard mask.

[0092] The monocrystalline silicon wafer was pre-cleaned using a piranha solution (98% sulfuric acid and 30% hydrogen peroxide in a 3:1 volume ratio) to remove residual organic contaminants. The wafer was then immersed for 5 minutes in a BOE solution (40% ammonium fluoride and 40% hydrofluoric acid in a 7:1 volume ratio) to remove the surface oxide layer introduced by the piranha solution and prevent the formation of an etching false mask. Wet etching to a thickness of 70 μm was performed using a 30% potassium hydroxide solution at 65°C, with the solution circulating and filtered during the etching process. After etching, the silicon nitride hard mask was removed using 40% hydrofluoric acid, followed by rinsing with deionized water to obtain a silicon cantilever wafer. The roughness of the etched surface was measured using a white light interferometer. Figure 5 The image shows a roughness measurement diagram of the etched surface of a silicon cantilever wafer after wet etching, as described in Embodiment 1 of the present invention. Figure 5 As shown, the surface roughness R at this time q =2.73nm.

[0093] The epitaxial single-crystal thin film used in this embodiment is a Bragg reflector epitaxial wafer with aluminum gallium arsenide / gallium arsenide stacked arrangement. One layer of aluminum gallium arsenide and one layer of gallium arsenide constitute one stacking cycle. The number of stacking cycles for the Bragg reflector epitaxial wafer is 5.5. The thickness of aluminum gallium arsenide is 119 nm, the thickness of gallium arsenide is 103 nm, and the thickness of the Bragg reflector is 1213 nm. In addition to the Bragg reflector, the Bragg reflector epitaxial wafer also includes a 500 nm aluminum gallium arsenide sacrificial layer and a 300 nm gallium arsenide buffer layer. The total thickness of the Bragg reflector epitaxial wafer is 2013 nm. It is prepared by molecular beam epitaxy on a 4-inch gallium arsenide substrate. Warpage of the epitaxial wafer is measured to obtain its warpage, in preparation for warpage compensation. Figure 7a The diagram shows a warpage measurement of a 4-inch aluminum gallium arsenide / gallium arsenide Bragg reflector epitaxial wafer according to Embodiment 1 of the present invention. The warpage was measured using a GSS plus Semi-auto 3D wafer surface scanning metrology system. Figure 7a As shown, the BOW (Bend-Off W) of the epitaxial wafer measured over the entire 4-inch area is 11 μm. Surface-activated bonding or plasma-activated bonding cannot be performed at this point. Similarly, the warpage of the gallium arsenide substrate was measured.

[0094] A 4-inch gallium arsenide substrate, identical to the epitaxial wafer, was used. Silicon nitride was deposited on the gallium arsenide substrate using ion-assisted reactive magnetron sputtering. After the sample was placed in the deposition chamber, the chamber was evacuated to a vacuum level of 5E-6 Torr. Argon gas was then introduced into the chamber at a flow rate of 40 sccm. Radio frequency bias was activated at a power of 120 W, and the surface of the epitaxial substrate was pre-cleaned for 60 seconds.

[0095] Figure 6a An optical microscope image of a gallium arsenide substrate without pre-cleaning and directly coated according to Embodiment 1 of the present invention is shown. Figure 6b An optical microscope image of a gallium arsenide substrate surface after pre-cleaning and subsequent coating according to Embodiment 1 of the present invention is shown. Figures 6a-6b As can be seen from the above pre-cleaning, the surface of the formed gallium arsenide substrate is cleaner and the surface energy is increased, thereby improving the adhesion of the silicon nitride stress compensation film and improving the stability of stress compensation.

[0096] Nitrogen and argon gases were then introduced into the chamber at a flow rate of 25 sccm and 40 sccm, respectively. Sputtering was then initiated using a silicon target powered by a pulsed DC power supply of 300W with a duty cycle of 80%. Simultaneously, an RF bias was applied to bombard the epitaxial substrate surface during the deposition process at a power of 100W. Wafer warpage was measured using a wafer warpage measurement device to obtain the warpage after deposition, from which the rate of warpage increase could be calculated. The deposition time was calculated based on this rate of increase, aiming to ensure that the warpage caused by the silicon nitride thin film was consistent with the warpage caused by the molecular beam epitaxy aluminum gallium arsenide / gallium arsenide Bragg mirror film. At this point, the preliminary determination of the deposition parameters was completed. Figure 7b The diagram shows a warpage measurement of a 4-inch gallium arsenide substrate used for confirming stress-compensated coating process parameters, as illustrated in Embodiment 1 of the present invention. Figure 7b As shown, in this embodiment, after 800 s of silicon nitride deposition by ion-assisted reactive magnetron sputtering, the BOW of the gallium arsenide substrate increased by 7 μm. Based on the previous measurement of the warpage of the epitaxial wafer as 11 μm, the deposition time for silicon nitride deposition on the back side of the epitaxial wafer can be determined to be 800 / 7×11=1257 s.

