High-sensitivity MEMS capacitive accelerometer based on novel folded beam structure and preparation method thereof

Abstract

The invention relates to a high-sensitivity MEMS capacitive accelerometer based on a novel folding beam structure and a preparation method thereof, and relates to the technical field of MEMS capacitive accelerometers, and the accelerometer comprises an upper cover which is provided with an upper fixed capacitor plate; a lower fixed capacitor plate is arranged on the lower cover; the sensitive silicon structure is arranged between the upper fixed capacitance plate and the lower fixed capacitance plate and comprises a fixed frame, a sensitive mass block located in the middle of the fixed frame and a folding beam used for connecting the sensitive mass block and the matching beam, and the folding beam is a snake-shaped beam and is of a variable cross-section structure; the width of the section is gradually reduced from the end connected with the matching beam to the end connected with the sensitive mass block. By adopting the variable cross-section snake-shaped beam structure, the wide part of the connecting end can bear large acceleration, and the structural stability is guaranteed; the rigidity of the narrower end is low, the sensitive mass block is easier to displace under the action of external acceleration, the capacitance change is more obvious, the detection sensitivity is improved, and the two aspects of requirements are considered.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of MEMS capacitive accelerometers, in particular to a high-sensitivity MEMS capacitive accelerometer based on a novel folded beam structure and a preparation method. BACKGROUND

[0002] The MEMS capacitive accelerometer is a sensor manufactured based on micro-electro-mechanical system technology, which measures acceleration by detecting the change of a detection capacitor caused by the displacement of a movable mass under the action of acceleration. The MEMS capacitive accelerometer is usually composed of a fixed frame, a sensitive mass, an elastic support structure (such as a beam), and a fixed electrode and a movable electrode for capacitive detection. When there is an external acceleration, the sensitive mass produces a relative displacement under the action of inertial force, resulting in a change in the capacitance value between the fixed electrode and the movable electrode. The acceleration information can be obtained by detecting the change in the capacitance.

[0003] In the process of pursuing high sensitivity, the existing MEMS capacitive accelerometer faces key technical problems. On the one hand, as the core elastic support structure, the folded beam is difficult to simultaneously consider high sensitivity and structural stability. If the traditional constant cross-section folded beam is reduced in stiffness to improve sensitivity, it is easy to cause insufficient structural strength and is easy to be damaged when bearing a large acceleration. If the structural strength is ensured, the sensitivity is difficult to improve. On the other hand, temperature changes will cause changes in parameters such as the stiffness of the beam and the size of the capacitive electrode plate, resulting in temperature drift and affecting the measurement accuracy. The existing structure and signal processing method have poor inhibition effect on temperature drift. Therefore, the high-sensitivity MEMS capacitive accelerometer based on the novel folded beam structure is designed and processed. SUMMARY

[0004] In view of the deficiencies of the prior art, the application provides a high-sensitivity MEMS capacitive accelerometer based on a novel folded beam structure and a preparation method. By designing the folded beam as a snake shape and adopting a variable cross-section structure for the folded part, the sensitivity is improved while the structural stability is ensured. By adopting a multi-layer structure for the sensitive mass and setting a comb-shaped protruding electrode, the capacitive detection accuracy is further improved, and the problems of the traditional folded beam being difficult to simultaneously consider high sensitivity and structural stability and temperature drift affecting the measurement accuracy are solved.

[0005] To achieve the above-mentioned purposes, the application adopts the following technical solutions:

[0006] The high-sensitivity MEMS capacitive accelerometer based on the novel folded beam structure comprises:

[0007] an upper cover, an upper fixed capacitive plate is arranged on the upper cover, the upper fixed capacitive plate comprises an upper detection electrode and an upper matching electrode surrounding the periphery of the upper detection electrode;

[0008] A lower cover, on which a lower fixed capacitance plate is arranged, the lower fixed capacitance plate comprising a lower detection electrode and a lower matching electrode surrounding the lower detection electrode;

[0009] A sensitive silicon structure arranged between the upper fixed capacitance plate and the lower fixed capacitance plate, the sensitive silicon structure comprising:

[0010] A fixed frame, each side of which is provided with a stress release slot, and a matching beam is formed between the stress release slot and the inner side of the fixed frame;

[0011] A sensitive mass block arranged in the middle of the fixed frame;

[0012] A folded beam for connecting the sensitive mass block and the matching beam, the folded beam being a serpentine beam, and the serpentine beam adopts a variable cross-section structure, the cross-section width of the variable cross-section structure gradually decreases from 25 μm-35 μm to 12 μm-18 μm from the connecting end of the matching beam to the connecting end of the sensitive mass block.

[0013] Preferably, the sensitive mass block adopts a multi-layer structure, comprising a top layer mass block, an intermediate layer mass block and a bottom layer mass block;

[0014] The top layer mass block, the intermediate layer mass block and the bottom layer mass block are connected into one body through a vertical connecting column, the diameter of the vertical connecting column being 8 μm-12 μm;

[0015] The side of the top layer mass block opposite to the upper detection electrode is provided with an upper protruding electrode in the shape of a comb tooth, the side of the bottom layer mass block opposite to the lower detection electrode is provided with a lower protruding electrode in the shape of a comb tooth, and the comb teeth of the upper protruding electrode and the comb teeth of the upper detection electrode are staggered with each other, and the comb teeth of the lower protruding electrode and the comb teeth of the lower detection electrode are staggered with each other.

