A characterization method for short-range chemical order in high-entropy alloys

Abstract

This application belongs to the field of high-entropy alloy microstructure characterization, and specifically discloses a characterization method for the short-range chemical order of high-entropy alloys, which includes: S1 hydrogenating the high-entropy alloy sample to introduce sufficient hydrogen atoms into the crystal lattice structure of the high-entropy alloy; S2 measuring the overall microstructure of the sample. 1 H NMR spectrum, S3 pair 1 The analysis focuses on the NMR Knight shift, represented by the horizontal axis of the H NMR spectrum. Specifically, it involves statistically analyzing the Knight shift values ​​generated by each hydrogen atom in its surrounding elemental environment. This establishes a relationship between all elements in the high-entropy alloy sample and their corresponding Knight shift values. The analysis then resolves the Knight shift value for each element and its relationship with its surrounding elements. 1 The relative positions of the H NMR spectra, according to 1 Peaks in the H NMR spectrum characterize the short-range chemical ordered structure of the corresponding elements. The method of this invention is a novel characterization method covering the short-range chemical ordered structure of the entire sample.

Description

Technical Field

[0001] This application belongs to the field of high-entropy alloy microstructure characterization technology, and more specifically, relates to a characterization method for the short-range chemical order of high-entropy alloys. Background Technology

[0002] With the continuous development of science and technology, the demand for materials in high-end fields is becoming increasingly strong. The emergence of high-entropy alloys has broken the limitations of traditional alloys in various aspects of performance, becoming one of the most widely studied hot topics in recent years. In high-entropy alloys, the complexity of elements and the interactions between components enable the formation of chemically ordered short-range structures (CSRO) with a size <1 nm at the atomic scale, which is one of the significant differences between high-entropy alloys and traditional alloys. As an important characteristic of high-entropy alloys, CSRO has been found to directly affect the mechanical properties, dislocation motion during deformation, and radiation resistance of high-entropy alloys, thus possessing significant research value.

[0003] Characterizing CSRO in high-entropy alloys is a major challenge. Although many theoretical simulations of CSRO have yielded results, the extremely small size of CSRO and the diverse composition and similar atomic sizes of elements in high-entropy alloys still present numerous obstacles to direct experimental observation and characterization.

[0004] Currently, experimental methods for characterizing CSRO include transmission electron microscopy (TEM) and related techniques, atom probe chromatography (APT), and high-energy X-ray diffraction. Among these, transmission electron microscopy and related techniques are the most mature. Although it is now possible to characterize short-range ordered chemical structures using transmission electron microscopy, there are still many challenges in its experimental operation, such as the time-consuming pretreatment of the sample, the difficulty of the characterization process, and the cumbersome nature of the analysis.

[0005] Therefore, a new method is urgently needed to characterize CSRO in high-entropy alloys and further explore the changes in the local environment in high-entropy alloys. Summary of the Invention

[0006] To address the shortcomings of existing technologies, the purpose of this application is to provide a characterization method for chemical short-range order in high-entropy alloys. This method aims to provide a novel, NMR-based approach that covers the entire sample to identify chemically ordered structures (CSRO), which will facilitate further exploration of changes in the local environment within high-entropy alloys.

[0007] To achieve the above objectives, this application provides a method for characterizing the short-range chemical order of high-entropy alloys, which includes the following steps: S1: The high-entropy alloy sample to be tested is subjected to hydrogen treatment to introduce sufficient hydrogen atoms into the crystal lattice structure of the high-entropy alloy, so that a hydrogen NMR spectrum signal can appear in the nuclear magnetic resonance experiment. S2: Measure the overall 1H NMR spectrum of the sample. S3: Analyze the NMR Knight shifts represented by the horizontal axis of the 1H NMR spectrum. Specifically, count the Knight shift values ​​generated by each hydrogen atom in its surrounding elemental environment to form a relationship between all elements of the high-entropy alloy sample to be tested and the Knight shift values. Determine the Knight shift value corresponding to each element and its relative position in the 1H NMR spectrum. Characterize the chemical short-range ordered structure of the corresponding element based on the peaks in the 1H NMR spectrum.

