A bimodal hybrid mode magnetoelectric sensor, preparation and detection method
Patent Information
- Application Number
- CN202611102404.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-23
- Publication Date
- 2026-10-30
AI Technical Summary
[0004]然而,依据这种传统方式制备的磁电悬臂梁传感器,其磁电悬臂梁在U形模态工作时,会叠加沿悬臂梁长轴方向的机械振荡,导致检测精度的不稳定,且极易受到环境温度漂移、夹具装夹状态变化、激励信号幅值波动以及材料老化等因素的影响,导致在使用过程中出现检测信号的基线漂移和灵敏度衰减,长期稳定性难以保证,无法兼顾高灵敏度和高稳定性的检测需求
本发明,在磁电传感器的制备过程中通过在单个悬臂梁中协同配置几何尺寸、磁致伸缩层覆盖区域及固定端边界约束条件,使振动产生的U形模态与另一机械模态形成受控耦合,使其在频域上产生可分辨的第一混杂谐振峰和第二混杂谐振峰,从而将传统上被视为噪声来源的长轴叠加振荡转化为可设计的双峰混杂响应;在此基础上,根据有限元仿真或预实验获得的两个混杂峰应变分布,将磁致伸缩层覆盖区域设置为与应变分布相匹配的图形,使两个混杂峰均能有效参与磁电转换,再通过对磁致伸缩层沿预定方向施加磁场进行退火处理,使退火感生的磁各向异性与目标混杂模态的主应变方向或有效磁弹响应方向匹配,从而实现对一个混杂峰的选择性增强;在检测时,通过同时采集两个混杂峰的磁电响应并采用双峰差分、比值或归一化差分读出方式,可将环境温度漂移、夹具装夹状态变化、激励信号幅值波动及材料老化等因素引起的两个峰的共同变化作为共模项予以扣除,仅将两个峰的相对变化作为磁场检测信号,从而在根本上克服了传统单峰读出方案难以抑制共模漂移、长期稳定性差的技术缺陷,实现了高灵敏度与高稳定性的协同提升,同时退火方向与混杂模态应变场匹配的设计使得无需改变通用层叠结构即可选择性增强目标峰输出,在单个悬臂梁内实现两个可标定读出点,显著提高了磁场检测的稳定性和信号可辨识性。
Smart Images

Figure CN123009752A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of magnetoelectric sensor technology, specifically relating to a dual-peak hybrid mode magnetoelectric sensor, its preparation and detection methods. Background Technology
[0002] A magnetoelectric cantilever beam sensor is a magnetic field-sensitive device based on the coupling of the magnetostrictive and piezoelectric effects, typically composed of a magnetostrictive layer and a piezoelectric layer stacked together. Its working principle is as follows: the magnetic field to be measured causes strain in the magnetostrictive layer, which is transmitted to the piezoelectric layer through the interlayer interface, and then converted into voltage, charge, or impedance changes as outputs by the piezoelectric effect. Due to its advantages such as strong magnetoelectric conversion capability at room temperature, high sensitivity, fast response, ease of miniaturization, and compatibility with MEMS processes, this type of sensor has broad application prospects in fields such as biomedical magnetic field measurement (e.g., magnetocardiography and magnetoencephalography), geophysical exploration, military navigation, non-destructive testing, and weak magnetic environment monitoring.
[0003] The magnetoelectric cantilever beam sensor mainly consists of a substrate and a first electrode layer, a piezoelectric thin film layer, a magnetostrictive layer, and a second electrode layer. In the fabrication of magnetoelectric cantilever beam sensors, the U-shaped mode (i.e., the high-frequency mechanical mode in which the cantilever beam bends along its minor axis during vibration, resulting in a U-shaped deformation in the beam's cross-section or width) has attracted widespread attention due to its high quality factor and strong magnetoelectric response. In traditional fabrication processes, the long-axis oscillation accompanying the U-shaped mode is considered a harmful parasitic mode or noise source. Therefore, complex boundary reinforcement structures (such as increasing the thickness of the fixed end or designing asymmetric support beams) or precise mass loading tuning are typically introduced during fabrication to attempt to suppress or eliminate this oscillation component in the fabrication process, thereby achieving the goal of suppressing this superimposed oscillation.
[0004] However, the magnetoelectric cantilever beam sensor prepared according to this traditional method will have mechanical oscillations superimposed along the long axis of the cantilever beam when the magnetoelectric cantilever beam is working in U-shaped mode, which will lead to unstable detection accuracy. It is also very susceptible to the effects of environmental temperature drift, changes in fixture clamping status, fluctuations in excitation signal amplitude, and material aging, resulting in baseline drift and sensitivity decay of the detection signal during use. Long-term stability is difficult to guarantee, and it is impossible to meet the detection requirements of high sensitivity and high stability at the same time. Summary of the Invention
[0005] To address the problems existing in the prior art, the present invention aims to provide a bimodal hybrid magnetoelectric sensor, its fabrication and detection method, which transforms modes that are traditionally considered to be unwanted superimposed oscillations into a designable bimodal hybrid response. Common-mode drift is suppressed by bimodal differential or ratio readout, and mode-selective gain is achieved by utilizing the annealing direction, thereby improving the stability and signal distinguishability of magnetic field detection.
[0006] The technical solution of this invention is: A method for fabricating a bimodal hybrid magnetoelectric sensor includes the following steps: The distinguishable first and second hybrid resonance peaks generated when the U-shaped mode of the target cantilever beam is controlled to couple with another mechanical mode are identified. The first hybrid resonance peak characterizes the U-shaped strain response along the short axis of the cantilever beam, and the second hybrid resonance peak characterizes the strain or torsional induced shear strain response along the long axis of the cantilever beam. The other mechanical mode is any one or more combinations of bending modes, torsional modes, or boundary-induced modes adjacent to the frequency of the U-shaped mode, excluding the U-shaped mode. The geometric dimensions of the target cantilever beam are determined based on the first and second hybrid resonance peaks; and the coverage area pattern of the magnetostrictive layer that matches the strain distribution is determined according to the strain distribution of the first and second hybrid resonance peaks. Based on the geometry of the target cantilever beam and the coverage area pattern of the magnetostrictive layer, a first electrode layer, a piezoelectric thin film layer, a magnetostrictive layer, and a second electrode layer are sequentially deposited on the substrate. One end of the substrate is used as a fixed end, and the other end is used as a free end. The material of the substrate at the free end is removed to form a target cantilever beam that can generate a U-shaped mode when vibrating. The magnetostrictive layer is subjected to magnetic field annealing, and a magnetic field is applied along a predetermined direction and heated and kept at a certain temperature to induce or reconstruct magnetic anisotropy in the magnetostrictive layer, thereby obtaining a bimodal hybrid magnetoelectric sensor.