[0016] Preferably, the material of the top layer mass block, the intermediate layer mass block and the bottom layer mass block is single crystal silicon;

[0017] The material of the vertical connecting column is silicon dioxide;

[0018] The material of the upper protruding electrode and the lower protruding electrode is doped polysilicon, and the doping concentration is .

[0019] Preferably, the surface of the upper fixed capacitance plate and the lower fixed capacitance plate is covered with an insulating layer with a thickness of 200 nm-500 nm, and the material of the insulating layer is silicon nitride;

[0020] A metal electrode layer with a thickness of 500 nm-1000 nm is arranged on the insulating layer, and the material of the metal electrode layer is aluminum.

[0021] Preferably, the outer edge of the fixed frame has multiple anchor points for fixing the entire accelerometer to the external carrier;

[0022] The fixed anchor point is circular in shape with a diameter of 50μm-100μm, and the material of the fixed anchor point is the same as that of the fixed frame, which is monocrystalline silicon.

[0023] Preferably, the upper and lower covers are connected to the sensitive silicon structure via a silicon-silicon bonding process; the silicon-silicon bonding method is as follows:

[0024] Before bonding, the bonding surfaces of the upper cover, lower cover and sensitive silicon structure are pretreated by plasma cleaning, wet chemical etching and deionized water rinsing in sequence.

[0025] The bonding temperature of the silicon-silicon bonding process is 830℃-870℃, the bonding pressure is 1.2MPa-1.8MPa, and the bonding time is 35min-45min.

[0026] Preferably, the accelerometer further includes a signal processing circuit electrically connected to the upper detection electrode, the lower detection electrode, the upper matching electrode, and the lower matching electrode, the signal processing circuit including:

[0027] A preamplifier circuit with an amplification factor of 10-50 times is used to initially amplify the weak capacitance change signals output by the upper and lower detection electrodes.

[0028] A differential amplifier circuit used to perform differential processing on the signal output from a preamplifier circuit to eliminate common-mode interference;

[0029] A second-order low-pass Butterworth filter circuit is used to filter the signal output of a differential amplifier circuit to remove high-frequency noise. Its cutoff frequency is 1kHz-10kHz.

[0030] An A / D conversion circuit used to convert analog signals output from a filter circuit into digital signals.

[0031] Preferably, the signal processing circuit further includes a temperature compensation circuit, which is used to compensate for the temperature drift of the accelerometer. The compensation method is as follows:

[0032] The temperature of the sensitive silicon structure is measured by an integrated temperature sensor integrated on the sensitive silicon structure;

[0033] The acceleration signal output by the signal processing circuit is corrected based on a pre-built temperature-drift compensation model.

[0034] The temperature-drift compensation model is constructed as follows: the accelerometer is calibrated at different temperatures, the temperature and the corresponding acceleration drift are recorded, the temperature-drift compensation curve is obtained through polynomial fitting, and then the temperature-drift compensation model is established.

[0035] The fabrication method of a high-sensitivity MEMS capacitive accelerometer based on a novel folded beam structure includes the following steps:

[0036] S1: The structural pattern of the fixed frame, sensitive mass block, folded beam, stress relief groove and matching beam is fabricated on a single crystal silicon wafer using deep ultraviolet lithography.

[0037] S2: Ion implantation doping treatment is performed on the single-crystal silicon wafer with the structural pattern fabricated in S1;

[0038] S3: On another single-crystal silicon wafer, electrode patterns for the upper and lower fixed capacitor plates are fabricated using conventional photolithography and magnetron sputtering processes.

[0039] S4: On a single-crystal silicon wafer with the electrode pattern fabricated, a silicon nitride insulating layer is grown using a chemical vapor deposition process.

[0040] S5: Perform silicon-silicon bonding between the monocrystalline silicon wafers treated in S2 and the monocrystalline silicon wafers of the upper and lower fixed capacitor plates treated in S4.

[0041] S6: Process the monocrystalline silicon wafers of the upper and lower covers to match their size and shape with the bonded structure, and open windows at corresponding positions to connect with the sensitive silicon structure and the fixed capacitor plate;

[0042] S7: Assemble the top cover, the bonded structure and the bottom cover, and connect them into one piece through silicon-silicon bonding process to form a complete accelerometer chip;

[0043] S8: Package the accelerometer chip and signal processing circuit.

[0044] Preferably, in step S1, when fabricating the structural pattern, a folded beam with a variable cross-section structure is formed using multiple photolithography and etching processes, specifically:

[0045] First, the overall outline of the folded beam is created through a first photolithography and etching process;

[0046] A special mask is made. In the mask pattern, the width of the folded beam gradually changes from the end connected to the matching beam to the end connected to the sensitive mass block in a stepped transition form. The number of steps and the step width are made according to the design requirements.

[0047] The non-mask area of ​​the folded beam is then thinned through a second photolithography and etching process to form a variable cross-section structure.

[0048] Compared with the prior art, the present invention provides a high-sensitivity MEMS capacitive accelerometer based on a novel folded beam structure, which has the following advantages:

[0049] (1) The high-sensitivity MEMS capacitive accelerometer based on a novel folded beam structure provided by the present invention adopts a variable cross-section serpentine beam structure. From the connection end with the matching beam to the connection end with the sensitive mass block, the cross-sectional width decreases. The wider part of the connection end can withstand a larger acceleration, ensuring structural stability; the narrower end has low stiffness, and the sensitive mass block is more likely to displace when subjected to external acceleration, which makes the capacitance change more significant and improves the detection sensitivity, thus meeting both requirements.

