Relaxation ferroelectric single crystal rod nondestructive cutting positioning method and system

Abstract

The application discloses a kind of relax ferroelectric single crystal rod nondestructive cutting positioning method and system, it is related to relax ferroelectric single crystal material processing field, this method includes: to reference single crystal rod along axial continuous slice determination electrical property, establish the benchmark atlas of axial position and electrical property;Take the measured performance of at least two axial positions of the single crystal rod to be processed, compare the deviation with the corresponding value in benchmark atlas to judge applicability;If applicable, according to target performance, retrieve benchmark atlas to obtain theoretical cutting interval, then combine measured value and reference value to convert theoretical position into actual cutting coordinates, and cut the single crystal rod to be processed.The application realizes the nondestructive, high-precision positioning cutting of relax ferroelectric single crystal rod, significantly reduces material loss and cost, while effectively compensating for individual axial drift error, ensuring the accurate hit of target electrical property and wafer consistency.

Description

Technical Field

[0001] This application relates to the field of processing relaxor ferroelectric single crystal materials, and in particular to a non-destructive cutting and positioning method and system for relaxor ferroelectric single crystal rods. Background Technology

[0002] Relaxor ferroelectric single crystals are a class of high-performance functional crystal materials. Among them, lead magnesium niobate-lead titanate (Pb(Mg1 / 3Nb2 / 3)O3-PbTiO3, abbreviated as PMN-PT) solid solution single crystals are particularly noteworthy for their excellent piezoelectric properties (piezoelectric constant d). 33 With a strength exceeding 2000 pC / N, far surpassing the 200 pC / N to 600 pC / N of traditional PZT ceramics, it is widely recognized as a core material for next-generation ultrasonic transducers, underwater acoustic detection, and precision actuation devices. Since Shrout et al. of Pennsylvania State University reported the morphotropic phase boundary (MPB) effect of PMN-PT single crystals in 1997, this material has rapidly become a research hotspot in the international functional crystal field and has gradually achieved industrial application.

[0003] Currently, the industry primarily uses the Bridgman method (crucible lowering method) to grow large-size PMN-PT single crystals. This method involves slowly lowering the crucible in a temperature field, causing the melt to solidify directionally from bottom to top, ultimately obtaining a cylindrical single crystal rod. However, during crystal growth, due to the Ti... 4+ With Mg 2+ / Nb 5+ Difference in segregation coefficient in melt (k) Ti ≈1.2, k Mg ≈0.8, k Nb (≈0.9), a significant compositional gradient exists along the axial direction of the crystal rod. As the Ti content evolves axially, the crystal inevitably traverses the MPB region, leading to a decrease in the piezoelectric constant d. 33 Dielectric constant ε r The core performance indicators exhibit a highly nonlinear spatial distribution along the axial direction. The axial width of the MPB transition zone is typically only 5mm~15mm, but within this region, d 33 It can reach 2 to 3 times that of the single-phase region, making it the most valuable "golden section" in device manufacturing.

[0004] To obtain information on the performance distribution inside a single crystal rod, existing technologies mainly employ the following two types of testing schemes: Category 1: Sparse Sampling Test Method. Only a small number of samples are cut from a very few discrete locations (spaced a few centimeters apart) at the head, tail, or middle of the single crystal rod. After grinding, electrode fabrication, and polarization, the actual measurement of d is performed. 33 ε rParameters such as these are then used. Subsequently, these discrete data points are used to estimate the overall performance distribution of the rod through linear interpolation or empirical formulas.

[0005] The second method is the in-rod full-scale testing method. This involves densely slicing each single crystal ingot (e.g., 5mm-10mm slices), testing each slice individually, and then creating a dedicated performance spectrum for that single crystal ingot. While this method yields relatively complete data, it is a destructive test with significant slicing losses, and each ingot requires independent repeated testing, resulting in extremely low efficiency.

[0006] The aforementioned prior art has the following technical problems: (1) Insufficient spatial resolution makes it impossible to capture nonlinear abrupt changes in phase transition. The sampling interval of sparse sampling is on the order of centimeters, which is far greater than the width of the MPB transition region (5mm~15mm), resulting in fewer than 2 sampling points in the transition region, which cannot meet the minimum requirement of the sampling law for nonlinear reconstruction (at least 3~5 points). Linear interpolation further masks the real performance "abrupt peaks" and cannot form a continuous and real performance evolution spectrum.

[0007] (2) Data from a single rod cannot be reused across rods, resulting in high testing costs. Although full-scale testing of each rod yields complete data, the data is only valid for that rod, and other rods in the same batch require repeated destructive testing. For single crystal rods priced at tens of thousands of yuan, testing losses account for more than 30% of the material cost, severely restricting industrialization efficiency.

[0008] (3) Cutting is done based on position rather than performance, resulting in extremely low material utilization. Due to the lack of high-resolution performance evolution maps, subsequent cutting can only be done blindly based on fixed lengths or empirical positions. The performance obtained after cutting is the actual performance, and it is impossible to pre-set target performance and cut as needed. Once the actual performance deviates from the estimated value, a large number of waste wafers are generated. Under the current model, the material utilization rate of expensive single crystals is less than 30%, and the performance dispersion of devices in the same batch is large, making it difficult to guarantee the yield. Summary of the Invention

[0009] The purpose of this application is to provide a non-destructive cutting and positioning method and system for relaxor ferroelectric single crystal rods, which can realize non-destructive and high-precision positioning and cutting of relaxor ferroelectric single crystal rods, significantly reduce material loss and cost, and effectively compensate for individual axial drift errors, ensuring accurate hit of target electrical properties and wafer consistency.

[0010] To achieve the above objectives, this application provides the following solution: In a first aspect, this application provides a non-destructive cutting and positioning method for relaxor ferroelectric single crystal rods, including: The reference single crystal rod is continuously sliced ​​along the axial direction, and the electrical performance reference values ​​of each slice at the corresponding axial position are measured to establish a reference spectrum between the axial position of the reference single crystal rod and the electrical performance reference values; the reference single crystal rod and the single crystal rod to be processed are relaxor ferroelectric single crystal rods grown in the same batch, with the same growth process formula and the same growth equipment parameters. Obtain measured electrical properties of the single crystal rod to be processed at at least two axial positions, and determine the applicability of the reference spectrum to the single crystal rod to be processed based on the deviation between the measured electrical properties and the electrical property reference values ​​at the corresponding axial positions in the reference spectrum. If the reference spectrum is applicable to the single crystal rod to be processed, then according to the target electrical performance index, the corresponding theoretical axial position is retrieved in reverse from the reference spectrum to obtain the theoretical target cut-off interval; Based on the theoretical target cut-off interval, the measured electrical performance value, and the electrical performance reference value, the theoretical axial position within the theoretical target cut-off interval is converted into the actual mechanical cutting coordinates of the single crystal rod to be processed, and the single crystal rod to be processed is cut according to the actual mechanical cutting coordinates.