[0097] A new aluminum gallium arsenide / gallium arsenide Bragg reflector epitaxial wafer, identical to those described above, was used. The reverse side of the bonding surface was cleaned with isopropanol and dried. Silicon nitride was then deposited using essentially the same ion-assisted reactive magnetron sputtering process parameters as described above, the only difference being that the deposition time was increased to 1257 s. After deposition, the surface shape of the bonding surface was measured using a laser interferometer or a white light interferometer. The deposition parameters were adjusted based on the warpage until the warpage reached a level suitable for plasma-activated bonding. Figure 7c The diagram shows a warpage measurement of a 4-inch aluminum gallium arsenide / gallium arsenide Bragg reflector epitaxial wafer after stress compensation, according to Embodiment 1 of the present invention. Figure 7c As shown, in this embodiment, the BOW of the coated aluminum gallium arsenide / gallium arsenide Bragg reflector epitaxial wafer is measured to be 1.3 μm using a wafer warpage measurement device, which meets the requirements for plasma-activated bonding.

[0098] The warp-compensated aluminum gallium arsenide / gallium arsenide Bragg reflector epitaxial wafer was plasma-activated and bonded to the previously prepared silicon cantilever wafer: both surfaces were simultaneously bombarded with oxygen ions for 75 seconds, followed by bonding under pressure of 10 N under a vacuum of 1E-3 Pa. The bonding area of ​​the bonded pair reached over 95%, and the cantilevered portion did not fracture. After bonding, the bonded pair was pressurized to 10 N, annealed at 50 °C for 10 h, and then annealed at 200 °C for 24 h.

[0099] The bonded pair was mechanically thinned from the gallium arsenide (GaAs) end to remove the stress compensation film and most of the GaAs epitaxial substrate, reducing the thickness from 635 μm to 50 μm. Subsequently, selective wet etching was used to remove the remaining GaAs epitaxial substrate and the aluminum gallium arsenide (AlGaAs) sacrificial layer. What remained on the silicon cantilever was the AlGaAs / GaAs Bragg mirror film to be tested, grown by molecular beam epitaxy. Finally, wafer dicing was performed to obtain the cantilever for measuring the low-temperature mechanical loss of the AlGaAs / GaAs Bragg mirror film.

[0100] To measure the mechanical loss of the aluminum gallium arsenide / gallium arsenide Bragg mirror film, the mechanical loss of the silicon cantilever first needs to be determined. Multiple silicon cantilever arms were fabricated from another silicon cantilever sheet (without bonded aluminum gallium arsenide / gallium arsenide Bragg mirror film) prepared simultaneously from the same batch of substrates using wafer dicing. The mechanical loss of the silicon cantilever arms was measured by clamping the thicker portion of the cantilever with a clamping assembly of a mechanical loss measuring device, while the thinner portion of the cantilever extended out from the clamping assembly. After cooling to 4K, the cantilever was driven to vibrate. An incident laser was incident on the silicon cantilever surface, forming a reflected laser beam. The mechanical loss of the silicon cantilever was determined to be 9.4E-8 to 3.9E-7 using the reflected laser beam.

[0101] Furthermore, the mechanical loss of the aluminum gallium arsenide / gallium arsenide Bragg mirror was detected. The thicker portion of the cantilever bonded with the aluminum gallium arsenide / gallium arsenide Bragg mirror film was held by a clamping component of a mechanical loss measuring device, while the thinner portion of the cantilever extended from the clamping component. After cooling to 4K, the cantilever was driven to vibrate. An incident laser was incident on the aluminum gallium arsenide / gallium arsenide Bragg mirror and reflected, which was used to determine the mechanical loss of the cantilever bonded with the aluminum gallium arsenide / gallium arsenide Bragg mirror film. Finally, using the aforementioned formula (1) for the mechanical loss of the film, the mechanical loss of the aluminum gallium arsenide / gallium arsenide Bragg mirror film was obtained as 2.3E-5~5.0E-5.