[0050] (2) The high-sensitivity MEMS capacitive accelerometer based on a novel folded beam structure provided by the present invention has a temperature compensation circuit that plays a key role in signal processing circuits. The temperature sensor integrated on the sensitive silicon structure monitors the temperature in real time. Since temperature changes will affect the beam stiffness and the size of the capacitor plates, the capacitance value will be affected. Based on the pre-built temperature-drift compensation model, the signal processed by the pre-amplifier, differential amplifier and other circuits is corrected to offset the influence of temperature factors, reduce temperature drift and improve measurement accuracy. Attached Figure Description

[0051] Figure 1 This is a schematic diagram of the fabrication process of the high-sensitivity MEMS capacitive accelerometer based on a novel folded beam structure provided by the present invention. Detailed Implementation

[0052] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments and accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] A high-sensitivity MEMS capacitive accelerometer based on a novel folded beam structure includes:

[0054] The upper cover has an upper fixed capacitor plate, which includes an upper detection electrode and an upper matching electrode surrounding the upper detection electrode.

[0055] The lower cover has a lower fixed capacitor plate, which includes a lower detection electrode and a lower matching electrode surrounding the lower detection electrode.

[0056] A sensitive silicon structure disposed between an upper fixed capacitor plate and a lower fixed capacitor plate, the sensitive silicon structure comprising:

[0057] A fixed frame is provided with stress relief grooves on each side, forming a matching beam between the stress relief grooves and the inner side of the fixed frame;

[0058] A sensitive mass block located in the middle of a fixed frame;

[0059] Folded beams used to connect sensitive mass blocks to matching beams;

[0060] In this invention, the folding beam is a serpentine beam, and the folding part of the serpentine beam adopts a variable cross-section design. From the end connected to the matching beam to the end connected to the sensitive mass block, the cross-sectional width gradually decreases from 25μm-35μm to 12μm-18μm.

[0061] In a preferred embodiment, the sensitive mass block adopts a multi-layer structure, consisting of a top mass block, an intermediate mass block, and a bottom mass block, which are connected as a single unit by vertical connecting posts. The diameter of the vertical connecting posts is preferably 8μm-12μm.

[0062] In this embodiment, the top mass block has an upper protruding electrode in the shape of a comb on the side opposite to the upper detection electrode, and the bottom mass block has a lower protruding electrode in the shape of a comb on the side opposite to the lower detection electrode. The comb teeth of the upper protruding electrode and the comb teeth of the upper detection electrode are interleaved, and the comb teeth of the lower protruding electrode and the comb teeth of the lower detection electrode are interleaved.

[0063] In a preferred embodiment, the top layer mass block, the middle layer mass block, and the bottom layer mass block are made of monocrystalline silicon; the vertical connecting pillars are made of silicon dioxide; and the upper and lower protruding electrodes are made of doped polycrystalline silicon with a doping concentration of [missing information]. .

[0064] In a preferred embodiment, both the upper and lower fixed capacitor plates are covered with an insulating layer with a thickness of 200nm-500nm, preferably made of silicon nitride. A metal electrode layer with a thickness of 500nm-1000nm is disposed on the insulating layer, preferably made of aluminum.

[0065] In a preferred embodiment, the outer edge of the fixing frame has multiple fixing anchor points for securing the entire accelerometer to the external carrier. Preferably, the fixing anchor points are circular in shape with a diameter of 50μm-100μm, and the material of the fixing anchor points is the same as that of the fixing frame, which is monocrystalline silicon.

[0066] In a preferred embodiment, the upper and lower covers are connected to the sensitive silicon structure via a silicon-silicon bonding process, the specific implementation of which is as follows:

[0067] Before bonding, the bonding surfaces of the upper cover, lower cover, and sensitive silicon structure are sequentially subjected to plasma cleaning, wet chemical etching, and deionized water rinsing pretreatment. Among them, argon gas is used as the cleaning gas, with a power of 100W-200W and a time of 5min-10min; wet chemical etching uses a mixed solution of hydrofluoric acid concentration of 5%-10% and nitric acid concentration of 20%-30%, with an etching time of 30s-60s; the bonding temperature of the silicon-silicon bonding process is 830℃-870℃, the bonding pressure is 1.2MPa-1.8MPa, and the bonding time is 35min-45min.

[0068] In a preferred embodiment, the accelerometer further includes a signal processing circuit electrically connected to the upper detection electrode, the lower detection electrode, the upper matching electrode, and the lower matching electrode. The signal processing circuit includes:

[0069] A preamplifier circuit with an amplification factor of 10-50 times is used to initially amplify the weak capacitance change signals output by the upper and lower detection electrodes.

[0070] A differential amplifier circuit used to perform differential processing on the signal output from a preamplifier circuit to eliminate common-mode interference;

[0071] A second-order low-pass Butterworth filter circuit is used to filter the signal output of a differential amplifier circuit to remove high-frequency noise. Its cutoff frequency is 1kHz-10kHz.

[0072] An A / D conversion circuit with a resolution of 16 bits, used to convert analog signals output from a filter circuit into digital signals for subsequent data processing and transmission.

[0073] In a preferred embodiment, the signal processing circuit further includes a temperature compensation circuit, which is used to compensate for the temperature drift of the accelerometer, specifically as follows:

[0074] The temperature of the sensitive silicon structure is measured by an integrated temperature sensor integrated on the sensitive silicon structure;

[0075] The acceleration signal output by the signal processing circuit is corrected based on the temperature-drift compensation model obtained through prior experimental testing.

[0076] It should be noted that the experimental test involved calibrating the accelerometer at different temperatures, recording the temperature and the corresponding acceleration drift, obtaining the temperature-drift compensation curve through polynomial fitting, and then establishing a temperature-drift compensation model.