[0011] Secondly, this application provides a non-destructive cutting and positioning system for relaxor ferroelectric single crystal rods, comprising: The benchmark establishment module is used to continuously slice the reference single crystal rod along the axial direction, measure the electrical performance reference value of each slice at the corresponding axial position, and establish a benchmark spectrum between the axial position and electrical performance reference value of the reference single crystal rod; the reference single crystal rod and the single crystal rod to be processed are relaxor ferroelectric single crystal rods grown in the same batch, with the same growth process formula and the same growth equipment parameters. The applicability determination module is used to obtain the measured electrical properties of the single crystal rod to be processed at at least two axial positions, and to determine the applicability of the reference spectrum to the single crystal rod to be processed based on the deviation between the measured electrical properties and the electrical property reference values ​​at the corresponding axial positions in the reference spectrum. The target retrieval module is used to retrieve the corresponding theoretical axial position in the reference spectrum according to the target electrical performance index if the reference spectrum is applicable to the single crystal rod to be processed, so as to obtain the theoretical target interception interval. The cutting module is used to convert the theoretical axial position within the theoretical target cutting interval into the actual mechanical cutting coordinates of the single crystal rod to be processed, based on the theoretical target cutting interval, the measured electrical performance value, and the electrical performance reference value, and to cut the single crystal rod to be processed according to the actual mechanical cutting coordinates.

[0012] According to the specific embodiments provided in this application, this application has the following technical effects: This application abandons the high loss and low efficiency of traditional trial cutting. First, by constructing a reference spectrum under the same furnace batch, process, and equipment parameters, and introducing a deviation correction and applicability judgment mechanism based on a small number of measured axial position values, only a limited number of test points are needed to achieve accurate mapping of the axial electrical properties of the entire ingot, thereby significantly reducing the number of destructive slices, significantly saving expensive relaxor ferroelectric single crystal materials, and effectively reducing processing material loss and production costs. Second, this method uses the reference spectrum to reverse retrieve the theoretical axial position corresponding to the target electrical properties, and combines the theoretical coordinates with the measured deviation to convert them into actual mechanical cutting coordinates. This can effectively compensate for axial position drift caused by individual ingot growth differences or clamping alignment errors, ensuring that the electrical properties of the cut wafers accurately hit the target indicators, greatly improving the electrical consistency and process yield of the final product. Furthermore, the introduced applicability prediction mechanism can proactively identify and avoid the risk of misuse of the reference spectrum by abnormal ingots, enhancing the reliability and robustness of the entire cutting process. Furthermore, this application does not rely on complex online full-length scanning devices; it can be completed using only conventional electrical testing and simple conversion. The process is simple and easy to operate, making it very suitable for industrial mass production and continuous production. It can significantly shorten the production cycle and improve overall processing efficiency. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a schematic flowchart of a non-destructive cutting and positioning method for a relaxor ferroelectric single crystal rod provided in an embodiment of this application.

[0015] Figure 2 This is another schematic flowchart illustrating a non-destructive cutting and positioning method for a relaxor ferroelectric single crystal rod provided in an embodiment of this application.

[0016] Figure 3 This is a schematic diagram of the functional modules of a non-destructive cutting and positioning system for a relaxor ferroelectric single crystal rod provided in an embodiment of this application. Detailed Implementation

[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0019] In one exemplary embodiment, such as Figure 1 and Figure 2 As shown, it includes the following steps 101 to 105.

[0020] Step 101: Continuously slice the reference single crystal rod along the axial direction, measure the electrical performance reference value of each slice at the corresponding axial position, and establish a reference spectrum between the axial position of the reference single crystal rod and the electrical performance reference value.

[0021] The reference single crystal rod and the single crystal rod to be processed are relaxor ferroelectric single crystal rods grown in the same batch, with the same growth process formula and the same growth equipment parameters. The effective total processing length of the equal-diameter section of the reference single crystal rod and the single crystal rod to be processed is L (unit: mm).

[0022] In a specific application example, step 101 includes steps 11 to 13.

[0023] Step 11: The reference single crystal rod is continuously sliced ​​along the axial direction with uniform thickness. Specifically, from the beginning to the end of the uniform diameter section of the reference single crystal rod, a full continuous slice is made along a section perpendicular to the growth direction. The determination of the slice thickness d must simultaneously satisfy three constraints: (1) Sampling law constraint: During the axial growth of relaxor ferroelectric single crystals (such as PMN-PT), the Ti content exhibits a gradient distribution due to the segregation effect, causing the crystal structure to cross quasi-isomorphic phase boundaries. According to material growth kinetics, the axial physical length of the phase transition region is usually between 5 mm and 15 mm. To accurately reconstruct the nonlinear performance abrupt change in this region, it is necessary to ensure that at least 3 to 5 measured data points are obtained within the transition region. If d > 3 mm, the sampling points are too sparse (less than 2 points), which will cause the spline interpolation to lose its nonlinear capture capability and obscure the true phase boundary coordinates.

[0024] (2) Material utilization constraints: If d < 0.5 mm, the cutting edge loss of multi-wire cutting (usually 0.1 mm to 0.3 mm) will account for too high a proportion of the material, resulting in the ineffective waste of expensive single crystals.

[0025] (3) Critical balance: Considering the dual constraints of physical resolution and material utilization, the value range of d is 0.5mm≤d≤3mm. This application preferably uses 2mm, which is the critical balance point obtained between the above constraints. For a typical transition zone width (10mm~15mm), a thickness of 2mm can ensure 5~7 effective sampling points (including the boundary points of the transition zone) and control the blade edge loss ratio to within 10%.

[0026] During the cutting process, the starting end of the equal-diameter section is taken as the absolute zero point (X=0). The serial number i of each wafer and its physical axial coordinates from the zero point are strictly calibrated and recorded. (Unit: mm) Establish a unique correspondence between the wafer entity and its physical coordinates.

[0027] Step 12: After standardizing, grinding, cleaning, and drying the continuous sequence of wafers, conductive contact electrodes are prepared on both sides of the wafers (preferably by magnetron sputtering gold plating, but vacuum evaporation gold plating or low-temperature silver paste screen printing can also be used, suitable for different cost budgets and adhesion requirements). The wafer with the coated electrodes is placed in an insulating medium (such as a silicone oil bath) for room temperature polarization. The relationship between the polarization electric field strength Ep and the room temperature coercive field Ec is: Ep = α·Ec, where α is a safety factor, 1.5 ≤ α ≤ 3.

[0028] For PMN-PT single crystals (Ec is typically 0.2kV / mm~0.5kV / mm), the absolute value of Ep ranges from 0.3kV / mm to 1.5kV / mm, preferably 1.0~1.5kV / mm. After polarization for 15 minutes, the crystals are left to stand at room temperature for 24 hours for aging.

[0029] With piezoelectric constant d 33 As the main retrieval index, dielectric constant ε r As an auxiliary verification indicator, the actual measurement procedure and calibration are strictly performed in accordance with GB / T 3389 "Test Methods for Performance of Piezoelectric Ceramic Materials" or the current IEEE piezoelectric standard. The d-value of the central region of each wafer is measured. 33 With ε r As the axial coordinate X i The representative value.

[0030] Step 13: Using axial position as the abscissa and electrical performance reference value as the ordinate, a continuous nonlinear reference curve is constructed through interpolation fitting to obtain the reference spectrum.

[0031] Specifically, using the physical axial coordinate X i The x-axis represents the piezoelectric constant d. 33 With dielectric constant ε rUsing the vertical axis, a primary retrieval baseline curve S(X) and an auxiliary verification curve T(X) are established. A high-resolution baseline curve with nonlinear evolution is generated using cubic spline interpolation, B-spline interpolation, or Akima interpolation, which is suitable for scenarios with high data noise or uneven sampling point distribution. Unless otherwise specified, S(X) will be used as the operating object in the following text.