[0007] Preferably, when the coverage area of the magnetostrictive layer is set to a pattern that matches the strain distribution, the coverage area is one of full-length coverage, half-length coverage, coverage near the fixed end, coverage near the free end, central strip coverage, width-direction partition coverage, or multi-segment coverage. By changing the coverage area of the magnetostrictive layer, the magnetoelastic participation and inter-peak response ratio of the first hybrid resonance peak and the second hybrid resonance peak are adjusted.
[0008] Preferably, the frequency interval Δ between the first hybrid resonance peak and the second hybrid resonance peak is... f With center frequency f The ratio of 0 to 20% is the center frequency. f 0 represents the average frequency of the first hybrid resonance peak and the second hybrid resonance peak.
[0009] Preferably, when performing magnetic field annealing on the magnetostrictive layer, a magnetic field of 100 Oe to 3000 Oe is applied along the long axis direction, short axis direction, or direction forming a predetermined angle with the long axis of the cantilever beam, and the layer is held at 200℃ to 350℃ for 10 min to 90 min; if selective enhancement is desired for the first hybrid resonance peak dominated by U-shaped short axis strain, then annealing is performed by applying a magnetic field along the short axis direction; if selective enhancement is desired for the second hybrid resonance peak dominated by long axis strain or torsional induced shear strain, then annealing is performed by applying a magnetic field along the long axis direction or obliquely; wherein, the predetermined angle between the long axis and the magnetic field is 0 degrees to 90 degrees.
[0010] Preferably, before removing the substrate material below the free end, the substrate of the target cantilever beam is further subjected to local thinning, overall thinning, edge trimming, or local mass loading to adjust the frequency proximity and coupling strength between the U-shaped mode and the other mechanical mode.
[0011] Preferably, the piezoelectric thin film layer is made of AlN or AlScN and has a thickness of 0.5 μm to 3 μm; the magnetostrictive layer is made of FeCoSiB, FeGaB, or FeGa and has a thickness of 0.2 μm to 5 μm; the first electrode layer and the second electrode layer are each independently made of Mo, Pt, Au, Cu, Ti / Pt composite layer or Cr / Au composite layer and have a thickness of 50 nm to 300 nm.
[0012] A dual-peak hybrid mode magnetoelectric sensor, fabricated according to any of the above-described methods for fabricating a dual-peak hybrid mode magnetoelectric sensor, comprises a target cantilever beam structure formed by a substrate, a first electrode layer, a piezoelectric thin film layer, a magnetostrictive layer, and a second electrode layer.
[0013] Preferably, the length of the cantilever beam is 5mm to 30mm, the width is 1mm to 5mm, and the thickness of the substrate is 50μm to 500μm.
[0014] A detection method for a bimodal hybrid magnetoelectric sensor, used to implement a bimodal hybrid magnetoelectric sensor according to any one of the above claims, includes the following steps: The frequency response curve is obtained by performing a frequency scan on the dual-peak hybrid mode magnetoelectric sensor, and the first hybrid resonance peak and the second hybrid resonance peak are determined on the frequency response curve. The magnetoelectric response of the dual-peak hybrid mode magnetoelectric sensor after annealing in a predetermined direction magnetic field is calibrated. Based on the output enhancement degree, sensitivity and signal-to-noise ratio of the first hybrid resonance peak and the second hybrid resonance peak, one of the hybrid resonance peaks is determined as the high-gain detection channel and the other hybrid resonance peak is determined as the reference channel. At the resonant frequencies corresponding to the high-gain detection channel and the reference channel, an AC carrier magnetic field is applied simultaneously or in a time-division manner, and the DC bias magnetic field is adjusted so that the dual-peak hybrid mode magnetoelectric sensor operates at a pre-calibrated operating bias point. The output of the high-gain detection channel and the output of the reference channel are acquired at the working bias point, and the amplitude of the magnetic field to be measured or the change in the magnetic field to be measured is obtained by using bimodal differential readout or bimodal ratio readout. The high-gain detection channel is used to improve the magnetic field response strength and signal discernibility, while the reference channel is used to characterize or suppress common-mode drift caused by changes in temperature, clamping state, AC carrier magnetic field amplitude, or detection circuit gain.
[0015] Compared with the prior art, the bimodal magnetoelectric sensor, its preparation and detection method of the present invention have the following beneficial effects: This invention, in the fabrication of a magnetoelectric sensor, achieves controlled coupling between a U-shaped vibration mode and another mechanical mode by collaboratively configuring the geometric dimensions, magnetostrictive layer coverage area, and fixed-end boundary constraints within a single cantilever beam. This results in distinguishable first and second hybrid resonance peaks in the frequency domain, transforming the long-axis superimposed oscillations, traditionally considered a noise source, into a designable bimodal hybrid response. Based on this, and according to the strain distributions of the two hybrid peaks obtained through finite element simulation or pre-experimentation, the magnetostrictive layer coverage area is configured to match the strain distribution, ensuring both hybrid peaks effectively participate in the magnetoelectric conversion. Furthermore, by applying a magnetic field along a predetermined direction to the magnetostrictive layer for annealing, the annealed magnetic anisotropy is matched to the principal strain direction or effective magnetoelastic response direction of the target hybrid mode. This allows for selective enhancement of a hybrid peak. During detection, by simultaneously acquiring the magnetoelectric responses of two hybrid peaks and employing bi-peak differential, ratio, or normalized differential readout methods, the common changes of the two peaks caused by factors such as ambient temperature drift, changes in fixture clamping status, fluctuations in excitation signal amplitude, and material aging can be subtracted as common-mode terms. Only the relative changes of the two peaks are used as the magnetic field detection signal. This fundamentally overcomes the technical defects of traditional single-peak readout schemes, such as difficulty in suppressing common-mode drift and poor long-term stability. It achieves a synergistic improvement in high sensitivity and high stability. At the same time, the design of matching the annealing direction with the hybrid mode strain field allows for selective enhancement of the target peak output without changing the general stacked structure. Two calibrable readout points are achieved within a single cantilever beam, significantly improving the stability and signal identifiability of magnetic field detection. Attached Figure Description
[0016] Figure 1 This is a schematic diagram of a magnetoelectric cantilever beam in an embodiment of the present invention; wherein, 1 is a substrate; 2 is a first electrode layer; 3 is a piezoelectric thin film layer; 4 is a magnetostrictive layer; and 5 is a second electrode layer; Figure 2 The following is a simulation diagram of the vibration mode in an embodiment of the present invention; wherein, (a) is the simulation mode shape diagram of the first resonance peak, and (2) is the simulation mode shape diagram of the second resonance peak; Figure 3 This is a diagram showing the dual-modal test results in an embodiment of the present invention; Figure 4 This is a diagram showing the test results after annealing with a magnetic field applied along the minor axis in an embodiment of the present invention. Figure 5 This is a flowchart of the preparation method in an embodiment of the present invention. Detailed Implementation
[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.