[0077] Please see Figure 1 A method for fabricating a high-sensitivity MEMS capacitive accelerometer based on a novel folded beam structure includes the following steps:

[0078] S1: Prepare a single-crystal silicon wafer and use deep ultraviolet lithography to fabricate the structural patterns of the fixed frame, sensitive mass block, folded beam, stress relief groove and matching beam. The lithography resolution of this deep ultraviolet lithography technology is 0.5μm-1μm. In the fabrication process, inductively coupled plasma dry etching process is used, with an etching depth of 100μm-200μm.

[0079] S2: Ion implantation doping is performed on the single-crystal silicon wafer with the fabricated structural pattern in S1. The ion type is boron ions, the implantation energy is 50keV-100keV, and the implantation dose is 1×10⁻⁶. 15 cm -2 -5×10 15 cm -2 To adjust the electrical properties of the structure;

[0080] S3: On another single-crystal silicon wafer, electrode patterns for the upper and lower fixed capacitor plates are fabricated using conventional photolithography and magnetron sputtering. The photolithography resolution of conventional photolithography is 1μm-2μm; the metal material for magnetron sputtering is aluminum, and the sputtering thickness is 500nm-1000nm.

[0081] S4: On a single-crystal silicon wafer with the electrode pattern fabricated, a silicon nitride insulating layer is grown by chemical vapor deposition. The thickness of the insulating layer is 200nm-500nm.

[0082] S5: The single-crystal silicon wafer (i.e., the sensitive silicon structure) after S2 treatment is bonded to the single-crystal silicon wafers of the upper and lower fixed capacitor plates after S4 treatment with silicon-silicon bonding. The bonding temperature is 800℃-1000℃, the bonding pressure is 1MPa-3MPa, and the bonding time is 30min-60min.

[0083] S6: Process the monocrystalline silicon wafers of the upper and lower covers to match their size and shape with the bonded structure, and open windows at corresponding positions to connect with the sensitive silicon structure and the fixed capacitor plate;

[0084] S7: Assemble the top cover, the bonded structure and the bottom cover, and connect them into one piece through silicon-silicon bonding process to form a complete accelerometer chip;

[0085] S8: The accelerometer chip and signal processing circuit are packaged using ceramic packaging technology, ensuring good electrical connection between the chip and the signal processing circuit during the packaging process.

[0086] As a preferred implementation, in step S1, when creating the structural pattern, multiple photolithography and etching processes are used to achieve the variable cross-section design of the folded beam, specifically:

[0087] First, the overall outline of the folded beam is created through a first photolithography and etching process;

[0088] Then, a second photolithography and etching process is used to thin a portion of the folded beam to form a variable cross-section structure. The mask pattern for the second photolithography is fabricated according to the design requirements of the variable cross-section. The etching process uses reactive ion etching, and the etching depth is controlled according to the size requirements of the variable cross-section.

[0089] Example 1:

[0090] This embodiment mainly introduces the processing of the folded beam variable cross-section structure; in MEMS capacitive accelerometers, the variable cross-section design of the folded beam plays a key role in its performance, which is related to the balance between structural stability and sensitivity; the processing steps will be described in detail below.

[0091] Select a thickness of 525μm <100> A single-crystal silicon wafer with crystal orientation was used as the substrate. In the photolithography process, deep ultraviolet photolithography was used to fabricate the overall outline of the folded beam. AZ4620 photoresist was selected with a thickness of 10μm. The photoresist was homogenized at a speed of 1000r / min for 30s. The pre-baking temperature was set at 105℃ for 2min, the exposure time was 40s, the development time was 90s, and the post-baking temperature was 120℃ for 30min. The photolithography resolution was precisely controlled at 0.8μm.

[0092] The etching process employs inductively coupled plasma (ICP) dry etching, using a mixture of SF6 and C4F8 as the etching gas with a flow ratio of 3:1, an RF power of 200W, and an etching depth of 150μm. This initially forms a serpentine folded beam structure, with an initial width of 30μm at the connection end with the matching beam and 20μm at the connection end with the sensitive mass block. The overall length of the folded section is 800μm.

[0093] To achieve the variable cross section, a special mask needs to be made. In the mask pattern, the width gradient area of ​​the folded beam from the end connected to the matching beam to the end connected to the sensitive mass block adopts a stepped transition design, which is divided into 5 steps, and the width of each step is reduced by 3μm.

[0094] Photolithography is performed again, using LC100 photoresist with a thickness of 2μm, an exposure time of 8s, and a development time of 45s to ensure that the mask pattern is precisely aligned with the contour of the first etching, with the alignment error controlled within ≤1μm.

[0095] The etching process employs reactive ion etching (RIE), with SF6 as the etching gas, a flow rate of 50 sccm, and an RF power of 100W. The thickness difference in different areas is achieved by precisely controlling the etching time.

[0096] The etching depth of the 30μm width region connected to the matching beam is 5μm, the etching depth of the intermediate stepped region increases by 3μm, and the etching depth of the 15μm width region connected to the sensitive mass block is 20μm. Finally, a variable cross-section structure that gradually changes from 30μm to 15μm was successfully formed, which meets the design requirements.

[0097] Example 2:

[0098] This embodiment focuses on the fabrication of a multilayer structure of a sensitive mass block; as a key component of the accelerometer, the structural design of the sensitive mass block directly affects the accuracy of capacitance detection. The following is a detailed fabrication process.

[0099] First, prepare three monocrystalline silicon wafers, which will serve as the substrates for the top layer, middle layer, and bottom layer mass blocks, respectively, with thicknesses of 100μm, 200μm, and 100μm.