[0032] Step 102: Obtain the measured electrical properties of the single crystal rod to be processed at at least two axial positions, and determine the applicability of the reference spectrum to the single crystal rod to be processed based on the deviation between the measured electrical properties and the electrical property reference values ​​at the corresponding axial positions in the reference spectrum.

[0033] In a specific application example, a "micro-scale verification, spectrum reuse" strategy is adopted for the single crystal rod to be processed. The total number of verification wafers for the entire rod is controlled between 2 and 5. Step 102 includes steps 21 to 23.

[0034] Step 21: Cut a verification piece (thickness d) from the beginning (X=0), midpoint, and end (X=L) of the constant-diameter section of the single crystal rod to be processed. Measure the electrical performance parameters (piezoelectric constant and / or dielectric constant) of each verification piece to obtain the measured electrical performance values ​​of the beginning, midpoint, and end of the constant-diameter section of the single crystal rod to be processed. The electrical performance parameters include the piezoelectric constant and / or dielectric constant.

[0035] This application allows for multi-wire cutting, water-guided laser cutting, or diamond wire sawing when slicing reference single crystal rods and single crystal rods to be processed, which is suitable for single crystals of different hardness or specific cutting precision requirements.

[0036] Step 22: Calculate the relative deviation between the measured electrical performance value and the reference electrical performance value at the corresponding axial position in the reference spectrum, and obtain the spectrum deviation factor: ;in, This is the spectral deviation factor. These are the measured electrical properties of the initial end of the constant-diameter section of the single crystal rod to be processed. This serves as a reference value for the electrical properties at the starting point of the corresponding equal-diameter section in the baseline spectrum. The measured electrical properties at the midpoint of the single crystal rod to be processed are shown. This is the reference value for electrical performance at the midpoint of the baseline spectrum.

[0037] Step 23: If the spectrum deviation factor is less than or equal to a preset deviation threshold If the reference spectrum is applicable to the single crystal rod to be processed, then the reference spectrum is determined to be applicable to the single crystal rod to be processed; otherwise, the reference spectrum is determined to be inapplicable to the single crystal rod to be processed.

[0038] The physical meaning is: the arithmetic mean of the relative performance deviations between the single crystal rod to be processed and the reference single crystal rod at two characteristic positions, the head and the midpoint, which is used to quantify the overall consistency of the axial performance distribution of the two rods.

[0039] like (Preferred) If the reference spectrum is determined to be applicable to the single crystal rod to be processed, the calibration process in step 103 is initiated; if If it is identified as a heterogeneous growth rod, a unique map is created for it, and the process returns to step 101. The reasonable range for this value is 0.10 to 0.20, with 0.15 being preferred. This threshold is an empirical critical value determined based on the statistical law of single crystal rod growth fluctuations in the same furnace.

[0040] Step 103: If the reference spectrum is applicable to the single crystal rod to be processed, then according to the target electrical performance index, the corresponding theoretical axial position is retrieved in reverse from the reference spectrum to obtain the theoretical target cut-off interval.

[0041] Based on the processing requirements of downstream devices, the target electrical performance indicators are set as follows: (such as piezoelectric constant d) 33 target range Searching on the baseline spectrum S(X) for conditions... All axial coordinates of the conditions constitute the theoretical target intercept interval. The search methods include the following three: (1) Equivalence search: Given a single target value Solve Since the axial properties of relaxor ferroelectric single crystals typically exhibit a non-monotonic distribution (with rising and falling edges before and after crossing the MPB), the equation may have multiple solutions. In this case, the solution closest to the MPB peak coordinate is taken as the center, and a neighborhood interval (e.g., ±2 mm) is extended to both sides as... .

[0042] (2) Interval retrieval: Given a target interval Extract the continuous interval X that S(X) falls within this interval as If two or more discontinuous intervals satisfy the interval condition, then the longest continuous interval containing the global maximum (i.e., the MPB peak neighborhood) is taken as the interval. If the target performance clearly points to the single-phase region, then the shortest continuous interval that matches the device thickness requirement should be selected.

[0043] (3) Extreme value retrieval: Extract the coordinates of the maximum / minimum values ​​of S(X) to locate the center of the MPB peak region or single-phase stable region.

[0044] Step 104: Based on the theoretical target cut-off interval, the measured electrical performance value, and the electrical performance reference value, convert the theoretical axial position within the theoretical target cut-off interval into the actual mechanical cutting coordinates of the single crystal rod to be processed, and cut the single crystal rod to be processed according to the actual mechanical cutting coordinates.

[0045] In a specific application example, step 104 includes steps 41 to 48.

[0046] Step 41: Based on the measured electrical properties of the starting end and the tail end of the uniform diameter section of the single crystal rod to be processed, and the reference electrical properties of the corresponding axial position in the reference spectrum, establish the mapping relationship between the theoretical axial position and the actual mechanical coordinates of the single crystal rod to be processed.

[0047] Specifically, firstly, based on the measured electrical properties of the starting end and the tail end of the uniform diameter section of the single crystal rod to be processed, and the reference electrical properties at the corresponding axial positions in the reference spectrum, the growth rate fluctuation compensation coefficient is calculated: ;in, This is the growth rate fluctuation compensation coefficient. These are the measured electrical properties of the initial end of the constant-diameter section of the single crystal rod to be processed. This serves as a reference value for the electrical properties at the starting point of the corresponding equal-diameter section in the baseline spectrum. These are the measured electrical properties of the tail end of the single crystal rod to be processed. This is the reference value for electrical properties at the corresponding tail end in the baseline spectrum. This represents the change in the reference value for electrical performance. This represents the change in the measured value of electrical performance.

[0048] The ratio reflecting the performance gradient between the single crystal rod to be processed and the reference single crystal rod: This indicates that the performance gradient of the single crystal rod to be processed is more gradual (the growth rate is slower). This indicates a steeper gradient (faster growth rate). If ( If the preset minimum value is 50 pC / N, it indicates that the performance gradient of the single crystal rod to be processed is too gentle and the linear scaling model is not applicable. Therefore, a dedicated spectrum is created for it, and the process returns to step 101.

[0049] Then, the geometric zero-point offset is calculated based on the axial zero-point deviation between the single crystal rod to be processed and the reference single crystal rod: ;in, This is the geometric zero-point offset, reflecting the zero-point shift caused by mechanical clamping. If the clamping and positioning of the single crystal rod to be processed and the reference single crystal rod are consistent, then... , These are the actual mechanical coordinates of the starting end of the constant-diameter section of the single crystal rod to be processed. The absolute zero coordinates of the reference spectrum ( ).

[0050] Finally, based on the growth rate fluctuation compensation coefficient and the geometric zero-point offset, a mapping relationship between the theoretical axial position and the actual mechanical coordinates of the single crystal rod to be processed is established: ;in, The theoretical axial position, These are the actual machine coordinates.

[0051] Step 42, calculate the theoretical target cutoff interval. Transition zone to quasi-isomorphic phase boundary Shortest distance: .in, For the shortest distance, To determine the starting position of the interval for the theoretical target, To determine the endpoint of the interval for the theoretical target, The starting position of the quasi-isomorphic phase boundary transition zone. This is the endpoint of the transition zone between phases of the same type.

[0052] The quasi-isomorphic phase boundary transition region is the region in the reference spectrum that satisfies S'(X)|≥λ; where S'(X) is the first derivative of the reference spectrum, X is the axial position, and λ is a set derivative threshold.