[0018] Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this invention.
[0019] Furthermore, the technical solutions of the various embodiments of the present invention can be combined with each other, but only if they are feasible for those skilled in the art. If the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.
[0020] Example 1 See Figure 1 As shown, this embodiment provides a bimodal hybrid magnetoelectric sensor, which includes a substrate 1, a bottom electrode layer (first electrode layer 2), a piezoelectric thin film layer 3, a magnetostrictive layer 4, and a top electrode layer (second electrode layer 5). To achieve controlled coupling between the U-shaped mode and another mechanical mode within a single cantilever beam and to construct a bimodal hybrid response, this embodiment provides a method for fabricating a bimodal hybrid magnetoelectric sensor, the core steps of which are described below. Figure 5 As shown, it can be basically summarized as follows:
[0021] S1. Select substrate 1 material and clean substrate 1.
[0022] S2. Deposit a first electrode layer 2 on top of substrate 1.
[0023] S3. Deposit a piezoelectric thin film layer 3 above the first electrode layer 2.
[0024] S4. A magnetostrictive layer 4 is formed above the piezoelectric thin film layer 3. Based on the strain distribution of the first and second hybrid resonance peaks obtained from finite element simulation or pre-experiment, the coverage area of the magnetostrictive layer 4 is set to a pattern matching the strain distribution. The first and second hybrid resonance peaks are adjacent peaks with different frequencies that appear sequentially in the frequency sweep direction. The first hybrid resonance peak characterizes the U-shaped strain response along the short axis of the cantilever beam, and the second hybrid resonance peak characterizes the strain or torsional induced shear strain response along the long axis of the cantilever beam. The first and second hybrid resonance peaks are generated based on the controlled coupling between the U-shaped mode and another mechanical mode.
[0025] S5. A second electrode layer 5 is deposited above the magnetostrictive layer 4, which together with the substrate 1, the first electrode layer 2, the piezoelectric thin film layer 3 and the magnetostrictive layer 4 form a stacked structure.
[0026] S6. The stacked structure is shaped to form a cantilever beam structure, with one end of substrate 1 becoming a fixed end and the other end becoming a free end, so that the cantilever beam can generate a U-shaped mode during vibration; wherein, the geometry of the cantilever beam is configured to allow controlled coupling between the U-shaped mode and another mechanical mode, so as to form a distinguishable first hybrid resonance peak and a second hybrid resonance peak in the frequency response of the actually fabricated cantilever beam. The other mechanical mode is any one or more combinations of bending mode, torsional mode, or boundary-induced mode adjacent to the frequency of the U-shaped mode, other than the U-shaped mode.
[0027] S7. Remove the substrate 1 material below the free end to free the free end from the constraint so that the cantilever beam can vibrate cantilever with the fixed end as support.
[0028] S8. Perform magnetic field annealing on the magnetostrictive layer 4, apply a magnetic field along a predetermined direction and heat and keep warm to induce or reconstruct magnetic anisotropy in the magnetostrictive layer 4, so that the first hybrid resonance peak or the second hybrid resonance peak is selectively enhanced, and the final bimodal hybrid magnetoelectric sensor is prepared.
[0029] To more clearly illustrate the fabrication method of the bimodal hybrid magnetoelectric sensor, the following details the process in conjunction with specific design requirements: In this embodiment, the fabrication method of the bimodal hybrid magnetoelectric sensor, and its device structure, can be referred to Figure 1 As shown. The effective cantilever beam region of this device is preferably a long strip structure, with a length of 5mm to 30mm and a width of 1mm to 5mm. In this embodiment, the width of the cantilever beam is preferably 2mm, the length is preferably 24mm, and the thickness of the substrate 1 is preferably 50μm to 500μm. The fixed end is used for mechanical clamping and electrical signal extraction, and the free end is used to generate cantilever vibration.
[0030] See Figure 1 As shown, in an embodiment of the present invention, substrate 1 is first selected according to requirements and then cleaned. Substrate 1 can be selected from silicon, SOI, glass, or insulating substrate 1, and the thickness of substrate 1 can be 50μm~500μm. In this embodiment, a silicon wafer is selected as the structural substrate of the cantilever beam. Before entering the thin film deposition, the silicon wafer undergoes conventional cleaning and drying treatment, such as RCA cleaning, plasma cleaning, or organic solvent cleaning, to remove surface particles, organic contaminants, and moisture residues, ensuring stable adhesion of the subsequent metal electrode layer and piezoelectric thin film layer 3.
[0031] Furthermore, in an embodiment of the present invention, a first electrode layer 2 is deposited on top of the substrate 1. The pattern of the first electrode layer 2 is defined by photolithography, and Mo, Pt, Au, Cu, Ti / Pt, Mo / Pt, or a composite electrode thereof are formed using magnetron sputtering, electron beam evaporation, or other thin film deposition methods. The thickness of the first electrode layer 2 can be 50 nm to 300 nm. This electrode layer preferably also serves as a seed layer for the preferential orientation growth of piezoelectric thin films, but is not limited to signal extraction via low-resistivity conductive silicon. In this embodiment, a 100 nm thick Pt thin film is deposited on the silicon substrate surface using magnetron sputtering as part of the bottom electrode layer. The magnetron sputtering power of this Pt thin film is 30 W, and the working gas pressure is 3 mTorr. After deposition, a Si / Pt stacked structure is obtained.