[0100] For the top-level mass block, a square structure with a size of 1200μm×1200μm was fabricated using deep ultraviolet lithography with a lithographic resolution of 0.5μm. Then, it was etched to a thickness of 80μm using DRIE, leaving 20μm for subsequent attachment of the connecting pillars.

[0101] The intermediate layer mass block is also etched using DRIE to create a structure with dimensions of 1100μm×1100μm and a thickness of 180μm, with a 100μm width reserved at the edge for connection with the folded beam.

[0102] The bottom mass block is fabricated using the same method as the top layer, producing a structure with dimensions of 1200μm×1200μm and a thickness of 80μm.

[0103] A 5 μm thick silicon dioxide layer was grown on the lower surface of the top mass block and the upper surface of the bottom mass block using plasma-enhanced chemical vapor deposition (PECVD).

[0104] The interconnect pattern is defined by photolithography, with a diameter of 10μm and a spacing of 50μm. Then, silicon dioxide is etched to the substrate using RIE to form the outline of the interconnect.

[0105] Then, a 50nm TiW layer was deposited as an adhesion layer by magnetron sputtering, followed by electroplating of silicon dioxide to a diameter of 10μm. Finally, the three-layer mass blocks were bonded together at 850℃ and 1.5MPa pressure for 40min.

[0106] Ion implantation doping was performed on the upper surface of the top mass block opposite the upper detection electrode, and on the lower surface of the bottom mass block opposite the lower detection electrode. Boron ions were used, with an energy of 80 keV and a dose of 3 × 10⁻⁶. 15 cm -2 This forms a conductive layer;

[0107] The comb pattern is defined by photolithography. The comb teeth of the upper and lower protruding electrodes are 10 μm wide, 150 μm long, and 5 μm apart, with 30 pairs of teeth each. The comb structure is formed by DRIE etching to a depth of 30 μm.

[0108] Ensure that the comb teeth of the upper raised electrode and the comb teeth of the upper detection electrode are staggered with an overlap length of 100μm, and that the comb teeth of the lower raised electrode and the comb teeth of the lower detection electrode are also staggered, thereby improving the capacitance detection sensitivity.

[0109] Example 3:

[0110] This embodiment mainly focuses on the processing of the fixed frame and stress relief groove; the reasonable design and processing of the fixed frame and stress relief groove are crucial to ensuring the structural stability of the accelerometer. The processing process is described in detail below.

[0111] A single-crystal silicon wafer with a thickness of 525μm was selected, and the outline of the fixing frame was defined using deep ultraviolet lithography. The outer frame size was set to 1500μm×1500μm, the inner frame size was 1300μm×1300μm, and the side width of the frame was 100μm. DRIE etching technology was used to achieve an etching depth of 500μm, leaving 25μm as a support substrate, thus initially forming the fixing frame structure.

[0112] Stress relief grooves are defined by photolithography on the four sides of the fixed frame. Eight stress relief grooves are evenly distributed on each side. The grooves are U-shaped with their openings facing the inside of the frame. The grooves are 20 μm wide, 500 μm deep, and 100 μm long. After etching, a matching beam with a width of 50 μm is formed between the stress relief grooves and the inside of the fixed frame. The matching beam is used to connect the folding beam and the fixed frame, which can effectively release structural stress and enhance the stability of the fixed frame.

[0113] Eight anchor points are evenly distributed on the outer edge of the fixing frame. The circular anchor point pattern is defined by photolithography, with a diameter of 80μm. The DRIE etching technology is used to penetrate the entire silicon wafer. The anchor points are made of the same material as the fixing frame, which is monocrystalline silicon. To facilitate subsequent electrical connection with the external carrier, a 500nm aluminum layer is deposited on the surface of the anchor points by magnetron sputtering.

[0114] Example 4:

[0115] This embodiment mainly describes the accelerometer materials and surface treatment processes; these processes play a key role in ensuring the performance and stability of the accelerometer, and the specific process of each treatment process will be described in detail below.

[0116] Silicon nitride insulating layers were grown on the surfaces of the upper and lower fixed capacitor plates using hot-wall chemical vapor deposition (CVD). The reaction gases were SiH4 and NH3 at a flow ratio of 1:3, the deposition temperature was 700℃, the pressure was 500Pa, and the deposition time was 30min, resulting in an insulating layer thickness of 300nm. After deposition, silicon nitride in non-insulating areas was removed by RIE etching to ensure that the electrode areas were exposed, preparing for the subsequent fabrication of metal electrode layers.

[0117] An aluminum electrode layer was fabricated on the insulating layer using magnetron sputtering. The sputtering target was 99.99% pure aluminum, the sputtering power was 300W, the argon flow rate was 50sccm, and the vacuum level was maintained at 5×10⁻⁶. -4 Pa sputtering time of 600s resulted in an aluminum electrode layer thickness of 800nm. The electrode pattern was then defined by photolithography and wet etching processes. The etching solution used was a mixed solution of phosphoric acid, nitric acid and acetic acid in a ratio of 5:1:1, thereby forming an upper detection electrode, an upper matching electrode, a lower detection electrode and a lower matching electrode.

[0118] Boron ion implantation was performed on sensitive silicon structures, including folded beams and sensitive mass blocks; the implantation energy was set to 75 keV and the dose to 3 × 10⁻⁶. 15 cm -2 After implantation, annealing is performed under a nitrogen atmosphere at 900℃ for 30 minutes to activate the dopant ions and reduce the resistivity of the doped polycrystalline silicon to 1×10⁻⁶. -3 Ω•cm, meeting the conductivity requirements of both the upper and lower raised electrodes, with a doping concentration of 3×10⁻⁶. 19 cm -3 .