[0102] Thus, based on the cantilever formed by integrating a single-crystal silicon cantilever substrate with an aluminum gallium arsenide / gallium arsenide Bragg mirror film, it is possible to accurately measure the mechanical loss of the aluminum gallium arsenide / gallium arsenide Bragg mirror film at low temperatures, which will help to further develop low-temperature systems such as atomic optical lattice clocks for application in the field of quantum precision measurement.

[0103] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for preparing a device for measuring mechanical loss of an epitaxial single-crystal thin film, characterized by, include: An epitaxial single-crystal thin film is grown on the front side of an epitaxial substrate by molecular beam epitaxy, resulting in an epitaxial substrate with an epitaxial single-crystal thin film on the front side. The epitaxial substrate includes a front side and a back side. A coating process is performed on the back side of the epitaxial substrate to compensate for the stress of the warping of the epitaxial single crystal film, resulting in an epitaxial substrate with an epitaxial single crystal film on the front side and a stress compensation film on the back side. The surface of the epitaxial single crystal film away from the epitaxial substrate is directly bonded to the cantilever substrate to form a bonding pair, thereby obtaining a composite material; The stress compensation film and the epitaxial substrate on the side of the epitaxial single crystal film away from the cantilever substrate in the composite material are removed to obtain the device; The cantilever substrate is made of single-crystal silicon, the epitaxial substrate is a group III-V semiconductor single-crystal substrate, and the epitaxial single-crystal thin film is a group III-V semiconductor single-crystal thin film.

2. The production method according to claim 1, characterized by, A sacrificial layer is also included between the epitaxial single crystal thin film and the epitaxial substrate; The epitaxial substrate comprises gallium arsenide; The epitaxial single crystal thin film includes either a single-layer film or a multilayer film. The stress compensation film is made of silicon nitride.

3. The method of claim 2, wherein, The coating process is performed on the back side of the epitaxial substrate by ion-assisted reactive magnetron sputtering.

4. The production method according to claim 3, characterized by, The coating process performed on the back side of the epitaxial substrate by ion-assisted reactive magnetron sputtering includes: Argon is used as the working gas and nitrogen is used as the reaction gas. A silicon target is used to perform a coating process on the back side of the epitaxial substrate by reactive magnetron sputtering. During the coating process, an radio frequency bias is activated to bombard the back side of the epitaxial substrate to deposit the stress compensation film on the back side of the epitaxial substrate.

5. The preparation method according to claim 4, characterized in that, The vacuum degree of the coating process is lower than 1E-5 Torr, and the ratio of nitrogen flow rate to argon flow rate is 1:(1~10); the power supply of the silicon target is 250~350W, and the power of the radio frequency bias voltage is 80~120W.

6. The preparation method according to claim 4, characterized in that, The warpage of the epitaxial single crystal film after stress compensation is less than 5 μm.

7. The preparation method according to claim 1, characterized in that, The conditions for direct bonding include: The surface of the epitaxial single crystal film away from the epitaxial substrate and the surface of the cantilever substrate are bombarded with oxygen ions, and pressure bonding is performed under a pressure of 5~20N, followed by annealing.

8. The preparation method according to claim 1, characterized in that, The cantilever substrate is prepared by the following method: Mask layers are deposited on two opposite surfaces of a single-crystal silicon substrate to obtain a single-crystal silicon substrate with mask layers on both surfaces. Patterning, resist removal, and cleaning are performed sequentially on the mask layer of either of the two surfaces to obtain a single-surface patterned monocrystalline silicon substrate. The cantilever substrate is obtained by wet etching on the patterned surface side of the single-surface patterned monocrystalline silicon substrate and cleaning.

9. The preparation method according to claim 8, characterized in that, Between depositing the mask layer and performing the patterning process, the following is also included: The monocrystalline silicon substrates with mask layers on both surfaces were cleaned using a piranha solution, and the cleaned monocrystalline silicon substrates with mask layers on both surfaces were then immersed in a buffer oxide etching solution. In the piranha solution, the volume ratio of sulfuric acid to hydrogen peroxide is (3~10):1; in the buffer oxide etching solution, the volume ratio of ammonium fluoride solution to hydrofluoric acid is (3~10):

1.

10. A method for measuring the mechanical loss of epitaxial single-crystal thin films at low temperatures, characterized in that, The device is prepared by the fabrication method of any one of claims 1 to 9 for measuring the mechanical loss of an epitaxial single crystal thin film.

Images