[0032] See Figure 1 As shown, in an embodiment of the present invention, a piezoelectric thin film layer 3 is deposited above the first electrode layer 2. The material of the piezoelectric thin film layer 3 is AlN or AlScN, and the thickness is 0.5 μm to 3 μm. A C-axis oriented AlN or AlScN piezoelectric thin film layer 3, preferably 1 μm to 2 μm thick, is deposited on the first electrode layer 2. During the deposition process, the target-substrate distance, argon / nitrogen ratio, sputtering power, deposition gas pressure, and substrate 1 temperature are adjusted to obtain a higher piezoelectric coefficient and a lower dielectric loss. After deposition, the quality of the piezoelectric thin film can be confirmed by XRD rocking curves, capacitance, and dielectric loss tests. In this embodiment, an AlN piezoelectric thin film layer 3 with a thickness of 1.5 μm is prepared on the Pt bottom electrode layer. In this embodiment, the AlN piezoelectric thin film can be completed by a foundry process, and the specific sputtering power, target-substrate distance, gas ratio, deposition temperature, and other parameters are subject to the foundry process records. In this invention, the AlN layer serves to convert mechanical strain under hybrid modes into measurable electrical signals. Therefore, as long as it meets the requirements for thickness, insulation, orientation, and adhesion for piezoelectric readout, it can be used in this embodiment.
[0033] Depending on the electrode structure requirements, SiO2, Al2O3, Ta, Ti, Mo, or a composite transition layer thereof are formed on the piezoelectric layer. Electrode lead-out windows, insulating openings, and magnetic layer deposition areas are defined using photolithography and etching. This step controls interface stress, magnetic layer adhesion, and electrical isolation, and can serve as a means for subsequent mass loading and stress adjustment to regulate hybrid modes. In this embodiment, a window pattern is defined in the area where the bottom electrode needs to be led out or a test electrical connection needs to be formed using photolithography. After development, the sample is placed in hot phosphoric acid at 135°C with a mass fraction of 85% to thermally etch and open the AlN in the window area, exposing the underlying Pt bottom electrode in the predetermined region. After opening the window, the photoresist is removed and the sample is cleaned to obtain a Si / Pt / AlN stacked structure with electrode lead-out windows.
[0034] Furthermore, in an embodiment of the present invention, a magnetostrictive layer 4 is formed above the piezoelectric thin film layer 3. The magnetostrictive layer 4 is made of FeCoSiB, FeGaB, or FeGa, and has a thickness of 0.2 μm to 5 μm, preferably 0.8 μm to 2 μm. During deposition, an in-situ magnetic field can be selectively applied to preset the initial easy magnetization direction; after deposition, a patterned magnetostrictive layer 4 is obtained by peeling. Based on the strain distribution of the first and second hybrid resonance peaks obtained from finite element simulation or pre-experiment, the coverage area of the magnetostrictive layer 4 is set to a pattern matching the strain distribution. The coverage area can be one of full-length coverage, half-length coverage, coverage near the fixed end, coverage near the free end, central strip coverage, width-direction partition coverage, or multi-segment coverage. By changing the coverage area of the magnetostrictive layer 4, the magnetoelastic participation and inter-peak response ratio of the first and second hybrid resonance peaks are adjusted.
[0035] In this embodiment, a strip-shaped magnetic film region with a width of 2 mm and a length of 24 mm is defined on the surface of the AlN piezoelectric thin film by photolithography. This strip-shaped region coincides with the effective vibration region of the cantilever beam, forming a fully covered magnetic film structure. Subsequently, a FeCoSiB magnetostrictive thin film with a thickness of 1 μm is deposited by magnetron sputtering at a power of 30 W and a working pressure of 3 mTorr. After deposition, a 2 mm × 24 mm strip-shaped FeCoSiB magnetic functional layer is obtained by lift-off. The coverage area of the magnetostrictive layer 4 is set to full-length coverage, so that the magnetostrictive layer 4 can participate in strain-magnetic elastic coupling at both hybrid resonance peaks.
[0036] Furthermore, a second electrode layer 5 is deposited above the magnetostrictive layer 4, forming a stacked structure together with the substrate 1, the first electrode layer 2, the piezoelectric thin film layer 3, and the magnetostrictive layer 4. The second electrode layer 5 and test pads are formed by photolithography and deposition. The second electrode material can be Pt, Au, Mo, Cu, or a Cr / Au composite layer, with a thickness of 50 nm to 300 nm. The electrode pattern can be designed for single-end readout, differential readout, partitioned readout, or multi-pad readout, to separately acquire the amplitude, phase, or impedance response of the first and second mixed peaks. In this embodiment, a 100 nm thick Pt film is deposited immediately after the FeCoSiB magnetostrictive film using magnetron sputtering as the top electrode layer and magnetic film protective layer. The magnetron sputtering power of this Pt film is 30 W, and the working gas pressure is 3 mTorr. Since the top Pt layer is continuously prepared after FeCoSiB deposition, surface oxidation of the magnetic layer can be reduced, and subsequent electrical connections and test stability can be improved.
[0037] Furthermore, the stacked structure is shaped to form a cantilever beam structure, with one end of substrate 1 forming a fixed end and the other end forming a free end, enabling the cantilever beam to generate a U-shaped mode during vibration. The stacked structure is shaped by etching or cutting to form the cantilever beam structure. Specifically, the cantilever beam structure can be formed by deep silicon etching, back-side etching, dicing, laser cutting, or wet etching. In this embodiment, laser cutting is performed along the boundary of the magnetic material to match the outer boundary of the cantilever beam device with the boundary of the FeCoSiB strip magnetic layer, resulting in a single magnetoelectric cantilever beam device with a width of approximately 2 mm and a length of approximately 24 mm. This cutting method keeps the magnetic layer coverage area, the piezoelectric readout area, and the effective vibration area consistent, which is beneficial for obtaining two hybrid mode response peaks with similar output amplitudes.
[0038] Specifically, the geometry of the cantilever beam is configured to achieve controlled coupling between the U-shaped mode and another mechanical mode, thereby forming a distinguishable first and second hybrid resonance peak in the frequency response of the actually fabricated cantilever beam. Three types of geometries—weakly split, critically split, and strongly split—are preferably fabricated, with the frequency interval between the two hybrid peaks Δ... f With center frequency f The ratio of 0 is in the range of 0.2% to 20%, and is further preferred to be 0.5% to 8%. Among these, f 0 refers to the arithmetic mean (midpoint value) of the first hybrid resonance peak frequency and the second hybrid resonance peak frequency. In this embodiment, simulation confirms that the device has a distinct U-shaped vibration mode; when the frequency of the U-shaped mode is close to that of the adjacent structural mode, they hybridize and form two distinguishable resonance response peaks.