[0119] Example 5:

[0120] This embodiment focuses on silicon-silicon bonding and packaging processes; these processes are key steps in integrating the various components of the accelerometer into a whole, directly affecting the performance and reliability of the accelerometer, and will be described in detail below.

[0121] Before silicon-silicon bonding, the bonding surfaces of the top cover, bottom cover, and sensitive silicon structure are pretreated:

[0122] First, plasma cleaning is performed. The sample is placed in a plasma cleaner, and argon gas is introduced at a flow rate of 200 sccm, a power of 150W, and a cleaning time of 8 minutes to remove organic matter and impurities from the surface.

[0123] Next, wet chemical etching was performed using a mixed solution of 8% hydrofluoric acid and 25% nitric acid. The sample was immersed for 45 seconds to remove the surface oxide layer.

[0124] Finally, rinse three times with deionized water for 1 minute each time, and then dry with nitrogen gas at a flow rate of 500 sccm for 2 minutes.

[0125] The pretreated sensitive silicon structure was precisely aligned with the upper and lower fixed capacitor plates, with the alignment error controlled within ≤2μm. Then, it was placed in a bonding furnace for silicon-silicon bonding. During the bonding process, the heating rate was set to 5℃ / min, the temperature was raised to 850℃, the pressure was applied to 1.5MPa, and the temperature was held for 40min. Nitrogen gas was introduced for protection during the process, with a flow rate of 100sccm. After bonding, the structure was allowed to cool naturally to room temperature at a cooling rate of 3℃ / min to ensure good bonding results.

[0126] The top and bottom covers are made of 1mm thick borosilicate glass (BF33). The dimensions matching the bonding structure are made by laser etching, i.e., 1500μm×1500μm, and windows with a size of 500μm×500μm are opened at the corresponding positions. During laser etching, the laser wavelength is 355nm, the power is 10W, and the etching depth is 500μm to ensure that the windows are precisely aligned with the electrode area of ​​the sensitive silicon structure, which facilitates subsequent lead connection.

[0127] The accelerometer chip and signal processing circuit are housed within a ceramic package.

[0128] First, the chip is fixed to the package base using UV adhesive, with a curing time of 30 seconds and a UV intensity of 300mW / cm². 2 ;

[0129] Then, electrical connections were made using gold wire ball bonding, with a gold wire diameter of 25μm, a bonding temperature of 150℃, and an ultrasonic power of 50mW.

[0130] Finally, the casing is sealed by laser welding with a power of 50W and a welding time of 2 seconds. The interior is filled with nitrogen gas of 99.99% purity to reduce the damping effect, thus completing the entire encapsulation process.

[0131] Example 6:

[0132] This embodiment mainly introduces the implementation process of the signal processing circuit; the signal processing circuit is an important component of the accelerometer, which converts the capacitance change signal into a usable digital signal. The design and working principle of each circuit module are described in detail below.

[0133] The preamplifier circuit uses an operational amplifier OPA2376 to build a non-inverting amplifier circuit; resistor R1 is 1kΩ and R2 is 29kΩ, achieving a magnification of 30 times. The bandwidth is set between 1kHz and 10kHz, which can amplify the weak μV level signal generated by capacitance change to the mV level, which is convenient for subsequent processing.

[0134] The differential amplifier circuit uses the INA128 chip, which effectively suppresses common-mode interference with a common-mode rejection ratio of up to 100dB, ensuring the stability of the output signal.

[0135] The second-order low-pass Butterworth filter circuit consists of resistor R3 (3.3kΩ), capacitor C1 (10nF), and operational amplifier LM358. The cutoff frequency is set to 5kHz, which can effectively remove high-frequency noise and improve signal quality.

[0136] The A / D conversion circuit uses the ADS1115 chip, which has 16-bit resolution and a sampling rate of 860 SPS. It can accurately convert analog signals into digital signals and transmit them to the subsequent processing unit through the I²C interface.

[0137] In the temperature compensation circuit, the DS18B20 integrated temperature sensor is selected and attached to the fixed frame of the sensitive silicon structure. The sensor has an accuracy of ±0.5℃, can measure temperature in real time, and has a measurement range of -40℃ to 125℃.

[0138] Under different temperature conditions, namely -20℃, 0℃, 20℃, 40℃, 60℃, and 80℃, 0g, 1g, and 2g accelerations were applied to the accelerometer via a tumbling platform, and the output drift was recorded. A third-order polynomial fitting method was used, i.e., drift = k0 + k1T + k2T 2 +k3T 3 (where T is temperature), generate compensation curves, and establish a temperature-drift compensation model;

[0139] During operation, the signal processing circuit calls the compensation model in real time to correct the digital signal after A / D conversion, so that the drift after compensation is ≤0.1mg / ℃, effectively reducing the impact of temperature drift on the measurement results.

[0140] Example 7:

[0141] This embodiment mainly introduces the testing, calibration and usage process of the accelerometer; these steps are crucial to ensuring that the accelerometer can accurately measure acceleration. The specific operation process will be described in detail below.

[0142] Fix the accelerometer on the tumbling platform and connect the signal processing circuit and data acquisition system; place the accelerometer in a temperature chamber and control the temperature to gradually increase from -20℃ to 80℃ in 20℃ increments; at each temperature point, output 0°, 90°, 180°, and 270° sequentially through the tumbling platform, corresponding to 0g, 1g, 0g, and -1g accelerations respectively, and record the circuit output values ​​simultaneously;

[0143] Based on the recorded data, the acceleration drift at different temperatures is calculated to update the temperature-drift compensation model, ensuring that the calibration error is controlled within ≤0.5%FS and guaranteeing the accuracy of accelerometer measurements.

