Method for controlling defects and performance of stem-based barium titanate crystal

Abstract

The present application relates to the technical field of barium titanate crystal material characterization and defect control, and particularly relates to a barium titanate crystal defect and performance control method based on STEM, which comprises the following steps: TEM sample preparation is performed on a tetragonal barium titanate crystal sample to obtain a flake sample; STEM is used to obtain high-resolution STEM images of different regions in the same grain and to perform fast Fourier transform on the images respectively, so as to obtain fast Fourier transform diffraction patterns of at least two regions; diffraction spot features and interplanar spacing are extracted and compared with standard PDF cards to determine the crystal system and the crystal plane orientation; dislocation defects are identified by means of the contrast difference of high-resolution images combined with the offset or splitting of diffraction spots, and stacking fault defects are identified by means of additional diffraction fringes; and the local lattice distortion degree and the asymmetric structure distortion boundary are determined in combination with the composition and phase analysis results, and the characterization results are output. The present application realizes the accurate stripping and identification of atomic-level double spatial defects.

Description

Technical Field

[0001] This invention relates to the field of barium titanate crystal material characterization and defect control technology, and in particular to a method for controlling defects and properties of barium titanate crystals based on STEM. Background Technology

[0002] Barium titanate, a representative perovskite-structured ferroelectric material, is widely used in microelectronics and optoelectronics fields such as dielectric ceramics, piezoelectric sensors, and electro-optic modulation devices due to its excellent dielectric, piezoelectric, and electro-optic properties. During the actual preparation and service of tetragonal barium titanate crystals, microscopic defects such as dislocations and stacking faults inevitably arise within the grains due to thermal stress, mechanical loads, or electric field driving. To evaluate the macroscopic electrical properties of the material and guide process improvement, transmission electron microscopy and other techniques are typically used to characterize and analyze the types of defects within the crystal, their spatial distribution, and the resulting local lattice distortions.

[0003] Traditional defect characterization methods mostly rely on macroscopic statistical analysis or morphological observation in a single spatial domain, lacking a dual coherent interference mechanism that combines the in-situ arrangement of points in the positive space lattice with the spectral characteristics of the reciprocal space. This single-level characterization method is highly susceptible to interference from the amorphous damage layer on the surface of strongly ferroelectric sensitive materials, and macroscopic statistical methods severely dilute local weak strain signals, making it impossible to accurately peel off and qualitatively and quantitatively identify dislocations and stacking faults at the atomic scale. Summary of the Invention

[0004] To overcome the above shortcomings, this invention provides a STEM-based method for controlling defects and properties of barium titanate crystals, aiming to improve the problem of being unable to precisely peel off and qualitatively and quantitatively identify dislocations and stacking faults in situ at the atomic scale.

[0005] This invention provides the following technical solution: a method for controlling defects and properties of barium titanate crystals based on STEM, comprising: S1. TEM sample preparation was performed on the tetragonal barium titanate crystal sample to prepare a thin-film sample that meets the requirements of atomic-level high-resolution characterization. S2. Use STEM to perform transmission imaging on the thin-film sample to obtain high-resolution STEM images of different regions within the same grain. S3. Perform Fast Fourier Transform on different regions of the high-resolution STEM image to obtain Fast Fourier Transform diffraction patterns of at least two different regions. S4. Extract the diffraction spot features and interplanar spacing from the fast Fourier transform diffraction patterns in each region, and compare them with the standard PDF card of tetragonal barium titanate to determine the crystal system and crystal orientation of the tetragonal barium titanate crystal sample. S5. Analyze the contrast difference in different regions within the same grain using the high-resolution STEM image, identify dislocation defects by combining the offset or splitting of diffraction spots in the fast Fourier transform diffraction pattern, and identify stacking fault defects by combining the additional diffraction fringes in the fast Fourier transform diffraction pattern. S6. Combining the results of composition analysis and phase structure analysis, determine the degree of local lattice distortion caused by the dislocation defects and the boundary of local asymmetric structural distortion caused by the stacking fault defects, and output the micro-defect distribution characterization results.

[0006] Preferably, in S1, the TEM sample preparation of the tetragonal barium titanate crystal sample includes: Tetragonal barium titanate crystal samples were subjected to stepwise peeling and thinning using focused ion beam plasma slicing or high-energy, low-loss ion beam thinning. During the peeling and thinning process, the thinning thickness of the sheet sample is controlled to meet the probability window of a single inelastic scattering of the electron beam. The amorphous damage layer on the surface of the thin-film sample was cleaned and removed to prepare an isotropic thin-film sample that meets the requirements of atomic-level high-resolution characterization and can present clear atomic occupancy contrast.

[0007] Preferably, in S2, acquiring high-resolution STEM images of different regions within the same grain includes: Obtain local thickness distribution information of thin-section samples; Based on the local thickness distribution information, the accelerating voltage and focusing beam of the STEM are dynamically matched and set to control the electron irradiation dose per unit area of ​​the thin-film sample below the preset safety threshold. While maintaining atomic-level spatial resolution, transmission imaging is performed on the matrix region and the defect local distribution region within the same grain to obtain high-resolution STEM images that can effectively suppress the radiation-induced phase transition effect of tetragonal barium titanate crystal samples.

[0008] Preferably, in S4, determining the crystal system and crystal plane orientation of the tetragonal barium titanate crystal sample includes: In the Fourier reciprocal space, low-index diffraction spot pairs with orientations parallel to and perpendicular to the spontaneous polarization axis are extracted, and the spatial coordinates of the central symmetry axis of each low-index diffraction spot pair are calculated. The spatial distance of diffraction spots in the fast Fourier transform diffraction pattern in each region is measured, and the lattice stretching distortion ratio of the non-equiaxed unit cell parameter is calculated. The lattice stretching distortion ratio was mapped and matched with the standard PDF card of tetragonal barium titanate to accurately determine the crystal system and crystal plane orientation of the tetragonal barium titanate crystal sample.

[0009] Preferably, in S5, the identification of dislocation defects includes: In the positive space of a high-resolution STEM image, a Burgers loop is constructed along the discontinuous boundary of the atomic arrangement, and the Burgers vector is calculated. In fast Fourier transform diffraction patterns, the spatial asymmetric splitting features and offset displacement of higher-order diffraction spots are extracted. By coupling the reciprocal space deformation quantity with the positive space distortion field, the insertion direction of the edge half-atom plane of the dislocation defect and the position of the local center are determined.