[0053] Specifically, when the target performance corresponds to the MPB transition region, the first derivative S'(X) with respect to S(X) is calculated to identify the slope abrupt change region. A derivative threshold λ is set for region division: when |S'(X)|≥λ, and the axial length of the continuous coverage of this state is ≥d (i.e., not less than one sampling interval), it is determined to be a quasi-isomorphic phase boundary transition region (MPB region); when |S'(X)|<λ, it is determined to be a stable single-phase region.

[0054] Set derivative threshold ,in, Let be the absolute value of the global maximum first derivative of the curve. This represents the phase transition sensitivity coefficient. The physical meaning is: the relative judgment ratio based on the global maximum slope. When When the threshold is too low, gentle performance fluctuations will be misjudged as phase transitions; when... If the threshold is too high, the edge region of the MPB transition zone will be missed and classified as a stable single-phase region. Therefore... The reasonable value range is 0.05~0.30. Verification has been performed on multiple batches of PMN-PT single crystals. At that time, the MPB axial range identified by the algorithm had the highest degree of agreement with the measured phase boundary of XRD / dithermal spectrum, and was therefore selected as the preferred range.

[0055] Step 43: If the shortest distance is greater than the set distance threshold, the theoretical axial position within the theoretical target interception interval is converted into the actual mechanical cutting coordinates of the single crystal rod to be processed according to the mapping relationship.

[0056] Step 44: If the shortest distance is less than or equal to a set distance threshold (10mm), then it is determined that the theoretical target interception interval falls into the adjacent region of the quasi-isomorphic phase boundary transition region, and the peak position of the quasi-isomorphic phase boundary transition region is correspondingly located on the single crystal rod to be processed. A verification wafer is cut from the sample, and the measured peak coordinates of the single crystal rod to be processed are determined based on the measured electrical properties of the verification wafer. .

[0057] Step 45: Calculate the local nonlinear displacement between the measured peak coordinates and the peak position: .

[0058] Step 46, if the absolute value of the local nonlinear displacement... Less than or equal to the preset displacement threshold Then, based on the mapping relationship, the theoretical axial position within the theoretical target intercept interval is converted into the actual mechanical cutting coordinates of the single crystal rod to be processed. Preferably... .

[0059] Step 47: If the absolute value of the local nonlinear displacement is greater than the preset displacement threshold, the mapping relationship is locally nonlinearly corrected based on the local nonlinear displacement and the peak position of the quasi-isomorphic phase boundary transition zone, and the theoretical axial position within the theoretical target intercept interval is converted into the actual mechanical cutting coordinates of the single crystal rod to be processed based on the mapping relationship after local nonlinear correction.

[0060] Specifically, the mapping relationship after local nonlinear correction is as follows: ;in, This is a local nonlinear displacement. The peak position of the quasi-isomorphic phase boundary transition region. The half-width of the quasi-isomorphic phase boundary transition zone, The Gaussian attenuation correction term ensures that: the closer to the MPB region, the greater the weight of the local displacement correction; and that when far from the MPB region, automatic regression linear calibration occurs.

[0061] Step 48: Cut the single crystal rod to be processed according to the actual mechanical cutting coordinates.

[0062] Specifically, the interval is selected based on the theoretical target. Mapped to the actual mechanical coordinate range This guides the cutting equipment to extract the wafer oriented within this range, ensuring that the measured performance of the obtained wafer meets the preset target electrical performance indicators. Require.

[0063] Step 105: If the reference spectrum is not applicable to the single crystal rod to be processed, then the single crystal rod to be processed is determined to be a heterogeneous growth rod, and the single crystal rod to be processed is used as a reference single crystal rod to establish a reference spectrum.

[0064] The inventors of this application, through long-term research on relaxor ferroelectric single crystal growth and device fabrication, discovered that multiple single crystal rods grown under the same batch and process conditions exhibit highly similar axial performance evolution patterns. The differences primarily manifest as overall curve shifts and expansions, rather than complete changes in local morphology. This discovery provides a physical basis for "single-rod testing and multi-rod reuse." Therefore, this application selects one single crystal rod from the same batch as a "reference single crystal rod," establishing a "physical position-core performance" baseline map of the entire rod using 2mm high-resolution continuous slices. For other "single crystal rods to be processed" in the same batch, only a small number of verification slices need to be cut from both ends for double-end calibration. The axial coordinates can then be retrieved in reverse based on the target performance requirements using the baseline map, achieving precise on-demand cutting based on the required performance. This method compresses the destructive testing from "full measurement per rod" to "full measurement of one rod + micro measurement of multiple rods". At the same time, it ensures the nonlinear capture accuracy and positioning accuracy of the MPB phase transition region through mathematical means such as cubic spline interpolation, first derivative threshold determination and Gaussian decay correction.

[0065] In addition, this application provides three embodiments to illustrate a non-destructive cutting and positioning method for relaxor ferroelectric single crystal rods.

[0066] Example 1: On-demand truncation of MPB high-performance region based on interval retrieval.

[0067] Experimental materials: Two PMN-0.30PT single crystals grown by the Bridgman method in the same batch were selected: reference single crystal rod A (L=80mm, diameter 25mm) and single crystal rod B to be processed (L=80mm).

[0068] Target performance requirement: Downstream transducer devices require a piezoelectric constant d. 33 High-performance wafers ∈ [2000, 2500] pC / N.

[0069] Step 1: High-resolution continuous slicing of the reference single crystal rod. Taking reference single crystal rod A as the object, set the slice thickness d=2mm, and use a diamond multi-wire cutting machine to perform full continuous slicing, for a total of 40 slices. With the beginning of the constant diameter section as the absolute zero point X=0, establish the sequence number i and the axial coordinate X. i The correspondence (X)i =(i-1)×d).

[0070] Step 2: Standardized Preparation and Performance Testing. Forty wafers were thinned to 1.0 mm by grinding and ultrasonic cleaning, followed by magnetron sputtering of double-sided gold-plated electrodes. They were then placed in a 25°C silicone oil bath and subjected to an Ep = 1.2 kV / mm (α = 2.4, corresponding to Ec ≈ 0.5 kV / mm) for 15 minutes. After polarization, they were allowed to stand for 24 hours for aging.

[0071] Determine the central region d according to GB / T3389 33 Some data are shown in Table 1.

[0072] Table 1

[0073] Step 3: Constructing a baseline map and reverse retrieval. Using X... i x-coordinate, d 33 Using y as the ordinate, a baseline curve S(X) is generated using cubic spline interpolation.

[0074] Reverse spectrum retrieval (range retrieval): Setting target electrical performance indicators The theoretical target cutoff interval is obtained by searching for a continuous X interval on S(X) that satisfies this condition. .

[0075] Phase boundary identification (optimization verification): Take the first derivative of S(X), |S'(X)| max =580 pC / N / mm (located at X≈37 mm). Taking β=0.15, λ=87 pC / N / mm. The MPB transition region is determined to be 34.5 mm~39.5 mm, which highly overlaps with the interval retrieval results, confirming the finding. It is located in the high-performance zone of MPB.

[0076] Step 4: Reverse positioning and on-demand cutting of the single crystal rod B to be processed.

[0077] Preliminary assessment of the applicability of the spectra: Verification pieces (d=2mm) were cut from the single crystal rod B at X=0 and X=40mm respectively, and measured... , Corresponding benchmark value , .in, This is the reference value for the piezoelectric constant at X=0. This is the reference value for the piezoelectric constant at X=40. The measured value of the piezoelectric constant at X=0 is given. The value is the measured piezoelectric constant at X=40. Determine the applicable reference spectrum.