[0008] Furthermore, the hydrogenation treatment specifically involves holding the high-entropy alloy sample to be tested, prepared by electric arc melting, at a hydrogen pressure of at least 3 MPa and a high temperature of 500°C to 600°C for at least 20 minutes.

[0009] Furthermore, the hydrogenation process is carried out until the number of H atoms in the high-entropy alloy sample to be tested is no less than 1.69 times the total number of metal element atoms.

[0010] Furthermore, in step S2, the sample after hydrogenation treatment is ground into powder, and at least 1g of the powder sample is randomly selected for nuclear magnetic resonance detection. The pulse is set to solid-state echo pulse, the rotation speed is set to 18 kHz, the 90° pulse time is set to 3μs, and the overall 1H nuclear magnetic resonance spectrum of the sample is measured at room temperature.

[0011] Furthermore, in step S3, a face-centered cubic supercell containing 32 metal atoms is constructed, and hydrogen atoms are filled in all the tetrahedral interstices to form a supercell structure containing 32 metal atoms and 64 hydrogen atoms. In the supercell structure, each hydrogen atom occupies a tetrahedral interstitial space, and is surrounded by four nearest-neighbor metal atoms. Calculate the coordination environment of these nearest-neighbor metal atoms. 1 The Knight shift value of H is obtained, and the Knight shift value generated by each hydrogen atom in the corresponding elemental environment is calculated.

[0012] Furthermore, through statistical analysis of the coordination of the nearest-neighbor metal atoms of hydrogen atoms, the types and numbers of the four nearest-neighbor metal atoms of each hydrogen atom were determined. This four-coordinate metal atom configuration directly corresponds to the short-range chemical order structure at the atomic scale. By comparing the theoretical Knight shifts of different chemical short-range ordered structures with the corresponding hydrogen atoms, the contribution weights of each metal element to the Knight shifts in the ¹H NMR spectrum were quantitatively analyzed.

[0013] Furthermore, the contribution of each metal element to the Knight shift in the ¹H NMR spectrum was set as an unknown variable, and a system of five linear equations containing 64 samples was constructed using the theoretical Knight shift as the response value: y=Xx Where y is the theoretical Knight displacement, X is the metal atom coordination configuration matrix, and x is the contribution vector of the element to be determined to the Knight displacement.

[0014] Furthermore, the random sample consensus algorithm is used to screen outliers and fit the model in the dataset. The minimum sample size n=6 is selected, and the probability convergence formula of the random sample consensus algorithm is used for calculation: 1 (1 (1 ) n) k =P in, denoted as the proportion of outliers, k as the number of iterations, P as the probability of successful model fit, and n as the minimum number of samples.

[0015] Furthermore, set the outlier ratio. =30%, maximum number of iterations k=1000, ensure P>0.99999, use median absolute deviation to automatically estimate residual threshold to achieve adaptive discrimination of outliers.

[0016] Overall, the technical solutions conceived in this application have the following beneficial effects compared with the prior art: Nuclear magnetic resonance (NMR) is a powerful method for characterizing the microstructure of materials. However, it is currently not applicable to the characterization of high-entropy alloys. This is because the metallic elements in high-entropy alloys generally have strong nuclear quadrupole moments, low thermal abundance, and small gyromagnetic ratios, resulting in problems such as broad spectral lines, weak signals, and long measurement times. Typically, even after tens of hours of measurement, only a broad peak of a few thousand ppm can be measured, making it impossible to characterize the microstructure of high-entropy alloys.