[0039] See Figure 2 As shown, in an embodiment of the present invention, Figure 2The simulation results of the vibration modes of the cantilever beam are shown. Specifically, the simulation mode shapes of the first and second resonance peaks of the U-mode of the bimodal hybrid magnetoelectric sensor with a size of 2 mm × 24 mm are shown. Among them, (a) is the simulation mode shape of the first resonance peak, and (2) is the simulation mode shape of the second resonance peak. A finite element model was established based on the above geometric dimensions and stacked structure, and the intrinsic frequencies, U-modes, and adjacent bending, torsion, or boundary-induced modes of the cantilever beam were simulated and analyzed. The 2 mm × 24 mm device was used as the basic model in the simulation. The actual silicon substrate thickness, piezoelectric layer thickness, metal electrode thickness, and magnetostrictive layer 4 thickness were set, and the boundary conditions of the fixed end were constrained. The simulation results are as follows. Figure 2 As shown, the device exhibits a distinct U-shaped vibration mode; when the frequencies of the U-shaped mode and the adjacent structural mode are close, they mix and form two distinguishable resonant response peaks. This simulation step is used to identify the structural origin of the bimodal split before fabrication, rather than simply attributing the bimodal to test fixtures, contact resistance, or random defects.
[0040] Furthermore, before removing the substrate 1 material below the free end, the back side of the cantilever beam is locally thinned, overall thinned, edge-trimmed, or locally mass-loaded to control the frequency proximity and coupling strength between the U-shaped mode and another mechanical mode. This step does not achieve the frequency gradient through the difference in clamping lengths of multiple sensing units, but rather controls the frequency proximity and coupling strength of the two modes within the same sensing unit.
[0041] The substrate material beneath the free end is removed to deconstrain the free end, allowing the cantilever beam to vibrate cantilevered with the fixed end as support. After thin film preparation, the free end of the cantilever beam is released, and warpage is reduced through low-temperature annealing, stress compensation layers, or geometric symmetry. After release, beam warpage, film integrity, and electrode continuity are tested to screen devices that meet the bimodal hybridization test conditions.
[0042] Furthermore, the magnetostrictive layer 4 undergoes magnetic field annealing. A magnetic field is applied along a predetermined direction, and the layer is heated and held at that temperature to induce or reconstruct magnetic anisotropy in the magnetostrictive layer 4, thereby selectively enhancing the first or second hybrid resonance peak. During the magnetic field annealing of the magnetostrictive layer 4, a magnetic field of 100 Oe to 3000 Oe is applied along the long axis, short axis, or at a predetermined angle of 0 to ~90 degrees to the long axis of the cantilever beam, and the layer is held at 200℃ to 350℃ for 10 min to 90 min. Preferred conditions are 250℃, 30 min, and 2000 Oe. If selective enhancement is desired for the first hybrid resonance peak dominated by U-shaped short-axis strain, annealing is performed by applying a magnetic field along the short axis; if selective enhancement is desired for the second hybrid resonance peak dominated by long-axis strain or torsional induced shear strain, annealing is performed by applying a magnetic field along the long axis or obliquely.
[0043] In this embodiment, the same device that has completed the initial test is subjected to magnetic field annealing. During annealing, a magnetic field of 2000 Oe is applied along the short axis of the device, and the device is held at 250°C for 30 min to induce or reconstruct magnetic anisotropy in the FeCoSiB magnetostrictive layer 4. Preferably, after the holding period, the temperature is lowered to room temperature or below the magnetic anisotropy stabilization temperature while maintaining the direction of the magnetic field, and then the sample is taken out for testing.
[0044] After annealing, the magnetoelectric frequency response of the device was tested again under the same or comparable clamping, electrical connection, magnetic field direction, and frequency sweep conditions as before annealing. The test results show that the first peak output is significantly enhanced, while the second peak output remains almost unchanged. This indicates that short-axis magnetic field annealing does not simply increase all resonant peaks of the entire device proportionally, but rather enables the hybrid modes that match specific strain distributions, magnetoelastic coupling directions, and magnetic anisotropy directions to obtain selective gain.
[0045] Post-fabrication calibration: After annealing, the VSM or MOKE hysteresis loop, magnetic anisotropy field Hk, coercivity Hc, saturation magnetization Ms, piezoelectric capacitance, impedance spectrum, and magnetoelectric output spectrum are measured respectively. If the two hybrid peaks are distinguishable in the frequency domain, and different annealing directions cause repeatable changes in the output ratio of the two peaks, then the fabrication of the bimodal hybrid sensor is complete.
[0046] This embodiment demonstrates that in a single 2 mm × 24 mm FeCoSiB / AlN / Pt / Si stacked cantilever beam device, the hybridization conditions of the U-shaped mode and its adjacent modes can be pre-observed through structural design and modal simulation, resulting in two bimodal responses with consistent output heights in the actual device. Furthermore, by applying a 2000 Oe magnetic field along the short axis and annealing at 250°C for 30 min, the first peak can be significantly enhanced while the second peak remains essentially unchanged, achieving mode-selective magnetoelastic gain through annealing programming.
[0047] Unlike schemes that rely on multiple magnetoelectric sensing units connected in series and change the cantilever length of each unit to form multiple resonant peaks or broaden the bandwidth, the core of this embodiment lies in the splitting of hybrid modes inside a single cantilever beam and the selective enhancement of one of the hybrid peaks after annealing. Its technical focus is not on changing the number of sensing units or forming a frequency gradient through clamping length, but on obtaining a distinguishable, tunable and readable bimodal magnetoelectric response through a single-device stacked structure, magnetic film covering method, U-shaped mode design and short-axis magnetic field annealing.
[0048] Example 2 See Figure 3 and Figure 4 As shown, this embodiment provides a detection method for a dual-peak hybrid mode magnetoelectric sensor, which is used to implement the dual-peak hybrid mode magnetoelectric sensor described in Embodiment 1. Figure 3The dual-mode test results of the dual-peak hybrid mode magnetoelectric sensor before annealing are shown, that is, the dual-mode output results of a single sensor before annealing enhancement. Figure 4 The test results of the sensor after annealing with a magnetic field along the short axis are shown. After annealing, one of the hybrid resonance peaks was selectively enhanced.
[0049] See Figure 3 The figure shows the test results of the magnetoelectric coupling coefficient of a 2 mm × 24 mm bimodal hybrid magnetoelectric sensor as a function of frequency. The optimal bias Hdc is 3 Oe. The magnetoelectric coupling coefficient is 107 at the first peak and 92 at the second peak. In an embodiment of the present invention, the cantilever beam of the bimodal hybrid magnetoelectric sensor is first frequency-scanned to determine the first and second hybrid resonance peaks in the frequency response. Specifically, the bimodal hybrid magnetoelectric sensor prepared in Example 1 is fixed in a test fixture, and the fixed end is kept in a stable mechanical constraint. The electrical signals of the first electrode layer (bottom electrode) and the second electrode layer (top electrode) are respectively extracted. An AC magnetic field is applied to the sensor and a frequency sweep test is performed. The changes in output voltage, output charge, or equivalent magnetoelectric response with frequency are recorded. The response peak that appears first on the low-frequency side or after the frequency sweep is defined as the first hybrid resonance peak, and the response peak that appears later on the high-frequency side or after the frequency sweep is defined as the second hybrid resonance peak.