[0010] Preferably, in S5, the identification of stacking fault defects includes: Capture anomalous information about discontinuous atomic arrangement at the interface of two or more crystal stacks with different orientations in high-resolution STEM images; In fast Fourier transform diffraction patterns, identify associated localized additional diffraction fringes that extend along a specific normal direction between the main diffraction spots of the matrix. The full width at half maximum (FWHM) and relative intensity distribution of the additional diffraction fringes were extracted, and the stacking fault disorder of the tetragonal barium titanate crystal sample along the stacking direction was quantitatively determined using these values.

[0011] Preferably, it also includes process control steps, the specific process of which includes: Obtain the degree of local lattice distortion and the boundary of local asymmetric structural distortion from the micro-defect distribution characterization results of the S6 output; The micro-defect distribution characterization results are input into a preset defect-process mapping model to parse out the corresponding process parameter adjustment strategy; Based on the process parameter adjustment strategy, the dynamic preparation process parameters of the tetragonal barium titanate crystal samples were adjusted in a closed loop.

[0012] Preferably, the closed-loop adjustment of the dynamic preparation process parameters for the tetragonal barium titanate crystal sample includes: Lowering the sintering temperature reduces shear stress concentration at the grain growth front, thereby inhibiting the nucleation and proliferation of dislocation sources at the source. The cooling rate of the thermal gradient within the temperature range of the paraelectric-ferroelectric phase transition is controlled to release the non-uniform transient thermal stress during the paraelectric-ferroelectric phase transition process. By eliminating the lattice mismatch of the spontaneous polarization axis, the probability of crystal stacking fault formation in the tetragonal barium titanate crystal sample during the cooling crystallization process is reduced.

[0013] Preferably, the closed-loop adjustment of the dynamic preparation process parameters for the tetragonal barium titanate crystal sample further includes: Based on the local distortion field distribution in the micro-defect distribution characterization results, trace doping elements with ionic radii matching the main lattice are mixed into the matrix material of the tetragonal barium titanate crystal sample. The trace doping elements are triggered to spontaneously agglomerate towards the local elastic distortion region at the dislocation center and stacking fault interface, forming a Cotillard atmosphere. The slip motion of the dislocation and stacking fault defects is pinned at the atomic level using the Cotillard gas cloud.

[0014] Preferably, the process control step is followed by a device fabrication and performance verification process: Dielectric ceramics, piezoelectric sensors, or electro-optic modulation devices were prepared using regulated tetragonal barium titanate crystal samples. By reducing the stress within the crystal domains and the non-periodic stacking disorder, the blocking effect on the spontaneous ferroelectric domain wall motion inside the crystal is reduced. Suppress domain wall fatigue degradation and leakage current failure of devices under periodic electric field drive or cyclic mechanical stress load.

[0015] The present invention has the following beneficial effects: 1. In this invention, by using scanning transmission electron microscopy to capture the discontinuous occupancy contrast of the atomic lattice in positive space, and by combining fast Fourier transform diffraction patterns to extract the splitting, shifting, and associated local extra diffraction fringes of higher-order diffraction spots in reciprocal space, the dual-space qualitative and quantitative characterization of microscopic defects inside tetragonal barium titanate crystals at the atomic scale is realized. This solves the problem that traditional macroscopic characterization methods severely dilute local weak strain signals and cannot accurately remove dislocations and stacking faults in situ at the atomic level.

[0016] 2. In this invention, a pre-defined defect-process mapping model is constructed for reverse closed-loop control. The sintering temperature and phase transformation cooling rate are controlled in the thermodynamic dimension to release transient thermal stress. In the chemical composition dimension, trace dopant elements are segregated according to the local distortion field to form Koch gas clusters. This reduces the jamming effect of spontaneous ferroelectric domain wall movement inside the crystal and significantly extends the physical properties and cycle life of the terminal device. It solves the problem of severe domain wall fatigue degradation and leakage current failure caused by strong defect pinning in ferroelectric materials under periodic electric field drive or cyclic mechanical stress load.

[0017] 3. In this invention, a solid-state physics mechanism based on the asymmetric structural distortion and symmetry breaking of the perovskite octahedral lattice is established, which has the universality of lattice reciprocal space analysis process and defect-process control paradigm. This enables the radiation and promotion of this defect analysis and performance regulation method to other perovskite ferroelectric materials, and solves the problem that the existing ferroelectric crystal defect analysis technology has poor universality and is difficult to standardize defect engineering regulation for the entire perovskite family of microelectronic and optoelectronic devices. Attached Figure Description

[0018] Figure 1The flowchart shows the STEM-based method for controlling defects and properties of barium titanate crystals proposed in this invention. Figure 2 This is a schematic diagram of the atomic structure of dislocations and stacking faults in the STEM-based method for controlling defects and properties of barium titanate crystals proposed in this invention. Figure 3 This is a schematic diagram of the first diffraction pattern of the barium titanate crystal defect and performance control method based on STEM proposed in this invention. Figure 4 This is a schematic diagram of the second diffraction pattern of the barium titanate crystal defect and performance control method based on STEM proposed in this invention. Figure 5 This is a schematic diagram of the third diffraction pattern of the barium titanate crystal defect and performance control method based on STEM proposed in this invention. Detailed Implementation

[0019] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] In embodiments of the present invention, the present invention provides a method for controlling defects and properties of barium titanate crystals based on STEM, such as... Figures 1-5 As shown, it includes: S1. TEM sample preparation was performed on the tetragonal barium titanate crystal sample to prepare a thin-film sample that meets the requirements of atomic-level high-resolution characterization. Furthermore, in S1, the TEM sample preparation for the tetragonal barium titanate crystal sample includes: Tetragonal barium titanate crystal samples were subjected to stepwise peeling and thinning using focused ion beam plasma slicing or high-energy, low-loss ion beam thinning. During the peeling and thinning process, the thinning thickness of the sheet sample is controlled to meet the probability window of a single inelastic scattering of the electron beam. The amorphous damage layer on the surface of the thin-film sample was cleaned and removed to prepare an isotropic thin-film sample that meets the requirements of atomic-level high-resolution characterization and can present clear atomic occupancy contrast.

[0021] Specifically, a transmission electron microscope (TEM) was used to prepare a tetragonal barium titanate crystal sample to fabricate a thin-film sample that meets the requirements for atomic-level high-resolution characterization. This sample preparation is fundamental for subsequent high-resolution imaging using scanning transmission electron microscopy and for accurately identifying microscopic defects. Since tetragonal barium titanate is a ferroelectric material, its crystal structure is extremely sensitive to mechanical stress and high-energy particle irradiation. Conventional mechanical polishing or high-energy ion bombardment can easily introduce an amorphous damage layer on the sample surface, and may even induce a phase transition to a paraelectric or other metastable phase due to localized stress concentration or temperature rise, thereby masking or fabricating real dislocations and stacking faults within the crystal. Therefore, this approach ensures the lattice integrity of the sample by employing a synergistic process of stepwise micro / nano fabrication and surface damage control layer elimination during the preparation of the thin-film sample.