[0017] Long-term research has revealed that the interstitial spaces in refractory high-entropy alloys possess reversible hydrogen storage properties, making them ideal carriers for introducing "hydrogen probes." Based on this, this invention proposes a novel method for indirectly analyzing the chemical short-range in high-entropy alloys using hydrogen nuclear magnetic resonance (NMR). The core of this method lies in the fact that hydrogen atoms, acting as interstitial solutes, can sensitively "sensitize" and "encode" the local chemical environment composed of surrounding metal atoms through their NMR signals, thereby indirectly reflecting the microstructure of CSRO.

[0018] Specifically, after a hydrogen atom enters the alloy lattice, it occupies an interstitial position (such as a tetrahedral or octahedral interstitial position) surrounded by neighboring metal atoms. The type, number, and spatial arrangement of the surrounding metal atoms—the specific manifestation of CSRO—determine the local electronic structure of the hydrogen atom, such as electron cloud distribution and chemical bonding characteristics. This unique local environment modulates the effective magnetic field experienced by the hydrogen nucleus, which in turn manifests as precise changes in characteristic chemical shifts in its NMR spectrum. Different CSRO configurations correspond to different local atomic coordination modes, thereby inducing hydrogen to produce distinctive "fingerprint" spectral peaks. Therefore, by analyzing the peak position, line shape, and relative intensity of different chemical shift peaks in the ¹H NMR spectrum, the various characteristic local environments in which the hydrogen atom resides can be deduced, thereby inferring the short-range ordered configuration and statistical distribution of specific metal atoms in the alloy, ultimately achieving indirect detection and quantitative analysis of CSRO. In short, the method of this invention treats the NMR signal of hydrogen as an atomic-scale "in-situ sensor", transforming the atomic arrangement and structure information that is difficult to observe directly into a high-resolution spectroscopic signal that can be accurately measured and analyzed, providing a new approach to understanding the complex structure of high-entropy alloys at the microscale.

[0019] Compared to NMR, which directly detects refractory metal nuclei (such as 51V and 93Nb), hydrogen NMR offers a series of fundamental advantages in characterizing the short-range chemical properties of high-entropy alloys. Specifically, in terms of physical parameters, the advantage of hydrogen NMR over refractory metal NMR is orders of magnitude. ¹H nuclei have a spin of I=1 / 2, a natural abundance approaching 100%, and a high gyromagnetic ratio, resulting in extremely high intrinsic signal intensity. In contrast, refractory metal nuclei not only typically have a lower gyromagnetic ratio, but their spin quantum number I>1 / 2 leads to a catastrophic strong quadrupole moment effect. This makes the detection sensitivity of hydrogen 4-5 orders of magnitude higher than that of refractory metal nuclei. In terms of resolution, ¹H NMR benefits from the absence of quadrupole moment interference, allowing for spectral linewidths as narrow as 1-10 ppm (approximately hundreds to thousands of Hz) under solid-state magic angle rotation, clearly resolving chemical shift differences as small as 1 ppm. In contrast, the spectral lines of refractory metal nuclei often broaden to hundreds to thousands of ppm due to quadrupole moment broadening. It is evident that, in terms of resolution, hydrogen nuclear magnetic resonance (¹H NMR) lines are 2-3 orders of magnitude higher than those of refractory metal nuclei. Therefore, ¹H NMR, as an indirect probe, possesses overwhelming experimental advantages in terms of signal intensity, spectral quality, and information interpretation. These specific numerical comparisons demonstrate that, from an experimental physics perspective, ¹H NMR is an ultra-sensitive, ultra-resolution probe, while directly probing refractory metal nuclei presents an extremely low-sensitivity, extremely low-resolution challenge. Therefore, utilizing a "hydrogen probe" to indirectly characterize CSRO is an efficient way to bypass physical limitations and transform unmeasurable signals into measurable, high-quality information.