[0050] See Figure 3 As shown, in an embodiment of the present invention, the magnetoelectric response curves of the first hybrid resonance peak and the second hybrid resonance peak are calibrated under different bias magnetic fields and different test magnetic fields. Specifically, within the sensor's operating frequency band, bias magnetic fields of different intensities (e.g., 0Oe, 50Oe, 100Oe, 200Oe, 500Oe, etc.) are applied, and a test alternating magnetic field of known amplitude is applied under each bias magnetic field. The output voltage amplitude, phase, or impedance change at the first and second hybrid resonance peaks are recorded to form the magnetoelectric response calibration curves for each of the two resonance peaks. These calibration curves are used for signal inversion and sensitivity calibration during subsequent actual detection.
[0051] See Figure 3 and Figure 4 As shown, in embodiments of the present invention, one or more of the following are selected as the detection result output modes: single-peak readout, bi-peak differential readout, bi-peak ratio readout, or normalized differential readout. Wherein:
[0052] Single-peak readout mode: Select either the first or second hybrid resonance peak, and output its amplitude change or frequency shift as the magnetic field detection signal. This mode is suitable for scenarios where detection stability requirements are not high and environmental factors change little.
[0053] Dual-peak differential readout mode: Simultaneously acquires the response signals of two mixed resonant peaks, subtracts the amplitude changes or frequency shifts of the two peaks, and outputs the difference as the magnetic field detection signal. This mode can effectively counteract common-mode interference such as ambient temperature drift, fixture clamping pressure changes, and excitation signal amplitude fluctuations.
[0054] Bimodal ratio readout mode: The ratio of the response amplitudes of the two mixed resonant peaks is taken and output as the magnetic field detection signal. This mode has a natural ability to suppress fluctuations in excitation amplitude and is insensitive to changes in sensor gain.
[0055] Normalized differential readout mode: The response signals of the two peaks are first normalized (e.g., divided by their respective zero-field reference values) before differential calculation is performed. This mode eliminates common-mode drift and corrects for systematic deviations introduced by the different frequency positions of the two peaks.
[0056] See Figure 3 and Figure 4 As shown, in an embodiment of the present invention, when the ambient temperature, clamping state, or excitation amplitude changes, the common change of the first hybrid resonance peak and the second hybrid resonance peak is subtracted as a common-mode term, and the relative change of the first hybrid resonance peak and the second hybrid resonance peak is used as a magnetic field detection signal or a self-calibration signal. Specifically, when the ambient temperature rises, causing the elastic modulus of the cantilever beam to decrease and the overall resonant frequency to shift downward, the two peaks will synchronously drift in frequency. This synchronous drift is subtracted as a common-mode term, while the difference in the response amplitude of the two peaks caused by the magnetic field is extracted as an effective magnetic field signal, thereby achieving self-compensation for temperature drift.
[0057] See Figure 3 and Figure 4 As shown, in embodiments of the present invention, depending on the application scenario, one of the enhanced first hybrid resonance peak and the second hybrid resonance peak is selected as a high-sensitivity channel, and the other is selected as a reference channel, or both the first and second hybrid resonance peaks are used simultaneously to expand the dynamic range. For example, when detecting weak magnetocardiogram or magnetoencephalogram signals, the annealed enhanced first hybrid resonance peak can be selected as the high-sensitivity channel to obtain optimal sensitivity, and the second hybrid resonance peak can be used as a reference channel for common-mode suppression. In scenarios where detection is required over a wide magnetic field range, the linear response intervals of the two peaks can be spliced together, with one peak responsible for low-field detection and the other for high-field detection, thereby expanding the dynamic range.
[0058] Test results are as follows Figure 3As shown, this bimodal hybrid magnetoelectric sensor exhibits two clearly distinguishable resonant response peaks in its frequency response, and the initial output heights of the two peaks are essentially the same. This result indicates that, before annealing enhancement, the sensor has already formed a stable bimodal split response, and both peaks can effectively participate in magnetoelectric conversion. The consistency of the initial output heights of the first and second hybrid resonant peaks is beneficial for subsequent comparison before and after annealing to determine whether magnetoelastic coupling selectively enhances a particular hybrid mode.
[0059] See Figure 4 The figure shows the test results of the magnetoelectric coupling coefficient as a function of frequency after annealing of a 2 mm × 24 mm bimodal hybrid mode magnetoelectric sensor. At the optimal bias Hdc of 0.75 Oe, the magnetoelectric coupling coefficient is 500 at the first peak and 155 at the second peak. In the embodiment of this invention, the output of the first hybrid resonant peak is significantly enhanced after annealing, while the output of the second hybrid resonant peak changes only slightly, or even remains almost unchanged. This indicates that short-axis magnetic field annealing provides selective gain to the hybrid mode dominated by U-shaped short-axis strain. Based on this, in practical detection, if it is necessary to detect weak magnetic field signals, the enhanced first hybrid resonant peak can be selected as the high-sensitivity channel, and the second hybrid resonant peak can be selected as the reference channel. The bimodal differential readout can suppress common-mode interference such as excitation amplitude fluctuations, temperature drift, and clamping pressure changes, effectively improving the long-term stability and signal-to-noise ratio of magnetic field detection.
[0060] The core of the dual-peak hybrid mode magnetoelectric sensor detection method described in this embodiment lies in fully utilizing the common-mode response characteristics of two hybrid resonant peaks within the same sensor to changes in external environmental factors. Through readout algorithms such as differential, ratio, or normalized differential, the relative change of the two peaks is used as the effective magnetic field detection signal, thereby significantly improving the long-term stability and environmental adaptability of magnetic field detection without increasing the hardware complexity of the sensor.
[0061] In summary, the detection method of the dual-peak hybrid mode magnetoelectric sensor of this embodiment can be executed in the following order: The first hybrid resonance peak f1 and the second hybrid resonance peak f2 are used as two magnetoelectric readout channels. Three types of quantities can be read from each peak:
[0062] Resonant frequencies: f1, f2 Peak output amplitude: V1, V2 Magnetoelectric coupling coefficient or sensitivity: α1=V1 / (Hac·t), α2=V2 / (Hac·t) Where Hac is the amplitude of the AC magnetic field used to test the magnetoelectric response, and Hdc is the DC bias magnetic field. t is the thickness of the piezoelectric layer.