[0022] Specifically, the stepwise stripping and thinning of samples can be achieved through focused ion beam plasma slicing or high-energy, low-loss ion beam thinning. When using focused ion beam plasma slicing, the ion beam generated by a gas plasma source precisely positions and cuts specific grains of a bulk tetragonal barium titanate crystal sample. To avoid lattice reconstruction caused by a single high-energy bombardment, a stepwise thinning strategy is adopted, i.e., as the thickness of the sample sheet decreases, the accelerating voltage and beam current intensity of the ion beam are gradually reduced to achieve micro-scale, uniform stepwise stripping. Throughout the stripping and thinning process, the thinning thickness of the sample sheet needs to be dynamically monitored and controlled at the endpoint to ensure that the final thickness is strictly limited within the probability window of a single inelastic scattering by the electron beam. According to the physical mechanism of electron energy loss spectrum, when the sample sheet is too thick, high-energy electrons will undergo multiple inelastic scatterings when penetrating the sample, which will lead to severe blurring of the background contrast in high-resolution scanning transmission electron microscopy images, making it impossible to resolve single-row atomic occupancy sites. To quantitatively characterize whether the sample thickness meets the single scattering requirement, a formula for calculating the relative thickness of inelastic scattering is introduced: ; In the above formula, the parameters Represents the local absolute thickness of the thin-film sample. This represents the mean free path of high-energy electrons in the tetragonal barium titanate crystal, representing the average inelastic scattering. This represents the total integrated intensity of the full spectrum in the transmission electron energy loss spectrum, while This represents the integral intensity of the zero-loss peak in the energy loss spectrum. By controlling the thinning process, the calculated relative thickness ratio is... It is within the preset single scattering dominance range, thus physically eliminating the interference of multiple scattering on spatial resolution.

[0023] Once the thickness of the thin-film sample reaches the aforementioned scattering window requirement, due to the ion bombardment in the previous process, the shallow surface of the sample often inevitably retains an amorphous damage layer or a low-energy ion implantation layer of tens of nanometers, which leads to discontinuous atomic occupancy contrast. Therefore, a surface cleaning and damage layer removal process is subsequently performed. This process typically involves using an argon ion beam with an extremely low accelerating voltage to perform a grazing-angle light bombardment on both sides of the thin-film sample, or using a chemical micro-etching solution for gentle selective cleaning. The purpose of this step is to completely remove the disordered amorphous phase from the surface without causing secondary damage to the intact tetragonal lattice inside, thereby constructing an isotropic thin-film sample with a smooth surface and a continuous internal lattice extending to the edges. In subsequent scanning transmission electron microscopy characterization, the prepared thin-film sample allows the electron beam to coherently interfere along a specific low-index zone axis, thus exhibiting high-contrast, clearly defined atomic occupancy contrast in the positive space image. This lays a solid sample foundation for the subsequent accurate extraction of interplanar spacing and the resolution of local distortion fields caused by dislocations and stacking faults.

[0024] S2. Use STEM to perform transmission imaging on thin-section samples to obtain high-resolution STEM images of different regions within the same grain. Furthermore, in S2, acquiring high-resolution STEM images of different regions within the same grain includes: Obtain local thickness distribution information of thin-section samples; Based on the local thickness distribution information, the accelerating voltage and focusing beam of the STEM are dynamically matched and set to control the electron irradiation dose per unit area of ​​the thin-film sample below the preset safety threshold. While maintaining atomic-level spatial resolution, transmission imaging is performed on the matrix region and the defect local distribution region within the same grain to obtain high-resolution STEM images that can effectively suppress the radiation-induced phase transition effect of tetragonal barium titanate crystal samples.

[0025] Specifically, high-resolution scanning transmission electron microscopy (STEM) imaging is performed on the thin-film samples prepared in the aforementioned steps to obtain STEM images of different regions within the same grain. The purpose of this imaging process is to address the technical bottleneck of ferroelectric materials being susceptible to radiation damage and induced phase transitions under continuous electron beam bombardment by precisely controlling the electron beam irradiation dose. Since tetragonal barium titanate is a charge- and structure-sensitive material, prolonged inelastic scattering of high-energy electron beams easily induces lattice relaxation and may even induce local phase transitions to the paraelectric phase, thereby masking the true morphology of microscopic defects.

[0026] To overcome the aforementioned characterization distortions, the process of acquiring local thickness distribution information of the thin-film samples was first implemented. Since the prepared samples generally exhibit minute wedge-shaped thickness differences, the energy absorption rate of the electron beam penetrating regions of different thicknesses varies significantly. The local absolute thickness of the thin-film samples can be determined online using micro-area electron energy loss spectroscopy, thereby establishing a local thickness distribution matrix corresponding to the imaging region in space, providing a quantitative benchmark for subsequent dynamic matching of beam parameters.

[0027] After acquiring the local thickness distribution information, the control system dynamically matches and sets the accelerating voltage and focusing beam of the scanning transmission electron microscope based on the thickness distribution matrix, and runs an active safety control mechanism for electron irradiation dose. To keep the electron irradiation dose per unit area of ​​the thin-film sample below a preset safety threshold, the system automatically reduces the focusing beam intensity when the scanning electron beam moves to a thinner region; and increases the accelerating voltage to enhance electron penetration when the beam moves to a relatively thicker region. Here, an irradiation dose rate control algorithm is introduced for dynamic adjustment: ; In the above formula, the parameters Represents the electron irradiation dose rate per unit area, parameter The parameter represents the dynamically adjusted physical convergence beam intensity of a scanning transmission electron microscope. Represents the charge of a single electron, parameter This represents the effective beam spot cross-sectional area of ​​the electron beam at local scanning points on the surface of the thin-film sample. Using this irradiation dose rate control algorithm, the control system adjusts the intensity of the physically converging beam in real time based on the absolute thickness of each local pixel during the two-dimensional scanning movement, ensuring that the irradiation dose rate of each imaging segment is below the intrinsic critical damage threshold.

[0028] Through the closed-loop operation of the aforementioned dose defense mechanism, transmission electron microscopy (TEM) imaging is performed on the matrix region and the locally distributed defect region within the same grain while maintaining atomic-level spatial resolution. A highly coherent physically converging beam scans point-by-point through the crystal lattice, acquiring high-contrast atomic occupancy contrast images. This imaging method effectively suppresses the radiation-induced phase transition effect in tetragonal barium titanate crystal samples due to thermal effects and charge distortion, ensuring that the acquired high-resolution scanning transmission electron microscopy images can objectively reflect the diffraction contrast difference caused by dislocations and the crystal stacking differences caused by stacking faults within the grain. This provides raw image data for subsequent high-precision defect analysis using Fourier transform.