[0020] Compared with other microstructure characterization techniques, nuclear magnetic resonance (NMR) has the following irreplaceable advantages in resolving the short-range chemical order of high-entropy alloys: First, NMR provides macroscopically statistically representative information. As a bulk detection technique, its signal originates from the contribution of all observed nuclei (e.g., ¹H) within the entire sample. Therefore, the obtained CSRO information reflects the statistical average state of the material, effectively avoiding the local sampling bias that may exist in micro-area observation techniques (e.g., TEM, APT), and its results are more representative of the overall material. Second, NMR has intrinsically high sensitivity and quantitative resolution potential for CSRO configurations. The chemical shift of ¹H NMR directly and sensitively reflects the local electronic environment of the hydrogen nucleus, which is mainly determined by the type, number, and arrangement of its neighboring metal atoms (i.e., CSRO). The chemical interactions between different refractory metal elements (e.g., Ti, V, Nb) and hydrogen are significantly different. Even with similar atomic numbers, their electronic structures and bonding characteristics are quite different, which provides a physical basis for distinguishing different CSRO configurations through chemical shifts. Meanwhile, the ¹H nucleus itself possesses extremely high NMR sensitivity, enabling the effective detection and conversion of these differences in microscopic chemical environment into distinguishable peak signals, laying the foundation for subsequent quantitative analysis based on peak intensity. Finally, NMR offers significant advantages in experimental operation. Sample preparation is extremely simple, requiring no electron-transparent thin films or atomic probe tips; bulk or powder samples can be used directly for testing, greatly reducing sample preparation difficulty and the risk of introducing artifacts. In terms of measurement efficiency, the acquisition of a single ¹H NMR spectrum can typically be completed within one minute, offering high efficiency and supporting high-throughput, rapid screening, and in-situ dynamic processes. Attached Figure Description

[0021] Figure 1 This is a characterization method for short-range chemical ordering of high-entropy alloys in this embodiment of the invention.

[0022] Figure 2a This is the atomic model of T50 alloy after hydrogenation.

[0023] Figure 2b This is a schematic diagram of the local environment of the face-centered cubic structure in the atomic model of T50 alloy after hydrogenation.

[0024] Figure 3 The peaks of CSRO dominated by each element obtained from simulation calculations were observed in NMR. 1 Relative position on the H spectrum.

[0025] Figure 4 Ti in the embodiments of the present invention 50 Zr 18 Nb 15 V 12The as-cast Al5 high-entropy alloy samples, homogenized annealed samples, and samples aged for 1, 3, and 5 days were measured at a rotation speed of 18 kHz. 1 H spectrum.

[0026] Figure 5 Ti in the embodiments of the present invention 50 Nb 17 Zr 13 Measurements were taken at 18 kHz for as-cast and homogenized annealed V8Al8Cr4 high-entropy alloy samples. 1 H spectrum.

[0027] Figure 6 Ti in the embodiments of the present invention 30 V 30 Nb 20 Zr 20 The high-entropy alloy sample was measured at a rotation speed of 18 kHz after homogenization annealing. 1 H spectrum.

[0028] Figure 7a A high-angle annular dark-field transmission electron microscope atomic image of a T50 high-entropy alloy sample aged for 5 days

[110] , in which, Figure 7a The upper right illustration corresponds to the Fast Fourier Transform spectrum.

[0029] Figure 7b for Figure 7a The image shows the inverse Fourier transform image after extracting additional diffraction spots, with typical CSRO domains marked by yellow dashed circles.

[0030] Figure 8a A high-angle annular dark-field transmission electron microscope atomic image of a T50 high-entropy alloy sample aged for 7 days

[110] , wherein, Figure 8a The upper right illustration corresponds to the Fast Fourier Transform spectrum.

[0031] Figure 8b for Figure 8a The image shows the inverse Fourier transform image after extracting additional diffraction spots, with typical CSRO domains marked by yellow dashed circles. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0033] The embodiments of the present invention are implemented under the premise of the technical solution of the present invention, and detailed implementation methods and processes are given. However, the protection scope of the present invention is not limited to the following embodiments. The process parameters in the following embodiments that do not specify specific conditions are generally in accordance with conventional conditions.