[0063] Dual-peak determination: The frequency of the dual-peak hybrid mode magnetoelectric sensor is scanned to obtain the frequency response curve, and the first hybrid resonance peak and the second hybrid resonance peak are determined on the frequency response curve. The resonant frequency f1 and output V1 of the first hybrid resonance peak, and the resonant frequency f2 and output V2 of the second hybrid resonance peak are recorded respectively. Single-peak readout mode: Select either the first or second hybrid resonance peak, and output its amplitude change or frequency shift as the magnetic field detection signal. This mode is suitable for scenarios where detection stability requirements are not high and environmental factors change little.
[0064] Channel determination after annealing direction programming: The magnetoelectric response of the bimodal hybrid mode magnetoelectric sensor after annealing with a predetermined direction magnetic field is calibrated. Based on the output enhancement degree, sensitivity, and signal-to-noise ratio of the first and second hybrid resonance peaks, one hybrid resonance peak is determined as the high-gain detection channel, and the other hybrid resonance peak is determined as the reference channel. The high-gain detection channel is the hybrid resonance peak that obtains mode-selective gain after magnetic field annealing, and the reference channel is another hybrid resonance peak used to characterize common-mode drift. Single-peak readout mode: Selects the highest gain detection channel and outputs its amplitude change or frequency shift as the magnetic field detection signal. This mode is suitable for scenarios where detection stability requirements are not high and environmental factors change little.
[0065] Bimodal difference readout, bimodal ratio readout, or normalized difference readout: Specifically, when magnetic field annealing is performed along a first predetermined direction to enhance the magnetoelectric output of the first hybrid resonance peak relative to the second hybrid resonance peak, the first hybrid resonance peak is used as a high-gain detection channel, and the second hybrid resonance peak is used as a reference channel; when magnetic field annealing is performed along a second predetermined direction to enhance the magnetoelectric output of the second hybrid resonance peak relative to the first hybrid resonance peak, the second hybrid resonance peak is used as a high-gain detection channel, and the first hybrid resonance peak is used as a reference channel. Operating point and calibration curve acquisition: At the resonant frequencies corresponding to the high-gain detection channel and the reference channel, a small-amplitude AC carrier magnetic field is applied simultaneously or in a time-division manner, and the DC bias magnetic field is adjusted so that the dual-peak hybrid mode magnetoelectric sensor operates at a pre-calibrated operating bias point; a known calibration magnetic field is applied at the operating bias point, and the output Vm of the high-gain detection channel and the output Vr of the reference channel are recorded respectively to obtain the output-magnetic field calibration relationship of the high-gain detection channel and the reference channel; Wherein, when the object to be detected is a DC magnetic field or a low-frequency magnetic field, the frequency of the AC carrier magnetic field is equal to or close to the resonant frequency of the high-gain detection channel and / or the reference channel, the DC magnetic field or low-frequency magnetic field to be detected is superimposed on the DC bias magnetic field, and modulates the magnetoelectric output amplitude or phase corresponding to the AC carrier magnetic field. Common-mode drift calibration: Under conditions of no input magnetic field to be measured or a known input magnetic field to the reference field, at least one of the following is changed: temperature, clamping state, AC carrier magnetic field amplitude, or detection circuit gain. The output changes of the high-gain detection channel and the reference channel are recorded respectively to obtain the correspondence between the two and the common-mode drift. The common-mode drift includes the output change of the non-test magnetic field caused by changes in temperature, clamping state, AC carrier magnetic field amplitude, or detection circuit gain, which simultaneously acts on the high-gain detection channel and the reference channel. Dual-peak differential readout: When detecting the magnetic field to be measured, the real-time output Vm of the high-gain detection channel and the real-time output Vr of the reference channel are acquired, and the differential compensation output Vd is obtained according to the formula Vd=(Vm-Vm0)-β(Vr-Vr0); where Vm0 is the output of the high-gain detection channel in the reference state, Vr0 is the output of the reference channel in the reference state, and β is the drift ratio coefficient obtained according to the common-mode drift calibration; according to the pre-calibrated Vd-H relationship, the differential compensation output Vd is converted into the amplitude of the magnetic field to be measured or the change in the magnetic field to be measured; Bimodal ratio readout: Alternatively, when detecting the magnetic field to be measured, the bimodal ratio signal R is calculated based on the real-time output Vm of the high-gain detection channel and the real-time output Vr of the reference channel. The bimodal ratio signal R is Vm / Vr, or a normalized ratio (Vm / Vm0) / (Vr / Vr0). According to the pre-calibrated RH relationship, the bimodal ratio signal R is converted into the amplitude of the magnetic field to be measured or the change in the magnetic field to be measured. Output results: The output is the amplitude or change of the magnetic field to be measured obtained by the bimodal differential readout or the bimodal ratio readout; wherein, the high-gain detection channel is used to improve the magnetic field response intensity and signal discriminability, and the reference channel is used to suppress common-mode drift. The common-mode drift is subtracted or weakened by the bimodal differential readout or the bimodal ratio readout, thereby improving the stability of magnetic field detection.
[0066] In the above detection method, the first hybrid resonance peak and the second hybrid resonance peak serve as two calibrable magnetoelectric readout channels. The magnetic field annealing direction is used to perform mode-selective gain programming on one of the hybrid resonance peaks, making it a high-gain detection channel; the other hybrid resonance peak serves as a reference channel, used to reflect common-mode drift caused by changes in temperature, clamping state, AC carrier magnetic field amplitude, or detection circuit gain.
[0067] By using bimodal differential readout or bimodal ratio readout, it is possible to distinguish the effective changes caused by the magnetic field under test in the high-gain detection channel from the drift changes shared by both channels. This allows for the suppression of common-mode drift while maintaining a high magnetic field response, thereby improving the stability and identifiability of the magnetic field signal under test.
[0068] Compared with the prior art, the present invention has the following beneficial effects: 1. The modal components, which are traditionally considered to be unwanted superimposed oscillations, are transformed into a designable bimodal hybrid response, turning "noise" into "signal". Two calibrable readout points are obtained within a single cantilever beam, improving the design freedom and information capacity of the device.
[0069] 2. By using bimodal differential, ratio, or normalized differential readout methods, the influence of common-mode factors such as excitation amplitude fluctuations, temperature drift, and clamping pressure changes can be effectively eliminated, significantly improving the long-term stability and detection reliability of the sensor.