[0029] S3. Perform Fast Fourier Transform on different regions of the high-resolution STEM image to obtain Fast Fourier Transform diffraction patterns of at least two different regions. Specifically, different regions of the acquired high-resolution scanning transmission electron microscopy images are processed using Fast Fourier Transform (FFT) to obtain FFT diffraction patterns from at least two different regions. This mathematical transformation and image segmentation process converts the lattice atomic morphology information in positive space into frequency field signals in reciprocal space, which is a fundamental technique for accurately extracting information reflecting lattice asymmetry distortion. Because the microscopic defects within tetragonal barium titanate grains exhibit strong localization characteristics, directly performing macroscopic transformations on the entire image would overwhelm the weak heterogeneous signals caused by the defects by the large area of ​​perfect matrix signals. Therefore, a regionalized and hierarchical local transformation strategy must be adopted.

[0030] The specific operation is as follows: First, the original high-resolution scanning transmission electron microscopy image is read into the memory matrix of the digital image processing system. Based on the contrast gradient distribution characteristics of the image pixel array, a window cropping algorithm is used to segment the image into two or more independent sub-region matrices. These sub-region matrices contain at least one matrix reference region exhibiting perfect lattice occupancy, and one or more candidate defect regions exhibiting atomic plane skewness, discontinuity, or abnormal contrast. This spatial segmentation ensures that each sub-region matrix corresponds to a specific microscopic physical state within the grain.

[0031] After region segmentation, a Fast Fourier Transform (FFT) is performed on each independent sub-region matrix. The FFT is essentially a fast decimation algorithm for the two-dimensional Discrete Fourier Transform (DFT). It decomposes a two-dimensional square pixel matrix into a multiplied one-dimensional row and column vector, mapping the frequency of grayscale changes in the two-dimensional image in the spatial domain to a pulse function matrix in the frequency domain. To achieve a quantitative conversion between the pixel size in the positive space and the diffraction spot features in the reciprocal space, a two-dimensional Discrete FFT frequency transformation formula is introduced: ; In the above formula, the parameters The coordinates in the reciprocal space after the transformation are The complex spectral function at that point represents the complex amplitude at each point in the corresponding Fast Fourier Transform diffraction pattern. Parameters and These represent the total number of rows and columns of pixels in the horizontal and vertical directions of the input sub-region image matrix, respectively. Parameters Represents the coordinates in the positive space image. The pixel grayscale value at that location, and Then it is the imaginary unit. Through the cascaded calculation of this two-dimensional discrete fast Fourier transform frequency transformation formula, the complex two-dimensional atomic occupancy morphology of each sub-region is completely projected onto the reciprocal space disk array composed of various diffraction frequency spots.

[0032] Through the closed-loop operation of the above transformation, the system ultimately outputs fast Fourier transform diffraction patterns from at least two different regions. These diffraction patterns are typically applied in the form of two-dimensional power spectrum images, which are essentially intensity distribution maps obtained by performing modulus square operations on complex spectral functions. These independent diffraction patterns accurately record the unique spatial frequency characteristics of the matrix region and the defect region, respectively. The geometric arrangement of the spots in their images has a strict reciprocal mapping relationship with the lattice constant and crystal orientation of tetragonal barium titanate. This application of regionalized transformation eliminates the dilution effect of large-scale macroscopic statistics on local weak strain signals, providing a high signal-to-noise ratio reciprocal lattice space data source for subsequent steps to qualitatively and quantitatively identify dislocations and stacking faults by comparing standard cards and capturing spot splitting or associated fringes.

[0033] S4. Extract the diffraction spot characteristics and interplanar spacing from the fast Fourier transform diffraction patterns of each region, and compare them with the standard PDF card of tetragonal barium titanate to determine the crystal system and crystal orientation of the tetragonal barium titanate crystal sample. Furthermore, in S4, determining the crystal system and crystal plane orientation of the tetragonal barium titanate crystal sample includes: In the Fourier reciprocal space, low-index diffraction spot pairs with orientations parallel to and perpendicular to the spontaneous polarization axis are extracted, and the spatial coordinates of the central symmetry axis of each low-index diffraction spot pair are calculated. The spatial distance of diffraction spots in the fast Fourier transform diffraction pattern in each region is measured, and the lattice stretching distortion ratio of the non-equiaxed unit cell parameter is calculated. The lattice stretching distortion ratio was mapped and matched with the standard PDF card of tetragonal barium titanate to accurately determine the crystal system and crystal plane orientation of the tetragonal barium titanate crystal sample.

[0034] Specifically, the fast Fourier transform diffraction patterns of each sub-region obtained from the calculations are feature-extracted and compared with standard cards to accurately determine the crystal system and crystal plane orientation of the tetragonal barium titanate crystal sample. The purpose of this process is to establish a quantitative correspondence between the geometric features of reciprocal space and the true crystallographic orientation in normal space. Because tetragonal barium titanate exhibits slight isoaxiality at room temperature, there is an extremely small deviation between the lattice constant along its spontaneous polarization axis and the lattice constant perpendicular to the spontaneous polarization axis. Therefore, high-precision reciprocal space vector calculations are needed to capture this lattice distortion, thereby achieving high-accuracy phase structure determination.

[0035] The specific operation involves first locating and extracting low-index diffraction spot pairs aligned with and perpendicular to the spontaneous polarization axis crystal planes within the reciprocal lattice space formed by the Fast Fourier Transform diffraction pattern. These low-index diffraction spot pairs originate from the family of crystal planes with larger interplanar spacing and highest diffraction intensity. The system automatically searches for local intensity maxima in the power spectrum image using a centroid-based subpixel spot localization algorithm, thereby accurately locking the transmission center spot and the surrounding low-index principal diffraction spots. Next, using coordinate geometry operations, the spatial coordinates of the central symmetry axis of each low-index diffraction spot pair are calculated, thus establishing the geometric center and origin of the entire reciprocal lattice pattern to eliminate systematic errors caused by sample tilt or camera constant drift.