[0034] The endpoints and any values ​​of the ranges disclosed in this invention are not limited to the precise ranges or values, and these ranges or values ​​should be understood to include values ​​close to these ranges or values. For numerical ranges, the endpoint values ​​of the various ranges, the endpoint values ​​of the various ranges and individual point values, and individual point values ​​can be combined with each other to obtain one or more new numerical ranges, which should be considered as specifically disclosed in this invention.

[0035] The embodiments of this application are described below with reference to the accompanying drawings. Process parameters not specifically specified in the following embodiments are generally performed under conventional conditions.

[0036] The method of this invention mainly includes the following steps: S1: The high-entropy alloy sample to be tested is subjected to hydrogen treatment to introduce sufficient hydrogen atoms into the crystal lattice structure formed by the high-entropy alloy atoms, so that the hydrogen NMR spectrum signal can be detected by nuclear magnetic resonance equipment. S2: Measurement of the entire sample 1 H NMR spectrum S3: Yes 1 The analysis focuses on the NMR Knight shifts, represented by the horizontal axis of the H NMR spectrum. Specifically, it involves statistically analyzing the Knight shift values ​​generated by each hydrogen atom in its corresponding elemental environment, establishing a relationship between the elements composing the high-entropy alloy sample and these Knight shift values, and resolving the Knight shift value corresponding to each element and its corresponding NMR value in the NMR spectrum. 1 The relative positions of the H NMR spectra, according to 1 Peaks in the H NMR spectrum represent the short-range chemical ordered structure of the corresponding element.

[0037] In one embodiment of the present invention, a high-entropy alloy sample prepared by arc melting was subjected to hydrogenation treatment at 600°C for 20 min under a hydrogen pressure of 3 MPa. Under these conditions, the hydrogen content of the sample was H / M = 1.69. The sample was then ground into powder and subjected to nuclear magnetic resonance magic angle rotation (MAS) experiments.

[0038] In practice, the hydrogenation treatment specifically involves holding the high-entropy alloy sample prepared by arc melting at a hydrogen pressure of at least 3 MPa at a high temperature of 500℃~600℃ for at least 20 minutes. The hydrogenation treatment should continue until the number of H atoms in the high-entropy alloy sample is no less than 1.69 times the total number of metal element atoms. If the hydrogen pressure is too low, insufficient hydrogen will enter the metal, resulting in a weak NMR signal. If the temperature is too low, the hydrogenation temperature will not be reached, and the diffusion rate of hydrogen atoms will be extremely low, making it difficult for them to enter the metal. If the temperature is too high, the dehydrogenation temperature may be reached, and hydrogen will be released from the metal, hindering hydrogen entry. If the holding time is too short, hydrogen diffusion will not be sufficient to allow enough hydrogen to enter the sample.

[0039] In one embodiment of the present invention, the nuclear magnetic resonance magic angle rotation experiment is performed as follows: The sample powder to be tested was encapsulated in a rotor adapted to the NMR instrument for MAS experiments. The pulse was set to solid-state echo pulse, the rotation speed was set to 18 kHz, and the 90° pulse duration was set to 3 μs. The sample was measured at room temperature. 1 H NMR spectrum. Due to the dense distribution of protons in the solid sample, strong... 1 H- 1 Homonuclear dipole-dipole interactions (H-D) lead to severe spectral broadening, masking differences in local chemical environments. High-speed magic-angle rotation at 18 kHz can partially average this anisotropic interaction, effectively narrowing the spectral lines and suppressing rotational sidebands, thus obtaining high-resolution spectra sufficient to resolve different interstitial sites (such as tetrahedral or octahedral voids). Setting the 90° pulse width to 3 μs aims to obtain a sufficiently wide apparent bandwidth to ensure coverage of the broad frequency shifts caused by lattice distortion and chemical disorder in high-entropy alloys. Simultaneously, this pulse power is within the probe's safe load range, ensuring the uniformity of the RF field.