[0070] 3. By matching the annealing direction with the hybrid modal strain distribution, selective gain of the target peak can be achieved without changing the general stacked structure, providing a flexible means for on-demand control of sensor performance.
[0071] 4. Unlike broadband solutions that rely on the superposition of frequency gradients from multiple devices, this invention can achieve a bimodal response within a single cantilever beam, reducing system complexity and manufacturing costs, and facilitating device miniaturization and integration.
[0072] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.
Claims
1. A method for fabricating a bimodal hybrid magnetoelectric sensor, characterized in that, Includes the following steps: The distinguishable first and second hybrid resonance peaks generated when the U-shaped mode of the target cantilever beam is controlled to couple with another mechanical mode are identified. The first hybrid resonance peak characterizes the U-shaped strain response along the short axis of the cantilever beam, and the second hybrid resonance peak characterizes the strain or torsional induced shear strain response along the long axis of the cantilever beam. The other mechanical mode is any one or more combinations of bending modes, torsional modes, or boundary-induced modes adjacent to the frequency of the U-shaped mode, excluding the U-shaped mode. The geometric dimensions of the target cantilever beam are determined based on the first and second hybrid resonance peaks; and the coverage area pattern of the magnetostrictive layer that matches the strain distribution is determined according to the strain distribution of the first and second hybrid resonance peaks. Based on the geometry of the target cantilever beam and the coverage area pattern of the magnetostrictive layer, a first electrode layer, a piezoelectric thin film layer, a magnetostrictive layer, and a second electrode layer are sequentially deposited on the substrate. One end of the substrate is used as a fixed end, and the other end is used as a free end. The material of the substrate at the free end is removed to form a target cantilever beam that can generate a U-shaped mode when vibrating. The magnetostrictive layer is subjected to magnetic field annealing, and a magnetic field is applied along a predetermined direction and heated and kept at a certain temperature to induce or reconstruct magnetic anisotropy in the magnetostrictive layer, thereby obtaining a bimodal hybrid magnetoelectric sensor.
2. The method for fabricating a bimodal hybrid magnetoelectric sensor according to claim 1, characterized in that, When the coverage area of the magnetostrictive layer is set to a pattern that matches the strain distribution, the coverage area can be one of full-length coverage, half-length coverage, coverage near the fixed end, coverage near the free end, central strip coverage, width-direction partition coverage, or multi-segment coverage. By changing the coverage area of the magnetostrictive layer, the magnetoelastic participation and inter-peak response ratio of the first hybrid resonance peak and the second hybrid resonance peak can be adjusted.
3. The method for fabricating a bimodal hybrid magnetoelectric sensor according to claim 1, characterized in that, The frequency interval Δ between the first hybrid resonance peak and the second hybrid resonance peak f With center frequency f The ratio of 0 to 20% is the center frequency. f 0 represents the average frequency of the first hybrid resonance peak and the second hybrid resonance peak.
4. The method for fabricating a bimodal hybrid magnetoelectric sensor according to claim 1, characterized in that, When performing magnetic field annealing on the magnetostrictive layer, a magnetic field of 100 Oe to 3000 Oe is applied along the long axis direction, short axis direction, or direction forming a predetermined angle with the long axis of the cantilever beam, and the layer is held at 200℃ to 350℃ for 10 min to 90 min. If selective enhancement is desired for the first hybrid resonance peak dominated by U-shaped short axis strain, annealing is performed by applying a magnetic field along the short axis direction. If selective enhancement is desired for the second hybrid resonance peak dominated by long axis strain or torsional induced shear strain, annealing is performed by applying a magnetic field along the long axis direction or obliquely. The predetermined angle between the long axis and the target long axis is 0 degrees to 90 degrees.
5. The method for fabricating a bimodal hybrid magnetoelectric sensor according to claim 1, characterized in that, Before removing the substrate material below the free end, the substrate of the target cantilever beam is locally thinned, overall thinned, edge trimmed, or locally mass-loaded to adjust the frequency proximity and coupling strength between the U-shaped mode and the other mechanical mode.
6. The method for fabricating a bimodal hybrid magnetoelectric sensor according to claim 1, characterized in that, The piezoelectric thin film layer is made of AlN or AlScN and has a thickness of 0.5 μm to 3 μm; the magnetostrictive layer is made of FeCoSiB, FeGaB, or FeGa and has a thickness of 0.2 μm to 5 μm; the first electrode layer and the second electrode layer are each independently made of Mo, Pt, Au, Cu, Ti / Pt composite layer or Cr / Au composite layer and have a thickness of 50 nm to 300 nm.
7. A bimodal hybrid magnetoelectric sensor, prepared according to the method for preparing a bimodal hybrid magnetoelectric sensor according to any one of claims 1-6, characterized in that, The target cantilever beam structure is formed by a substrate, a first electrode layer, a piezoelectric thin film layer, a magnetostrictive layer, and a second electrode layer.
8. A dual-peak hybrid mode magnetoelectric sensor according to claim 7, characterized in that, The length of the cantilever beam is 5mm to 30mm, the width is 1mm to 5mm, and the thickness of the substrate is 50μm to 500μm.
9. A detection method for a bimodal hybrid magnetoelectric sensor, used to implement the bimodal hybrid magnetoelectric sensor according to any one of claims 7-8, characterized in that, Includes the following steps: The frequency response curve is obtained by performing a frequency scan on the bimodal hybrid magnetoelectric sensor, and the first hybrid resonance peak and the second hybrid resonance peak are determined on the frequency response curve. The magnetoelectric response of the dual-peak hybrid mode magnetoelectric sensor after annealing in a predetermined direction magnetic field is calibrated. Based on the output enhancement degree, sensitivity and signal-to-noise ratio of the first hybrid resonance peak and the second hybrid resonance peak, one of the hybrid resonance peaks is determined as the high-gain detection channel and the other hybrid resonance peak is determined as the reference channel. At the resonant frequencies corresponding to the high-gain detection channel and the reference channel, an AC carrier magnetic field is applied simultaneously or in a time-division manner, and the DC bias magnetic field is adjusted so that the dual-peak hybrid mode magnetoelectric sensor operates at a pre-calibrated operating bias point. The output of the high-gain detection channel and the output of the reference channel are acquired at the working bias point, and the amplitude of the magnetic field to be measured or the change in the magnetic field to be measured is obtained by using bimodal differential readout or bimodal ratio readout. The high-gain detection channel is used to improve the magnetic field response strength and signal discernibility, while the reference channel is used to characterize or suppress common-mode drift caused by changes in temperature, clamping state, AC carrier magnetic field amplitude, or detection circuit gain.