[0036] After establishing the origin, the system executes the characteristic parameter measurement and tensile distortion analysis process. During operation, the program automatically measures the spatial distance of diffraction spots in the fast Fourier transform diffraction patterns of each analysis region, i.e., calculates the pixel distance vector from the main diffraction spot to the central origin. Since the reciprocal space distance is inversely proportional to the interplanar spacing in normal space, the system, combined with the calibration camera constant of a high-resolution scanning transmission electron microscope, converts the pixel distance into the actual interplanar spacing. To quantitatively characterize the anisotropic distortion degree of the tetragonal barium titanate lattice, a formula for calculating the lattice tensile distortion ratio using non-isometric unit cell parameters is introduced: ; In the above formula, the parameters The lattice stretching distortion ratio, representing the parameters of an isometric unit cell, physically reflects the degree of anisotropic elongation or compression of the barium titanate unit cell during the transformation from cubic to tetragonal phase. Parameters The parameter represents the geometrical spatial distance from the origin of reciprocal lattice space to a low-index diffraction spot oriented parallel to the spontaneous polarization axis crystal plane. This represents the geometric spatial distance from the low-index diffraction spot oriented perpendicular to the spontaneous polarization axis to the origin of reciprocal space. Using this ratio calculation formula, the minute lattice anisotropy that is difficult to discern with the naked eye in high-resolution images can be extended and magnified into an asymmetric proportional deviation of the reciprocal space spot distance.

[0037] After completing the measurement and comparison calculations, the system enters the final mapping and alignment application stage. The calculated lattice stretching distortion ratio and the extracted interplanar spacings are mapped and matched with the tetragonal barium titanate standard powder diffraction cards (PDF cards) pre-stored in the system database, performing data mapping and residual minimization matching. By comparing the conformity between the experimental lattice ratio and the tetragonal intrinsic lattice constant ratio in the standard cards, it is possible not only to accurately determine whether the currently measured grain maintains a complete tetragonal crystal system, but also to directly solve for the crystal orientation matrix of each local region within the grain based on the spatial distribution direction of the low-index diffraction spot vectors. This application result of determining crystal orientation provides an absolute crystallographic coordinate reference for accurately identifying the Burgers vector direction of dislocation defects and determining the stacking boundaries of stacking fault defects in subsequent steps.

[0038] S5. Analyze the contrast difference in different regions within the same grain using high-resolution STEM images, identify dislocation defects by combining the offset or splitting of diffraction spots in the fast Fourier transform diffraction pattern, and identify stacking fault defects by combining additional diffraction fringes in the fast Fourier transform diffraction pattern. Furthermore, in S5, identifying dislocation defects includes: In the positive space of a high-resolution STEM image, a Burgers loop is constructed along the discontinuous boundary of the atomic arrangement, and the Burgers vector is calculated. In fast Fourier transform diffraction patterns, the spatial asymmetric splitting features and offset displacement of higher-order diffraction spots are extracted. By coupling the reciprocal space deformation quantity with the positive space distortion field, the insertion direction of the edge half-atom plane of the dislocation defect and the position of the local center are determined.

[0039] Furthermore, in S5, identifying stacking fault defects includes: Capture anomalous information about discontinuous atomic arrangement at the interface of two or more crystal stacks with different orientations in high-resolution STEM images; In fast Fourier transform diffraction patterns, identify associated localized additional diffraction fringes that extend along a specific normal direction between the main diffraction spots of the matrix. The full width at half maximum (FWHM) and relative intensity distribution of the additional diffraction fringes were extracted, and the stacking fault disorder of the tetragonal barium titanate crystal sample along the stacking direction was quantitatively determined using these values.

[0040] Specifically, the system runs a microscopic defect identification program, analyzes the contrast differences in different regions within the same grain using high-resolution scanning transmission electron microscopy images, and extracts specific distortion features from the fast Fourier transform diffraction patterns to identify dislocation and stacking fault defects within the grain in situ. When dislocation or stacking fault strain occurs in the tetragonal barium titanate crystal lattice, the atoms in the defect core region deviate from their equilibrium positions, leading to a weakening of coherent scattering intensity, which manifests as localized darkening or skewness in the positive space image. However, relying solely on positive space contrast grayscale analysis is easily affected by sample thickness fluctuations or noise interference. Therefore, this scheme achieves accurate identification and separation of defect types through dual coupling of positive space atomic lattice arrangement and reciprocal space spectral characteristics.

[0041] Specifically, in the dislocation defect identification process, the system first constructs a closed Burgers loop in the positive space matrix of the digital image processor, along the discontinuous boundaries of atomic arrangement on the high-resolution scanning transmission electron microscopy image. The atomic loops in the normal matrix region are closed, but when the loop surrounds the dislocation core, due to the presence of extra half-atomic planes, a lattice vector deviation occurs between the loop's endpoint and starting point. This is the calculation of the Burgers vector, thus completing the preliminary qualitative characterization of the dislocation properties in positive space. Simultaneously, the system enters the reciprocal space analysis process, extracting the spatial asymmetric splitting characteristics and displacement of higher-order diffraction spots in the corresponding sub-region's fast Fourier transform diffraction pattern. Because higher-order diffraction spots are much more sensitive to small lattice strains than lower-order spots, when a continuous non-uniform strain field exists around the dislocation, higher-order spots in the reciprocal space will spontaneously exhibit asymmetric stretching, splitting, or overall deviation from the ideal lattice coordinates. To quantitatively correlate the distortion states of the two spaces, a spatial strain coupling mapping equation is introduced: ; In the above formula, the parameters Represents the Burgers vector of dislocation defects calculated in positive space, with parameters Represents the geometric reciprocal vector of a specific higher-order diffraction spot selected in reciprocal space, with parameters This represents the spatial displacement of the higher-order diffraction spot relative to the theoretical position of a perfect crystal, while the parameter... This represents the specific electromagnetic lens magnification constant matrix used for image spatial calibration. By solving this spatial strain coupling mapping equation, the speckle deformation in reciprocal space is bidirectionally coupled with the local distortion field in normal space, eliminating the discrete error of calculation in a single normal space loop, thereby accurately determining the insertion direction of the edge-shaped half-atom plane of dislocation defects generated during the fabrication process and the position of the local core.

[0042] In the process of identifying stacking fault defects, the system first uses an image grayscale matrix discrimination algorithm to capture the anomalous information of discontinuous atomic arrangement at the interface of two or more crystals with different orientations in a high-resolution scanning transmission electron microscopy image. When a tetragonal barium titanate crystal experiences local slippage due to thermal stress, the normal stacking order of the lattice is disrupted, manifesting as atomic stacking dislocations in positive space. Subsequently, the system switches to the reciprocal space processing unit to identify the accompanying localized additional diffraction fringes extending along a specific normal direction between the matrix diffraction spots in the corresponding fast Fourier transform diffraction pattern. These fringes are triggered by lattice translational discontinuities caused by stacking faults, and their extension direction is physically strictly perpendicular to the stacking fault plane. The system uses a Gaussian fitting peak shape analysis algorithm to extract the full width at half maximum (FWHM) and relative intensity distribution of these additional diffraction fringes, and uses this to quantitatively analyze the stacking fault disorder degree of the tetragonal barium titanate crystal sample along this stacking direction. The quantification of stacking fault disorder can be used to assess the severity of local aperiodic deviations within a crystal, providing high-confidence data support for subsequent determination of asymmetric distortion boundaries in crystal domains.