[0144] The accelerometer is mounted on an external carrier, such as a car chassis or drone fuselage, by fixing anchor points; M2 screws are used for fixing, and the tightening torque is controlled at 0.5 N•m to ensure that the mounting surface is perpendicular to the sensitive axis, with a perpendicularity error of ≤0.1°.

[0145] The capacitance change signal first enters the preamplifier circuit for 30x amplification, then passes through the differential amplifier circuit to suppress common-mode interference, and then through the second-order low-pass Butterworth filter circuit with a cutoff frequency of 5kHz to remove high-frequency noise. Next, the analog signal is converted into a digital signal by the 16-bit resolution A / D conversion circuit. Finally, the digital signal is corrected by the temperature compensation circuit according to the temperature-drift compensation model, and finally the digital signal is output through the UART interface at a baud rate of 115200.

[0146] A conventional accelerometer of the same type on the market was selected as a control group. The performance of the accelerometer of the present invention and the control group accelerometer were tested under the same temperature (25°C) and humidity (50%RH) conditions.

[0147] The same acceleration change was applied to two sets of accelerometers, gradually increasing from 0g to 5g and then gradually decreasing back to 0g. The output capacitance change signal at different acceleration values ​​was recorded and converted into acceleration measurement values. The test results are shown in Table 1 below:

[0148] Table 1 Comparison of accelerometer measurement results

[0149]

[0150] The experimental data show that the accelerometer of the present invention measures values ​​that are closer to the true values ​​across the entire range, with smaller measurement errors. At low accelerations, such as 0g and 1g, the measurement accuracy advantage is obvious.

[0151] At high accelerations—such as 5g—the control group accelerometers showed large errors, while the accelerometers of this invention could still maintain relatively accurate measurements.

[0152] Furthermore, after temperature compensation, when the above tests were repeated under different temperature environments, the measurement accuracy of the accelerometer of the present invention was less affected by temperature, while the measurement accuracy of the accelerometer of the control group fluctuated more.

[0153] Experimental results show that the accelerometer of the present invention is superior to the control group in terms of measurement accuracy and stability, and can better meet the needs of practical applications.

[0154] In summary, the design and fabrication of this high-sensitivity MEMS capacitive accelerometer based on a novel folded beam structure utilizes a variable cross-section serpentine beam structure. From the connection end with the matching beam to the connection end with the sensitive mass block, the cross-sectional width decreases. The wider portion at the connection end can withstand greater acceleration, ensuring structural stability; the narrower end has lower stiffness, making the sensitive mass block more prone to displacement under external acceleration, resulting in a more significant change in capacitance and improving detection sensitivity, thus satisfying both requirements.

[0155] Furthermore, in the design and manufacturing process of this high-sensitivity MEMS capacitive accelerometer based on a novel folded beam structure, the temperature compensation circuit plays a crucial role in the signal processing circuit. A temperature sensor integrated on the sensitive silicon structure monitors the temperature in real time. Since temperature changes affect the beam stiffness and capacitor plate size, thus affecting the capacitance value, the signal processed by the pre-amplifier, differential amplifier, and other circuits is corrected based on a pre-built temperature-drift compensation model to offset the influence of temperature factors, reduce temperature drift, and improve measurement accuracy. This solves the problem that traditional folded beams cannot simultaneously achieve high sensitivity and structural stability, as well as the impact of temperature drift on measurement accuracy.

[0156] The relevant modules involved in this system are all hardware system modules or functional modules that combine computer software programs or protocols with hardware in the prior art. The computer software programs or protocols involved in these functional modules are technologies known to those skilled in the art and are not improvements to this system. The improvement of this system lies in the interaction or connection between the modules, that is, in improving the overall structure of the system to solve the corresponding technical problems that this system aims to address.

[0157] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A high-sensitivity MEMS capacitive accelerometer based on a novel folded beam structure, characterized in that, include: The upper cover has an upper fixed capacitor plate, which includes an upper detection electrode and an upper matching electrode surrounding the upper detection electrode. The lower cover has a lower fixed capacitor plate, which includes a lower detection electrode and a lower matching electrode surrounding the lower detection electrode. A sensitive silicon structure disposed between an upper fixed capacitor plate and a lower fixed capacitor plate, the sensitive silicon structure comprising: A fixed frame is provided with stress relief grooves on each side, and a matching beam is formed between the stress relief grooves and the inner side of the fixed frame. A sensitive mass block located in the middle of a fixed frame; A folded beam is used to connect the sensitive mass block and the matching beam. The folded beam is a serpentine beam with a variable cross-section structure. The cross-sectional width of the variable cross-section structure gradually decreases from 25μm-35μm to 12μm-18μm from the end connected to the matching beam to the end connected to the sensitive mass block.

2. The high-sensitivity MEMS capacitive accelerometer based on a novel folded beam structure according to claim 1, characterized in that, The sensitive mass block adopts a multi-layer structure, including a top-layer mass block, an intermediate-layer mass block, and a bottom-layer mass block; The top layer mass block, the middle layer mass block, and the bottom layer mass block are connected as a whole by vertical connecting columns, the diameter of which is 8μm-12μm. The top mass block has an upper protruding electrode in the shape of a comb on the side opposite to the upper detection electrode, and the bottom mass block has a lower protruding electrode in the shape of a comb on the side opposite to the lower detection electrode. The comb teeth of the upper protruding electrode and the comb teeth of the upper detection electrode are interleaved, and the comb teeth of the lower protruding electrode and the comb teeth of the lower detection electrode are interleaved.