[0040] Subsequently, simulation calculations and analyses of the NMR Knight shift were performed on the experimentally determined system. The calculations were performed using the VASP software package based on density functional theory (DFT). The interaction between ions and electrons was described using the projection enhanced wave (PAW) pseudopotential method, while the exchange correlation effect between electrons was calculated using the Perdew-Ernzerhof (PE) functional based on the generalized gradient approximation (GGA).

[0041] By statistically analyzing the chemical shifts generated by the chemical environment corresponding to each hydrogen atom, a large table of elements and chemical shifts was created. Machine learning was then used to statistically analyze this data, ultimately clearly revealing the Knight's shift value for each element in the high-entropy alloy system and their corresponding chemical shifts. 1The relative positions on the H spectrum can be used to quickly and accurately identify which chemical short-range ordered structure the peak in the proton NMR spectrum corresponds to, and its change process.

[0042] It should be noted that, due to the special nature of nuclear magnetic resonance equipment, the characterization in this application is limited to non-magnetic materials, mainly for the characterization of the chemical short-range ordered structure of refractory high-entropy alloys.

[0043] Nuclear magnetic resonance (NMR) characterization can directly reflect the existence of short-range chemical order and changes in local coordination environment. Combined with inter-element mixing enthalpy, the evolution trend of CSRO (Chemical Short-Range Order) can be qualitatively inferred. This is a technical discovery made by the applicant during long-term research. This invention introduces first-principles calculations based on density functional theory (DFT) to simulate ¹H NMR chemical shifts and combine them with NMR spectra to jointly analyze T50 high-entropy alloys (T50 high-entropy alloys refer to Ti...). 50 Zr 18 Nb 15 V 12 The types and dynamic evolution of CSROs in Al5 are described. In the actual experiment, the VASP software package can be used to perform DFT calculations. The interaction between ions and electrons is described by the projection-enhanced wave (PAW) pseudopotential, and the electron exchange correlation functional is selected from the PBE functional under the generalized gradient approximation (GGA) framework. The above describes the specific calculation processing methods in the experiment, which is not the core content of this invention application.

[0044] In one embodiment of the present invention, to simulate the local atomic environment of the hydrogenated alloy, a face-centered cubic supercell containing 32 metal atoms was constructed, and hydrogen atoms were filled in all tetrahedral interstices, ultimately forming a supercell structure containing 32 metal atoms and 64 hydrogen atoms. See [link to relevant documentation]. Figure 1 , Figure 1Figure 2 shows the atomic model of the hydrogenated T50 alloy. The supercell was structurally optimized to obtain the stable configuration with the lowest energy (in first-principles calculations (such as VASP) or molecular dynamics simulations, structural optimization is often called geometric optimization or relaxation, which refers to the process of finding the lowest energy point of the system by continuously adjusting atomic coordinates and cell parameters). Figure 2 is a schematic diagram of the local environment of the face-centered cubic structure in the atomic model of the hydrogenated T50 alloy. Each hydrogen atom occupies a tetrahedral interstitial space, denoted by a T-site, and is coordinated by four nearest-neighbor metal atoms. This atomic-level coordination environment directly corresponds to the local chemical configuration of CSRO. First-principles calculations simulated the NMR chemical shifts of hydrogen atoms in this environment, establishing a quantitative correlation with the experimental ¹H NMR (¹H Nuclear Magnetic Resonance) peak positions, thus providing atomic-scale theoretical support for resolving the types and evolution trends of CSRO.