[0043] S6. Combining the results of compositional analysis and phase structure analysis, determine the degree of local lattice distortion caused by dislocation defects and the boundary of local asymmetric structural distortion caused by stacking fault defects, and output the micro-defect distribution characterization results.

[0044] Specifically, the data processing system executes a comprehensive judgment and characterization result output program. Combining the compositional analysis and phase structure analysis results obtained in the previous steps, it determines the degree of local lattice distortion caused by dislocation defects within the grains and the boundaries of local asymmetric structural distortion caused by stacking fault defects, and finally outputs the microscopic defect distribution characterization results. This step, as the final stage of the entire analysis method, has the core operating logic of multi-field coupling between the microscopic inhomogeneity of chemical element distribution and the geometric distortion of reciprocal space, thereby transforming the abstract electron microscopic image into a data matrix that can quantitatively guide material evaluation.

[0045] The specific operational process is as follows: the data processing system first loads the local composition analysis data measured by the energy dispersive spectroscopy (EDS) instrument, as well as the phase structure analysis data calculated from the high-resolution image. Because the non-equiproportional agglomeration of titanium, barium, and oxygen atoms during the nucleation and growth of tetragonal barium titanate directly induces variations in the local lattice constant, the system uses a spatial coordinate alignment algorithm to perform pixel-level overlap mapping between the absolute abundance spectra of each element and the previously locked defect spatial coordinates.

[0046] After aligning multi-source data, the system runs a quantitative determination program for local strain fields and symmetry breaking. For dislocation defects, the system extracts the local atomic occupancy deviation and calculates the strain gradient caused by the insertion of half-atom planes using the elasticity tensor equation, thereby determining the degree of local lattice distortion. For stacking fault defects, the system focuses on analyzing the intrinsic symmetry of the ferroelectric phase broken due to changes in stacking order. By capturing the critical position of the transition from tetragonal acentrosymmetry to local centrosymmetry of the lattice point group, the system accurately delineates the distortion boundary of the local asymmetric structure. To quantitatively calculate the distortion field intensity around the dislocation core, a local distortion field tensor integral formula is introduced: ; In the above formula, the parameters This scalar value represents the final determined degree of local lattice distortion, used to quantify the destructive stress of dislocations on surrounding lattice points. Parameter This represents the anisotropic elastic modulus correction matrix, dynamically adjusted based on compositional analysis data, to eliminate lattice stiffness inhomogeneities caused by local element segregation. Parameters The gradient field represents the spatial displacement vector of atoms surrounding a dislocation nucleus in a positive space image relative to an ideal tetragonal lattice, with parameters... Represents the closed integral path along the periphery of the dislocation core. The spatial infinitesimal vector element. By discretizing the integral formula of the tensor of this localized distorted field, the system transforms discrete atomic displacements into a continuous strain energy distribution.

[0047] After calculating the distortion degree and distortion boundary, the data processing system generates a unified format two-dimensional lattice mapping dataset, which is then output to an external terminal as a micro-defect distribution characterization result. This characterization result includes the spatial topological characteristics of dislocation density within the grain, the geometric distribution of stacking fault boundaries, and the accompanying local strain field gradient cloud map. Through the output of this micro-defect distribution characterization result, process engineers can directly grasp the intrinsic structural defect distribution state of the crystal at the micro-nano scale, providing a reliable microphysical basis for evaluating the resistance to ferroelectric domain wall movement and determining the electrical degradation mechanism of the material.

[0048] Furthermore, it also includes process control steps, the specific process of which includes: Obtain the degree of local lattice distortion and the boundary of local asymmetric structural distortion from the micro-defect distribution characterization results of the S6 output; The micro-defect distribution characterization results are input into a preset defect-process mapping model to parse out the corresponding process parameter adjustment strategy; Based on the process parameter adjustment strategy, the dynamic preparation process parameters of the tetragonal barium titanate crystal samples were adjusted in a closed loop.

[0049] Furthermore, the closed-loop adjustment of the dynamic preparation process parameters for the tetragonal barium titanate crystal samples includes: Lowering the sintering temperature reduces shear stress concentration at the grain growth front, thereby inhibiting the nucleation and proliferation of dislocation sources at the source. The cooling rate of the thermal gradient within the temperature range of the paraelectric-ferroelectric phase transition is controlled to release the non-uniform transient thermal stress during the paraelectric-ferroelectric phase transition process. By eliminating the lattice mismatch of the spontaneous polarization axis, the probability of crystal stacking fault formation in tetragonal barium titanate crystal samples during the cooling crystallization process is reduced.

[0050] Furthermore, the closed-loop adjustment of the dynamic preparation process parameters for tetragonal barium titanate crystal samples also includes: Based on the local distortion field distribution in the micro-defect distribution characterization results, trace doping elements with ionic radii matching the main lattice were mixed into the matrix raw material of the tetragonal barium titanate crystal sample. Trace doping elements are triggered to spontaneously agglomerate towards the local elastic distortion region at the dislocation center and stacking fault interface, forming a Cotillard atmosphere. Atomic-level pinning of dislocation and stacking fault defects is achieved by utilizing Cotillard gas clouds.

[0051] Specifically, the system operates a process control program that transforms the microscopic defect distribution characterization results output from the previous steps into closed-loop control of the dynamic preparation process parameters for the tetragonal barium titanate crystal sample. The design logic of this process is to quantify the local strain and boundary asymmetry within the grains, trace back and correct the macroscopic thermodynamic preparation conditions, thereby blocking the evolution path of defects.

[0052] The specific operation involves the central processing unit first acquiring the degree of local lattice distortion and the boundary of local asymmetric structural distortion from the microscopic defect distribution characterization results. Subsequently, the system inputs these characteristic data into a pre-defined defect-process mapping model. This defect-process mapping model is an analytical matrix of process parameters based on multivariate partial differential regression, used to establish a quantitative causal mapping between the defect density field and the sintering temperature gradient, the cooling rate of the phase transition zone, and the chemical potential of the dopant elements. The model uses a Newton-Raphson iterative algorithm to solve the process control equations in reverse, analytically determining the process parameter adjustment strategy.