[0078] Two-end linear calibration and coordinate mapping: Actual measurements at both ends of the single crystal rod B to be processed: , Reference single crystal rod corresponding value , . K≈1.038 indicates that the performance gradient of the single crystal rod to be processed is basically consistent with that of the reference single crystal rod, and no significant scaling correction is required.

[0079] Geometric zero-point offset: , Therefore .

[0080] Establish mapping relationship: .

[0081] Local nonlinearity verification: The minimum distance |35.2-39.5|=4.3mm ≤ 10mm between the theoretical target cutoff interval [35.2, 41.8]mm and the MPB center (37mm) triggers reinforcement calibration. Cut the verification piece and measure the actual d 33 =2460pC / N, interpolation determined . |δ|=0.2mm≤2mm, accepting two-ended linear calibration results. .

[0082] Precise cut-out as needed: The target range is mapped to actual mechanical coordinates: Starting end: 1.038×35.2+0.3=36.8mm; Ending end: 1.038×41.8+0.3=43.7mm. The equipment is guided to cut within the 36.8mm~43.7mm range, yielding 3 wafers. Retest d. 33 The values ​​are 2340, 2420, and 2180 pC / N, respectively, all falling within the target range [2000, 2500] pC / N.

[0083] Experimental results: A closed loop of "input target performance range → output cutting coordinates" was achieved, and 100% of the three wafers met the requirements. Requirements: Edge wear accounts for 10.5%, and material utilization is significantly better than the traditional method.

[0084] Example 2: Location of Specific Performance Points Based on Equivalence Search Experimental material: PMN-0.30PT single crystal rod C, L=80mm, from the same batch as in Example 1.

[0085] Target performance requirements: The device needs d 33 =2200±50pC / N wafer (equivalent search), i.e. target range [2150,2250]pC / N.

[0086] Steps 1-3: Reference spectrum reuse and reverse retrieval directly adopt the reference spectrum S(X) established in Example 1 (based on L=80mm reference single crystal rod A).

[0087] Reverse spectrum retrieval (equivalence retrieval): Setting target electrical performance indicators Solving for S(X) = 2200, we get: Take the neighborhood interval ±2mm, that is... As the target of interception.

[0088] Step 4: Reverse positioning of the single crystal rod C to be processed.

[0089] Preliminary assessment of the applicability of the spectrum: Verification was performed at X=0 and X=40mm on the single crystal rod C to be processed, and the measurements were... , Corresponding benchmark value , (Interpolated from S(X)). The determination is applicable.

[0090] Two-end linear calibration and coordinate mapping: Actual measurements at both ends of the single crystal rod to be processed: , Reference values ​​for single crystal rods (extrapolated from the spectrum after being converted according to the length ratio): , . K=1.212>1, indicating that the performance gradient of this rod is gentler than that of the reference single crystal rod, and coordinate scaling correction is required.

[0091] Geometric zero-point offset: Therefore .

[0092] Establish mapping relationship: .

[0093] Local nonlinearity verification: When the distance between the theoretical target cutoff interval [34.2, 38.2] mm and the MPB center is ≤10 mm, reinforcement calibration is triggered. Cut the verification piece and measure the actual d 33 =2150pC / N, interpolation was used to determine the measured peak coordinates. .

[0094] The rod exhibits significant local nonlinear displacement, and the original deviation of the measured peak value from the theoretical peak value is: |δ|=6.2mm>2mm, triggering local nonlinear correction. The theoretical width of the MPB transition region is approximately 5mm, σ=2.5mm. Correction formula: The key coordinate corrections are shown in Table 2.

[0095] Table 2

[0096] Precise cropping as needed: cropping within the 44.2mm~51.3mm range. Additional measurements were performed at 46mm and 48mm for verification. Actual measured d... 33 The values ​​are 2180 pC / N and 2240 pC / N respectively, both meeting the requirement of 2200 ± 50 pC / N, and the d at 48 mm... 33 =2240pC / N is the peak neighborhood, which corresponds to the target isopleths.

[0097] Experimental results: K=1.212 accurately quantifies the gentleness of the gradient of C in the single crystal rod to be processed relative to the reference single crystal rod. If K=1 is used directly without introducing the K coefficient (i.e., ignoring the gradient difference), the target isopleth point will be located at 36.2mm, and the actual value at this position is d. 33 Only about 1680pC / N (low-performance rhombohedral phase region) will cause the entire batch of wafers to be scrapped. The economics of "full mapping of single rod and micro reuse of multiple rods" have been verified: only 4 verification wafers + target cutting are needed for the single crystal rod C to be processed, without the need to repeat 40 full slicing.

[0098] Example 3: Extreme accuracy truncation of MPB peak center based on extreme value retrieval (ΔX significant offset case).

[0099] Test material: PMN-0.30PT single crystal rods from the same batch, D, L=60mm, were selected. During the clamping of this rod, significant zero-point offset occurred due to wear of the clamping shims. The user required an extremely narrow high-performance region of ±2mm from the center of the MPB peak value.

[0100] Target performance requirement: Extreme value retrieval—extract the coordinates of the maximum value of S(X) and locate the MPB peak area.

[0101] Steps 1-3: Construction of Dedicated Short Rod Patterns. Dedicated baseline patterns were constructed using reference single crystal rods E (L=60mm) from the same furnace, with d=2mm, totaling 30 pieces. The MPB transition region was measured to be located between 32mm and 38mm, with a central peak X=35mm, |S'(X)| max =620 pC / N / mm. β = 0.20 (close to the MPB center, fast evolution rate, upper limit), λ = 124 pC / N / mm.

[0102] Graph Reverse Search (Extremum Search): Extract the coordinates of the maximum value of S(X) to obtain the coordinates of the theoretical peak value. The target intercept range is set to ±2mm of the peak value, i.e. .

[0103] Step 4: Reverse positioning of the single crystal rod D to be processed.

[0104] Preliminary determination of the applicability of the spectrum: The deviation factor γ of the verification sample was 0.08 ≤ 0.15, indicating applicability.

[0105] Two-end linear calibration and coordinate mapping: Actual measurement of both ends D of the single crystal rod to be processed: , Reference value for E in single crystal rod: , . With K≈1, the performance gradient is basically the same as that of the reference single crystal rod, with no significant scaling.

[0106] Geometric zero-point offset: (There is a significant clamping offset of +2.8mm), therefore ΔX = 2.8mm.

[0107] Establish mapping relationship .

[0108] Local nonlinearity verification: The theoretical target cutoff interval [33,37] mm completely falls within the MPB neighborhood (distance from the peak ≤ 10 mm), triggering enhanced calibration. Cut the verification piece and measure the actual d 33 =2320pC / N. Interpolation was performed on the verification piece and adjacent pieces to determine the measured peak coordinates. . |δ| = 3.0 mm > 2 mm, triggering Gaussian local nonlinear correction. The full width at half maximum (FWHM) of the MPB transition region is σ = (38 - 32) / 2 = 3 mm. Correction formula: The endpoint corrections for the target interval are shown in Table 3.

[0109] Table 3

[0110] Precise cut-out as needed: Two target wafers were precisely cut out within the [37.4, 41.4] mm range. Full performance testing was performed after cutting. 33 The values ​​were 2340 pC / N and 2280 pC / N, respectively, both within the MPB high-performance platform (reference peak 2350 pC / N). Metallurgical microscopy confirmed that the maximum deviation between the physical cutting location and the MPB center was ±0.8 mm, better than the ±2 mm specification.