3. The high-sensitivity MEMS capacitive accelerometer based on a novel folded beam structure according to claim 2, characterized in that, The material of the top layer mass block, the middle layer mass block and the bottom layer mass block is monocrystalline silicon; The material of the vertical connecting column is silicon dioxide; The upper and lower protruding electrodes are made of doped polycrystalline silicon, and the doping concentration is [missing information]. .

4. The high-sensitivity MEMS capacitive accelerometer based on a novel folded beam structure according to claim 1, characterized in that, The surfaces of both the upper and lower fixed capacitor plates are covered with an insulating layer with a thickness of 200nm-500nm, and the insulating layer is made of silicon nitride. A metal electrode layer with a thickness of 500nm-1000nm is disposed on the insulating layer, and the material of the metal electrode layer is aluminum.

5. The high-sensitivity MEMS capacitive accelerometer based on a novel folded beam structure according to claim 1, characterized in that, The outer edge of the fixed frame has multiple anchor points for fixing the entire accelerometer to the external carrier. The fixed anchor point is circular in shape with a diameter of 50μm-100μm, and the material of the fixed anchor point is the same as that of the fixed frame, which is monocrystalline silicon.

6. The high-sensitivity MEMS capacitive accelerometer based on a novel folded beam structure according to claim 1, characterized in that, The upper and lower covers are connected to the sensitive silicon structure via silicon-silicon bonding; the silicon-silicon bonding method is as follows: Before bonding, the bonding surfaces of the upper cover, lower cover and sensitive silicon structure are pretreated by plasma cleaning, wet chemical etching and deionized water rinsing in sequence. The bonding temperature of the silicon-silicon bonding process is 830℃-870℃, the bonding pressure is 1.2MPa-1.8MPa, and the bonding time is 35min-45min.

7. The high-sensitivity MEMS capacitive accelerometer based on a novel folded beam structure according to claim 1, characterized in that, The accelerometer further includes a signal processing circuit electrically connected to the upper detection electrode, the lower detection electrode, the upper matching electrode, and the lower matching electrode, the signal processing circuit comprising: A preamplifier circuit with an amplification factor of 10-50 times is used to initially amplify the weak capacitance change signals output by the upper and lower detection electrodes. A differential amplifier circuit used to perform differential processing on the signal output from a preamplifier circuit to eliminate common-mode interference; A second-order low-pass Butterworth filter circuit is used to filter the signal output of a differential amplifier circuit to remove high-frequency noise. Its cutoff frequency is 1kHz-10kHz. An A / D conversion circuit used to convert analog signals output from a filter circuit into digital signals.

8. The high-sensitivity MEMS capacitive accelerometer based on a novel folded beam structure according to claim 7, characterized in that, The signal processing circuit also includes a temperature compensation circuit, which is used to compensate for the temperature drift of the accelerometer. The compensation method is as follows: The temperature of the sensitive silicon structure is measured by an integrated temperature sensor integrated on the sensitive silicon structure; The acceleration signal output by the signal processing circuit is corrected based on a pre-built temperature-drift compensation model. The temperature-drift compensation model is constructed as follows: the accelerometer is calibrated at different temperatures, the temperature and the corresponding acceleration drift are recorded, the temperature-drift compensation curve is obtained through polynomial fitting, and then the temperature-drift compensation model is established.

9. The method for fabricating a high-sensitivity MEMS capacitive accelerometer based on a novel folded beam structure as described in any one of claims 1 to 8, characterized in that, Includes the following steps: S1: The structural pattern of the fixed frame, sensitive mass block, folded beam, stress relief groove and matching beam is fabricated on a single crystal silicon wafer using deep ultraviolet lithography. S2: Ion implantation doping treatment is performed on the single-crystal silicon wafer with the structural pattern fabricated in S1; S3: On another single-crystal silicon wafer, electrode patterns for the upper and lower fixed capacitor plates are fabricated using conventional photolithography and magnetron sputtering processes. S4: On a single-crystal silicon wafer with the electrode pattern fabricated, a silicon nitride insulating layer is grown using a chemical vapor deposition process. S5: Perform silicon-silicon bonding between the monocrystalline silicon wafers treated in S2 and the monocrystalline silicon wafers of the upper and lower fixed capacitor plates treated in S4. S6: Process the monocrystalline silicon wafers of the upper and lower covers to match their size and shape with the bonded structure, and open windows at corresponding positions to connect with the sensitive silicon structure and the fixed capacitor plate; S7: Assemble the top cover, the bonded structure and the bottom cover, and connect them into one piece through silicon-silicon bonding process to form a complete accelerometer chip; S8: Package the accelerometer chip and signal processing circuit.

10. The method for fabricating a high-sensitivity MEMS capacitive accelerometer based on a novel folded beam structure according to claim 9, characterized in that, When fabricating the structural pattern in S1, a folded beam with a variable cross-section structure is formed using multiple photolithography and etching processes, specifically: First, the overall outline of the folded beam is created through a first photolithography and etching process; A special mask is made. In the mask pattern, the width of the folded beam gradually changes from the end connected to the matching beam to the end connected to the sensitive mass block in a stepped transition form. The number of steps and the step width are made according to the design requirements. The non-mask area of ​​the folded beam is then thinned through a second photolithography and etching process to form a variable cross-section structure.

Images