[0053] Based on the analyzed process parameter adjustment strategy, the control system performs closed-loop adjustment of the dynamic preparation process parameters of the tetragonal barium titanate crystal sample. In the thermodynamic parameter control, the sintering furnace control unit first receives instructions to lower the sintering temperature, reducing the shear stress concentration at the grain growth front and suppressing the nucleation and proliferation of dislocation sources from the source. Subsequently, when the temperature drops to within the paraelectric-ferroelectric phase transition temperature range, the control system regulates the cooling rate of the temperature field gradient to release the non-uniform transient thermal stress during the paraelectric-ferroelectric phase transition, eliminating lattice mismatch during lattice reconstruction of spontaneous polarization axes, thereby reducing the probability of stacking fault formation in the tetragonal barium titanate crystal sample during the cooling crystallization process.

[0054] In the chemical composition pinning control, the control system, based on the local distortion field distribution, incorporates trace dopant elements with ionic radii matching the main lattice into the matrix material of the tetragonal barium titanate crystal sample. During heat treatment, the system triggers these trace dopant elements to spontaneously segregate towards the local elastic distortion regions at dislocation centers and stacking fault interfaces, forming Cotillard clouds. To quantitatively characterize the atomic-level pinning effect of Cotillard clouds on defect movement, the interaction energy limit control formula for the clouds is introduced here: ; In the above formula, the parameters The parameter represents the maximum binding energy between trace doped element atoms and the local strain field of the dislocation core. The intrinsic Poisson's ratio representing the tetragonal barium titanate crystal, parameter The parameter represents the local shear modulus of the crystal. The parameter represents the modulus of the edge component of the dislocation defect. The intrinsic atomic radius represents the principal lattice sites of tetragonal barium titanate, while the parameter This represents the relative radius mismatch between the introduced trace dopant atoms and the main lattice atoms. When the material is subsequently subjected to an alternating external field, the system utilizes the formed Cotillard atmosphere to perform atomic-level pinning of the slip motion of dislocation defects and stacking fault defects, ultimately achieving stability control of the crystal microstructure.

[0055] Furthermore, following the process control steps, there is also a device fabrication and performance verification process: Dielectric ceramics, piezoelectric sensors, or electro-optic modulation devices were prepared using regulated tetragonal barium titanate crystal samples. By reducing the stress within the crystal domains and the non-periodic stacking disorder, the blocking effect on the spontaneous ferroelectric domain wall motion inside the crystal is reduced. Suppress domain wall fatigue degradation and leakage current failure of devices under periodic electric field drive or cyclic mechanical stress load.

[0056] Specifically, after executing the aforementioned process control steps, the system runs the device fabrication and performance verification process. This process, as the practical application and physical verification of the entire defect control scheme at the end-product level, aims to transform microscopic lattice optimization into macroscopic electrical and mechanical device reliability indicators. By using tetragonal barium titanate crystal samples that have undergone thermodynamic adjustment or chemical composition pinning as the core functional matrix, the final fabricated device can exhibit a stable electrical response mechanism under service conditions.

[0057] The specific operation involves first preparing dielectric ceramics, piezoelectric sensors, or electro-optic modulation devices using a regulated tetragonal barium titanate crystal sample. In preparing the dielectric ceramics or piezoelectric sensors, barium titanate crystal powder or bulk material, whose internal defect density has been suppressed through a closed-loop process, is mixed with a necessary inorganic binder. This mixture is then shaped, subjected to subsequent heat treatment, and electro-polarization curing to construct a functional ceramic body with macroscopic piezoelectric or dielectric activity. In preparing the electro-optic modulation devices, a regulated high-crystallinity thin film or crystal sample is etched into a micro / nano photonic structure and electrodes are integrated. The linear electro-optic effect unique to its non-centrosymmetric structure is used to modulate the optical signal.

[0058] When the device is put into actual service or for performance verification, the system reduces the jamming effect on spontaneous ferroelectric domain wall movement within the crystal by decreasing the stress within the crystal domains and the aperiodic stacking disorder. Since the electrical, mechanical, and optical activity of tetragonal barium titanate is highly dependent on the spatial flipping and migration of spontaneous ferroelectric domain walls, the local distortion field caused by dislocations and the aperiodic stacking disorder caused by stacking faults removed in previous steps directly eliminate the slip pinning points during domain wall movement. When the device operates under an external field, the domain walls can undergo continuous and uniform elastic wall displacement at a lower critical driving field. To quantitatively characterize the degree of reduction in the intrinsic jamming effect on domain wall movement after micro-defect removal, a control formula for the energy dissipation of ferroelectric domain wall movement is introduced: ; In the above formula, the parameters This represents the local energy dissipation rate, which is lost due to the blocking effect as spontaneous ferroelectric domain walls migrate within the crystal lattice. Its value directly determines the rate of thermal energy accumulation and fatigue degradation within the device. (Parameter) The intrinsic viscous damping coefficient, determined by the anisotropy of the material, is a parameter. This represents the overall density of residual localized micro-defects after process control, determined and guided by the aforementioned analytical methods. Parameters This represents the dynamic average shear migration velocity of spontaneously ferroelectric domain walls under external field driving. Parameter The intrinsic local electrostatic potential energy field gradient represents the result of residual dislocation strain or stacking fault asymmetric boundaries, while the parameter and These represent the total geometric area of ​​the spontaneous ferroelectric domain wall and the area of ​​its infinitesimal element, respectively. Numerical calculations using this control formula show that a decrease in defect density directly leads to a proportional decrease in the energy dissipation rate.

[0059] Through the continuous action of the aforementioned lattice dynamics mechanism during device service, the system suppresses domain wall fatigue degradation and leakage current failure under periodic electric field driving or cyclic mechanical stress loads at the final application level. When dielectric ceramics, piezoelectric sensors, or electro-optic modulation devices are subjected to alternating electric field cycles or periodic mechanical loads for extended periods, the strong pinning effect of dislocations and stacking faults on domain walls is eliminated, preventing deadlock and uneven charge accumulation at local defects and thus inhibiting the initiation of microcracks. Simultaneously, the pinning of micro-defects by the Cothurst atmosphere also limits the disordered proliferation of residual dislocations under the linear electric field, fundamentally cutting off the physical path of rapid diffusion of oxygen vacancies along dislocation lines to form conductive channels. This effectively curbs the sudden increase in leakage current and dielectric breakdown failure under high-voltage electric fields, achieving in-situ verification of the device's macroscopic property stability.