[0111] Test Results: Key Verification of Zero-Point Offset Compensation: If ΔX=2.8mm is not detected and directly applied... The cutting will actually cut to the 35.8mm~39.8mm range, with the tail end falling into the tetragonal low-performance region (at this position d). 33 (Approximately 1,980 pC / N). Introducing ΔX correction avoids the scrapping of an entire expensive single crystal.

[0112] Dynamic threshold β adaptability verification: In this embodiment, β=0.20 (upper limit) was used, and a narrow transition zone with a width of only 6mm was successfully identified; if a fixed threshold (such as β=0.10) is used, the transition zone will be misjudged as 8mm wide, causing the target interception interval to be mixed into the edge of the single phase region, thus verifying the scientific value of the dynamic threshold.

[0113] This application also provides three comparative examples to verify the validity of this application.

[0114] Comparative Example 1: Existing sparse sampling test method (three-point linear interpolation).

[0115] Experimental method: A PMN-0.30PT single crystal rod F (L=80mm) from the same batch as in Example 1 was selected. Using the existing sparse sampling method, only one piece was cut from each of the head (X≈0mm), middle (X≈40mm), and tail (X≈80mm), and the d values ​​were measured. 33 The values ​​are 1200 pC / N, 2300 pC / N, and 2500 pC / N, respectively. Assuming that the performance evolves linearly along the axial direction, the MPB peak is estimated to be located at X≈35 mm through linear interpolation. Five slices are then cut directly from this location within a ±5 mm range (30 mm~40 mm).

[0116] Experimental results: 5 wafers measured d 33 The corresponding values ​​are 1680 pC / N, 1920 pC / N, 2100 pC / N, 2380 pC / N, and 2420 pC / N. Only two of these meet the high-performance requirements (d). 33 (≥2000pC / N), with a material utilization rate of only 40%.

[0117] The peak position estimated by linear interpolation (35mm) deviates by 2mm from the actual MPB peak center (37mm). Furthermore, due to the failure to capture nonlinear abrupt changes, the two top pieces (30mm~34mm) actually fall into the low-performance rhombohedral phase region, resulting in ineffective losses of expensive single crystals.

[0118] In Example 1 of this application, 5 to 7 measured points are obtained in the transition zone by high-resolution sampling of 0.5 mm to 3 mm and cubic spline interpolation. The positioning error is controlled within ±1.2 mm, and the effective utilization rate is close to 100%.

[0119] Comparative Example 2: Existing bar-by-bar full-scale testing method.

[0120] Experimental method: PMN-0.30PT single crystal rod G (L=80mm) from the same batch as in Example 2 was selected. The existing full-scale testing method for each rod was adopted: the entire rod was divided into 40 dense slices with d=2mm. Each slice was ground, electrode fabricated, polarized, and d was measured. 33 After creating a unique graph for the rod, it was cut and truncated.

[0121] Experimental results: The testing process consumed 40 wafers, with a cumulative cutting edge loss of approximately 8mm, accounting for 10% of the rod length. Although the MPB region could be accurately located after establishing the dedicated spectrum, this spectrum was only effective for single crystal rod G and could not be reused for other rods in the same batch. Single-rod testing costs were eliminated, and a complete single crystal rod was destroyed.

[0122] In Embodiment 2 of this application, the single crystal rod C to be processed only requires 4 verification wafers + target cutting, eliminating the need for repeated full-scale slicing of 40 wafers. The cost of a one-time full-scale slicing of reference single crystal rod A is amortized among multiple rods in the same furnace, reducing the cost by approximately 90%, without damaging the entire single crystal rod to be processed.

[0123] Comparative Example 3: Within the scheme of this application - without calibration, the coordinates of the reference single crystal rod are directly applied.

[0124] Experimental Method: The same single crystal rod C to be processed as in Example 2 (L=80mm, known K=1.212, ΔX=-0.5mm, δ=6.2mm after calibration in steps 4.2~4.3) was selected. The calibration process was deliberately omitted, and K=1, ΔX=0, and δ=0 were directly set. The theoretical coordinates of the reference single crystal rod A from Example 1 were then used. (corresponding to d) 33 =2200pC / N) is directly used as the cutting coordinate (i.e., X) of the single crystal rod to be processed. actual =36.2mm), cut 3 pieces within a ±2mm range at this position.

[0125] Experimental results: 3 wafers measured d 33 The values ​​were 1620 pC / N, 1580 pC / N, and 1650 pC / N, respectively, all far below the target value of 2200 pC / N.

[0126] Cause Analysis: The single crystal rod C to be processed, after double-ended calibration, showed K=1.212>1, indicating that its performance gradient was gentler than that of the reference single crystal rod, and that there was a local nonlinear displacement of δ=6.2mm. Directly applying the calibration without calibration resulted in cutting into the low-performance rhombohedral phase region before the MPB transition region, causing the measured performance of the cut wafer to deviate entirely from the preset values. The effective utilization rate of materials is 0%.

[0127] In Example 2 of this application, the target contour point is accurately mapped to 49.2 mm using K-coefficient scaling correction (1.212 times), zero-point offset compensation (-0.5 mm), and Gaussian local nonlinearity correction (6.2 mm), resulting in wafers that 100% meet the 2200±50 pC / N requirement. This comparative example demonstrates the necessity of two-end linear calibration and local nonlinearity correction.

[0128] In summary, the beneficial effects of this application include at least the following: (1) On-demand performance-oriented cutting achieves closed-loop control of "target performance input → cutting coordinate output". Through the introduced spectrum reverse retrieval mechanism, it allows for cutting based on target electrical performance indicators. Using the retrieval criteria, the corresponding axial coordinates are looked up in reverse on the baseline spectrum S(X) to form the theoretical target intercept interval. By combining coordinate mapping and correction, the theoretical target interception interval is transformed into actual mechanical cutting coordinates, realizing the reverse mapping and closed-loop control from performance requirements to physical coordinates. This breaks through the traditional blind processing mode of "cutting first and then measuring" or "cutting according to a fixed length," and directly determines the processing based on the performance requirements of downstream devices (such as d). 33 (≥2000pC / N) Reverse search the benchmark spectrum and extract the corresponding axial interval to make the cutting result accurately match the target performance.

[0129] (2) By combining the following three parts, the performance mutation of relaxor ferroelectric single crystal at the quasi-isomorphic phase boundary can be accurately captured, and the positioning error can be controlled within the millimeter level (within ±2mm), which improves the positioning accuracy and nonlinear capture capability.

[0130] Scientific sampling frequency: The established slice thickness of 0.5mm≤d≤3mm conforms to the sampling law constraint, ensuring that at least 3 to 5 measured points are obtained in the phase transition region of 5mm to 15mm, providing a data foundation for high-quality curve reconstruction.

[0131] Advanced mathematical model: Cubic spline interpolation is used to generate nonlinear evolution curves, and the slope abrupt changes are identified by combining the first derivative threshold λ. Compared with the traditional linear extrapolation method, cubic spline interpolation uses a cubic polynomial to connect every two adjacent sampling points, ensuring the continuity of the first and second derivatives of the curve, and more realistically restoring the nonlinear evolution of physical properties.

[0132] Gaussian decay correction algorithm: Introducing a local nonlinear correction term based on Gaussian distribution. This results in a greater correction weight the closer the cutting point is to the MPB peak; when it is far away, the weight decays exponentially and automatically reverts to linear calibration, effectively compensating for the nonlinear displacement at the phase boundary.

[0133] (3) Significantly improved the utilization rate of comprehensive materials and reduced the overall loss of expensive single crystals. It breaks through the high material loss mode caused by the destructive slicing of each single crystal rod in the traditional "full-scale testing of each rod". This application adopts the strategy of "full-scale mapping of one rod and micro-scale verification of multiple rods", which only performs high-resolution full-scale slicing on one reference single crystal rod. The remaining single crystal rods to be processed in the same batch only need 2-5 verification slices to complete the accurate positioning, thereby increasing the comprehensive material utilization rate of the entire batch of single crystal rods to more than 90%, which greatly saves expensive single crystal materials.

[0134] Specifically, by constraining both "physical resolution" and "material utilization," a critical equilibrium point of 2mm is selected to achieve critical equilibrium optimization. This thickness ensures 5-7 effective sampling points (including boundary points) within the transition zone while keeping the cutting edge loss ratio below 10%. A "one-rod mapping, multi-rod reuse strategy" is adopted, introducing a spectrum reuse and calibration mechanism. Considering the high one-time loss of a single crystal rod, applicability pre-judgment, double-end linear calibration, and local nonlinear correction ensure that the reference spectrum can be accurately applied to multiple single crystal rods to be processed in the same batch. For the single crystal rods to be processed, there is no need for repeated, expensive full-scale slicing; only a very small amount of verification wafer is consumed, enabling precise on-demand cutting and minimizing the average material loss of the entire batch processing.

[0135] (4) Excellent engineering applicability and compatibility with errors within the same furnace. This method can automatically identify and compensate for the performance gradient differences caused by growth rate fluctuations in different single crystal rods within the same furnace, making the same set of reference spectra widely applicable.

[0136] Preliminary determination of the applicability of the spectrum: By calculating the spectrum deviation factor γ, suitable rods with the same performance distribution as the reference single crystal rod (γ≤0.15) are selected to avoid unnecessary cutting of heterogeneous growth rods and ensure the reliability of the reuse of the reference spectrum.

[0137] Two-end linear calibration mechanism: introducing a growth rate fluctuation compensation coefficient By comparing the performance difference between the two ends of the single crystal rod to be processed and the reference single crystal rod, the axial coordinate scaling caused by the fluctuation of the component segregation rate is dynamically corrected (K>1 indicates a gentler gradient, K<1 indicates a steeper gradient).

[0138] Geometric zero-point offset compensation: through calculation Correct zero-point errors caused by mechanical clamping to ensure strict alignment between mechanical coordinates and physical performance coordinates.

[0139] Based on the same inventive concept, this application also provides a system for implementing the methods described above. The solution provided by this system is similar to the solution described in the methods above; therefore, specific limitations in one or more system embodiments provided below can be found in the limitations of the methods described above, and will not be repeated here.

[0140] In one exemplary embodiment, such as Figure 3 As shown, a non-destructive cutting and positioning system for relaxor ferroelectric single crystal rods is provided, which includes the following functional modules.

[0141] The benchmark establishment module 301 is used to continuously slice the reference single crystal rod along the axial direction, measure the electrical performance reference values ​​of each slice at the corresponding axial position, and establish a benchmark spectrum between the axial position and electrical performance reference values ​​of the reference single crystal rod. The reference single crystal rod and the single crystal rod to be processed are relaxor ferroelectric single crystal rods grown in the same batch, with the same growth process formula and the same growth equipment parameters.

[0142] The applicability determination module 302 is used to obtain the measured electrical properties of the single crystal rod to be processed at at least two axial positions, and to determine the applicability of the reference spectrum to the single crystal rod to be processed based on the deviation between the measured electrical properties and the electrical property reference values ​​at the corresponding axial positions in the reference spectrum.

[0143] The target retrieval module 303 is used to retrieve the corresponding theoretical axial position in the reference spectrum according to the target electrical performance index if the reference spectrum is applicable to the single crystal rod to be processed, so as to obtain the theoretical target interception interval.

[0144] The cutting module 304 is used to convert the theoretical axial position within the theoretical target cutting interval into the actual mechanical cutting coordinates of the single crystal rod to be processed based on the theoretical target cutting interval, the measured electrical performance value, and the electrical performance reference value, and to cut the single crystal rod to be processed according to the actual mechanical cutting coordinates.

[0145] The heterogeneity determination module 305 is used to determine that the single crystal rod to be processed is a heterogeneous growth rod if the reference spectrum is not applicable to the single crystal rod to be processed, and to establish a reference spectrum using the single crystal rod to be processed as a reference single crystal rod.

[0146] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0147] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0148] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for non-destructive cutting and positioning of relaxor ferroelectric single crystal rods, characterized in that, include: The reference single crystal rod is continuously sliced ​​along the axial direction, and the electrical performance reference values ​​of each slice at the corresponding axial position are measured to establish a reference spectrum between the axial position of the reference single crystal rod and the electrical performance reference values; the reference single crystal rod and the single crystal rod to be processed are relaxor ferroelectric single crystal rods grown in the same batch, with the same growth process formula and the same growth equipment parameters. Obtain measured electrical properties of the single crystal rod to be processed at at least two axial positions, and determine the applicability of the reference spectrum to the single crystal rod to be processed based on the deviation between the measured electrical properties and the electrical property reference values ​​at the corresponding axial positions in the reference spectrum. If the reference spectrum is applicable to the single crystal rod to be processed, then according to the target electrical performance index, the corresponding theoretical axial position is retrieved in reverse from the reference spectrum to obtain the theoretical target cut-off interval; Based on the theoretical target cut-off interval, the measured electrical performance value, and the electrical performance reference value, the theoretical axial position within the theoretical target cut-off interval is converted into the actual mechanical cutting coordinates of the single crystal rod to be processed, and the single crystal rod to be processed is cut according to the actual mechanical cutting coordinates.

2. The non-destructive cutting and positioning method for relaxor ferroelectric single crystal rods according to claim 1, characterized in that, The reference single crystal rod is continuously sliced ​​along the axial direction, including: The reference single crystal rod is continuously sliced ​​along the axial direction with equal thickness; the slice thickness satisfies 0.5mm≤d≤3mm; where d is the slice thickness.

3. The non-destructive cutting and positioning method for relaxor ferroelectric single crystal rods according to claim 1, characterized in that, Establishing a baseline spectrum between the axial position of the reference single crystal rod and its electrical performance reference value includes: Using axial position as the abscissa and electrical performance reference value as the ordinate, a continuous nonlinear reference curve is constructed through interpolation fitting to obtain the reference spectrum.

4. The non-destructive cutting and positioning method for relaxor ferroelectric single crystal rods according to claim 1, characterized in that, Obtain measured electrical properties of the single crystal rod to be processed at at least two axial positions, specifically including: A verification piece is cut from the beginning, middle and end of the constant diameter section of the single crystal rod to be processed, and the electrical performance parameters of each verification piece are measured to obtain the measured values ​​of the electrical performance at the beginning, middle and end of the constant diameter section of the single crystal rod to be processed; the electrical performance parameters include piezoelectric constant and / or dielectric constant.