[0060] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for controlling defects and properties of barium titanate crystals based on STEM, characterized in that, include: S1. TEM sample preparation was performed on the tetragonal barium titanate crystal sample to prepare a thin-film sample that meets the requirements of atomic-level high-resolution characterization. S2. Use STEM to perform transmission imaging on the thin-film sample to obtain high-resolution STEM images of different regions within the same grain. S3. Perform Fast Fourier Transform on different regions of the high-resolution STEM image to obtain Fast Fourier Transform diffraction patterns of at least two different regions. S4. Extract the diffraction spot features and interplanar spacing from the fast Fourier transform diffraction patterns in each region, and compare them with the standard PDF card of tetragonal barium titanate to determine the crystal system and crystal orientation of the tetragonal barium titanate crystal sample. S5. Analyze the contrast difference in different regions within the same grain using the high-resolution STEM image, identify dislocation defects by combining the offset or splitting of diffraction spots in the fast Fourier transform diffraction pattern, and identify stacking fault defects by combining the additional diffraction fringes in the fast Fourier transform diffraction pattern. S6. Combining the results of composition analysis and phase structure analysis, determine the degree of local lattice distortion caused by the dislocation defects and the boundary of local asymmetric structural distortion caused by the stacking fault defects, and output the micro-defect distribution characterization results.

2. The method for controlling defects and properties of barium titanate crystals based on STEM according to claim 1, characterized in that, In S1, the TEM sample preparation of the tetragonal barium titanate crystal sample includes: Tetragonal barium titanate crystal samples were subjected to stepwise peeling and thinning using focused ion beam plasma slicing or high-energy, low-loss ion beam thinning. During the peeling and thinning process, the thinning thickness of the sheet sample is controlled to meet the probability window of a single inelastic scattering of the electron beam. The amorphous damage layer on the surface of the thin-film sample was cleaned and removed to prepare an isotropic thin-film sample that meets the requirements of atomic-level high-resolution characterization and can present clear atomic occupancy contrast.

3. The method for controlling defects and properties of barium titanate crystals based on STEM according to claim 1, characterized in that, In S2, acquiring high-resolution STEM images of different regions within the same grain includes: Obtain local thickness distribution information of thin-section samples; Based on the local thickness distribution information, the accelerating voltage and focusing beam of the STEM are dynamically matched and set to control the electron irradiation dose per unit area of ​​the thin-film sample below the preset safety threshold. While maintaining atomic-level spatial resolution, transmission imaging is performed on the matrix region and the defect local distribution region within the same grain to obtain high-resolution STEM images that can effectively suppress the radiation-induced phase transition effect of tetragonal barium titanate crystal samples.

4. The method for controlling defects and properties of barium titanate crystals based on STEM according to claim 1, characterized in that, In S4, determining the crystal system and crystal plane orientation of the tetragonal barium titanate crystal sample includes: In the Fourier reciprocal space, low-index diffraction spot pairs with orientations parallel to and perpendicular to the spontaneous polarization axis are extracted, and the spatial coordinates of the central symmetry axis of each low-index diffraction spot pair are calculated. The spatial distance of diffraction spots in the fast Fourier transform diffraction pattern in each region is measured, and the lattice stretching distortion ratio of the non-equiaxed unit cell parameter is calculated. The lattice stretching distortion ratio was mapped and matched with the standard PDF card of tetragonal barium titanate to accurately determine the crystal system and crystal plane orientation of the tetragonal barium titanate crystal sample.

5. The method for controlling defects and properties of barium titanate crystals based on STEM according to claim 1, characterized in that, In S5, the identification of dislocation defects includes: In the positive space of a high-resolution STEM image, a Burgers loop is constructed along the discontinuous boundary of the atomic arrangement, and the Burgers vector is calculated. In fast Fourier transform diffraction patterns, the spatial asymmetric splitting features and offset displacement of higher-order diffraction spots are extracted. By coupling the reciprocal space deformation quantity with the positive space distortion field, the insertion direction of the edge half-atom plane of the dislocation defect and the position of the local center are determined.

6. The method for controlling defects and properties of barium titanate crystals based on STEM according to claim 1, characterized in that, In S5, the identification of stacking fault defects includes: Capture anomalous information about discontinuous atomic arrangement at the interface of two or more crystal stacks with different orientations in high-resolution STEM images; In fast Fourier transform diffraction patterns, identify associated localized additional diffraction fringes that extend along a specific normal direction between the main diffraction spots of the matrix. The full width at half maximum (FWHM) and relative intensity distribution of the additional diffraction fringes were extracted, and the stacking fault disorder of the tetragonal barium titanate crystal sample along the stacking direction was quantitatively determined using these values.

7. The method for controlling defects and properties of barium titanate crystals based on STEM according to claim 1, characterized in that, It also includes process control steps, the specific process of which includes: Obtain the degree of local lattice distortion and the boundary of local asymmetric structural distortion from the micro-defect distribution characterization results of the S6 output; The micro-defect distribution characterization results are input into a preset defect-process mapping model to parse out the corresponding process parameter adjustment strategy; Based on the process parameter adjustment strategy, the dynamic preparation process parameters of the tetragonal barium titanate crystal samples were adjusted in a closed loop.

8. The method for controlling defects and properties of barium titanate crystals based on STEM according to claim 7, characterized in that, The closed-loop adjustment of the dynamic preparation process parameters for the tetragonal barium titanate crystal sample includes: Lowering the sintering temperature reduces shear stress concentration at the grain growth front, thereby inhibiting the nucleation and proliferation of dislocation sources at the source. The cooling rate of the thermal gradient within the temperature range of the paraelectric-ferroelectric phase transition is controlled to release the non-uniform transient thermal stress during the paraelectric-ferroelectric phase transition process. By eliminating the lattice mismatch of the spontaneous polarization axis, the probability of crystal stacking fault formation in the tetragonal barium titanate crystal sample during the cooling crystallization process is reduced.

9. The method for controlling defects and properties of barium titanate crystals based on STEM according to claim 7, characterized in that, The closed-loop adjustment of the dynamic preparation process parameters for the tetragonal barium titanate crystal sample also includes: Based on the local distortion field distribution in the micro-defect distribution characterization results, trace doping elements with ionic radii matching the main lattice are mixed into the matrix material of the tetragonal barium titanate crystal sample. The trace doping elements are triggered to spontaneously agglomerate towards the local elastic distortion region at the dislocation center and stacking fault interface, forming a Cotillard atmosphere. The slip motion of the dislocation and stacking fault defects is pinned at the atomic level using the Cotillard gas cloud.

10. The method for controlling defects and properties of barium titanate crystals based on STEM according to claim 7, characterized in that, Following the process control steps, a device fabrication and performance verification process is also included: Dielectric ceramics, piezoelectric sensors, or electro-optic modulation devices were prepared using regulated tetragonal barium titanate crystal samples. By reducing the stress within the crystal domains and the non-periodic stacking disorder, the blocking effect on the spontaneous ferroelectric domain wall motion inside the crystal is reduced. Suppress domain wall fatigue degradation and leakage current failure of devices under periodic electric field drive or cyclic mechanical stress load.

Images