5. The non-destructive cutting and positioning method for relaxor ferroelectric single crystal rods according to claim 1, characterized in that, The measured electrical properties of the single crystal rod to be processed at at least two axial positions include the measured electrical properties at the beginning and midpoint of the constant diameter section of the single crystal rod to be processed. The applicability of the reference spectrum to the single crystal rod to be processed is determined based on the deviation between the measured electrical performance values ​​and the reference electrical performance values ​​at the corresponding axial positions in the reference spectrum, including: The relative deviation between the measured electrical performance value and the reference electrical performance value at the corresponding axial position in the reference spectrum is calculated to obtain the spectrum deviation factor: ;in, This is the spectral deviation factor. These are the measured electrical properties of the starting end of the constant-diameter section of the single crystal rod to be processed. This serves as a reference value for the electrical properties at the starting point of the corresponding equal-diameter section in the baseline spectrum. The measured electrical properties at the midpoint of the single crystal rod to be processed are shown. This is the reference value for electrical properties at the midpoint of the baseline spectrum. The effective total length of the constant-diameter section of the single crystal rod to be processed; If the spectrum deviation factor is less than or equal to a preset deviation threshold, the reference spectrum is determined to be applicable to the single crystal rod to be processed; otherwise, the reference spectrum is determined to be unapplicable to the single crystal rod to be processed.

6. The non-destructive cutting and positioning method for relaxor ferroelectric single crystal rods according to claim 1, characterized in that, The measured electrical properties of the single crystal rod to be processed at at least two axial positions include the measured electrical properties of the beginning and end of the constant diameter section of the single crystal rod to be processed. Based on the theoretical target cut-off interval, the measured electrical performance values, and the electrical performance reference values, the theoretical axial position within the theoretical target cut-off interval is converted into the actual mechanical cutting coordinates of the single crystal rod to be processed, including: Based on the measured electrical properties of the starting end and the tail end of the equal-diameter section of the single crystal rod to be processed, and the reference electrical properties of the corresponding axial position in the reference spectrum, a mapping relationship between the theoretical axial position and the actual mechanical coordinates of the single crystal rod to be processed is established. Calculate the shortest distance from the theoretical target intercept interval to the quasi-isomorphic phase boundary transition region; the quasi-isomorphic phase boundary transition region is the region in the reference spectrum that satisfies S'(X)|≥λ; where S'(X) is the first derivative of the reference spectrum, X is the axial position, and λ is a set derivative threshold. If the shortest distance is greater than a set distance threshold, the theoretical axial position within the theoretical target interception interval is converted into the actual mechanical cutting coordinates of the single crystal rod to be processed according to the mapping relationship. If the shortest distance is less than or equal to a set distance threshold, it is determined that the theoretical target interception interval falls into the adjacent region of the quasi-isomorphic phase boundary transition region, and a verification piece is cut from the single crystal rod to be processed at the peak position corresponding to the quasi-isomorphic phase boundary transition region. The measured peak coordinates of the single crystal rod to be processed are determined based on the measured electrical performance values ​​of the verification piece. Calculate the local nonlinear displacement between the measured peak coordinates and the peak position; If the absolute value of the local nonlinear displacement is less than or equal to a preset displacement threshold, then the theoretical axial position within the theoretical target intercept interval is converted into the actual mechanical cutting coordinates of the single crystal rod to be processed according to the mapping relationship. If the absolute value of the local nonlinear displacement is greater than the preset displacement threshold, the mapping relationship is locally nonlinearly corrected based on the local nonlinear displacement and the peak position of the quasi-isomorphic phase boundary transition region, and the theoretical axial position within the theoretical target intercept interval is converted into the actual mechanical cutting coordinates of the single crystal rod to be processed based on the mapping relationship after local nonlinear correction.

7. The non-destructive cutting and positioning method for relaxor ferroelectric single crystal rods according to claim 6, characterized in that, Based on the measured electrical properties of the starting and ending ends of the constant-diameter section of the single crystal rod to be processed, and the reference electrical properties at the corresponding axial positions in the reference spectrum, a mapping relationship between the theoretical axial position and the actual mechanical coordinates of the single crystal rod to be processed is established, including: Based on the measured electrical properties of the starting and ending ends of the constant-diameter section of the single crystal rod to be processed, and the reference electrical properties at the corresponding axial positions in the reference spectrum, the growth rate fluctuation compensation coefficient is calculated: ;in, This is the growth rate fluctuation compensation coefficient. These are the measured electrical properties of the initial end of the constant-diameter section of the single crystal rod to be processed. This serves as a reference value for the electrical properties at the starting point of the corresponding equal-diameter section in the baseline spectrum. These are the measured electrical properties of the tail end of the single crystal rod to be processed. This is the reference value for electrical properties at the corresponding tail end in the baseline spectrum. The effective total length of the constant-diameter section of the single crystal rod to be processed; Calculate the geometric zero-point offset based on the axial zero-point deviation between the single crystal rod to be processed and the reference single crystal rod: ;in, This is the geometric zero-point offset. These are the actual mechanical coordinates of the starting end of the constant-diameter section of the single crystal rod to be processed. The coordinates of the absolute zero point of the baseline spectrum; Based on the growth rate fluctuation compensation coefficient and the geometric zero-point offset, a mapping relationship between the theoretical axial position and the actual mechanical coordinates of the single crystal rod to be processed is established: ;in, The theoretical axial position, These are the actual machine coordinates.

8. The non-destructive cutting and positioning method for relaxor ferroelectric single crystal rods according to claim 7, characterized in that, The mapping relationship after local nonlinear correction is as follows: ;in, This is a local nonlinear displacement. σ represents the peak position of the quasi-isomorphic phase boundary transition region, and σ represents the half-width at half-maximum (WHM) of the quasi-isomorphic phase boundary transition region.

9. The non-destructive cutting and positioning method for relaxor ferroelectric single crystal rods according to claim 1, characterized in that, Also includes: If the reference spectrum is not applicable to the single crystal rod to be processed, the single crystal rod to be processed is determined to be a heterogeneous growth rod, and the single crystal rod to be processed is used as a reference single crystal rod to establish the reference spectrum.

10. A non-destructive cutting and positioning system for relaxor ferroelectric single crystal rods, applied to the non-destructive cutting and positioning method for relaxor ferroelectric single crystal rods according to any one of claims 1-9, characterized in that, The system includes: The benchmark establishment module is used to continuously slice the reference single crystal rod along the axial direction, measure the electrical performance reference value of each slice at the corresponding axial position, and establish a benchmark spectrum between the axial position and electrical performance reference value of the reference single crystal rod; the reference single crystal rod and the single crystal rod to be processed are relaxor ferroelectric single crystal rods grown in the same batch, with the same growth process formula and the same growth equipment parameters. The applicability determination module is used to obtain the measured electrical properties of the single crystal rod to be processed at at least two axial positions, and to determine the applicability of the reference spectrum to the single crystal rod to be processed based on the deviation between the measured electrical properties and the electrical property reference values ​​at the corresponding axial positions in the reference spectrum. The target retrieval module is used to retrieve the corresponding theoretical axial position in the reference spectrum according to the target electrical performance index if the reference spectrum is applicable to the single crystal rod to be processed, so as to obtain the theoretical target interception interval. The cutting module is used to convert the theoretical axial position within the theoretical target cutting interval into the actual mechanical cutting coordinates of the single crystal rod to be processed, based on the theoretical target cutting interval, the measured electrical performance value, and the electrical performance reference value, and to cut the single crystal rod to be processed according to the actual mechanical cutting coordinates.

Images