[0045] Based on first-principles simulations using density functional theory, this invention obtained T50 metal hydride (specifically, Ti). 50 Zr 18 Nb 15 V 12 The theoretical ¹H NMR chemical shifts of all hydrogen atoms in Al5 were obtained. Statistical analysis of the local coordination environment of hydrogen atoms clarified the types and numbers of the four nearest-neighbor metal atoms for each hydrogen atom, as shown in Table 1. This four-coordinate metal atom configuration directly corresponds to a short-range chemically ordered structure at the atomic scale. By correlating different CSRO configurations with the theoretical chemical shifts of the corresponding hydrogen atoms, the contribution weight of each metal element to the ¹H NMR chemical shifts can be quantitatively analyzed.

[0046] In one embodiment of the present invention, the chemical shift contribution of each metal element is set as an unknown variable, and the theoretical chemical shift is used as the response value to construct a system of five linear equations containing 64 sets of samples: y=Xx Where y is the chemical shift vector, X is the metal atom coordination configuration matrix, and x is the contribution vector of the element to be determined.

[0047] Because outliers may exist in theoretical calculation data due to structural relaxation or numerical errors, conventional least squares (OLS) is susceptible to fitting bias. To obtain robust and reliable statistical solutions, this invention employs the Random Sample Consensus (RANSAC) algorithm to screen outliers and fit the model. To avoid instability caused by data coplanarity or degradation, a minimum sample size of n=6 is selected; calculations are performed according to the following RANSAC probability convergence formula: 1 (1 (1 ) n ) k =P in, denoted as the proportion of outliers, k as the number of iterations, P as the probability of successful model fit, and n as the minimum number of samples.

[0048] When performing specific data processing, assume the proportion of outliers. =30%, to ensure P>0.99999, the maximum number of iterations k=1000; simultaneously, the Median Absolute Deviation (MAD) is used to automatically estimate the residual threshold to achieve adaptive outlier identification. Data filtering results show that 12 significant outliers were excluded. Linear fitting was performed on the remaining inliers to obtain the element contribution vector: x=[Al: 7.945745, Nb: 7.293364, Zr: 6.915278, Ti: 5.569865,V: 5.395299] The corresponding contributions of each element in T50 to the chemical shift are shown in Table 2. The goodness of fit of this result is R0. 2 =0.884799, and the mean absolute error (MAE) is 0.311644, indicating that the model fits well and can quantitatively analyze the contribution of each metal element to the ¹H NMR chemical shift, providing a direct theoretical basis for understanding the correspondence between CSRO structure and NMR peaks.

[0049] Table 1. Statistical analysis of chemical shift values ​​of each hydrogen atom after hydrogenation treatment of T50 metal hydrides.

[0050] In Table 1, 1, 2, 3, and 4 represent the number of metal atoms surrounding a hydrogen atom, and the serial number in the table indicates which hydrogen atom it is. For example, the row with serial number 1 indicates that the four metal atoms surrounding the hydrogen atom are 1 Ti, 1 V, 1 Zr, and 1 Al, while the row with serial number 64 indicates that the four metal atoms surrounding the hydrogen atom are 2 V and 2 Zr.

[0051] Table 2 shows the chemical shift values ​​of each element in T50.

[0052] Based on the chemical shift contribution vectors of each metal element obtained by RANSAC fitting, this invention can quantitatively predict the ¹H NMR chemical shifts corresponding to different chemical short-range order (CSRO) configurations, and then plot the relative peak distribution of various CSROs on the NMR spectrum. Figure 3 The peaks of CSRO dominated by each element obtained from simulation calculations were observed in NMR. 1 As shown in the figure, the relative positions on the NMR spectrum indicate that the leftmost peak represents the Al-rich CSRO structure, followed by Nb-rich, Zr-rich, and Ti-rich peaks to the right. Based on the aforementioned enthalpy of mixing relationship between elements and the calculation results in the figure, it can be inferred that in the T50 alloy, the initial As-cast sample was Al-rich. After WQ (homogenization annealing and water quenching), the peak shifted significantly to the right, indicating Zr enrichment. As the aging time increased, the peak continued to shift to the right, thus transforming the Zr-rich structure into a Ti-rich structure during aging. This provides direct theoretical support for verifying the CSRO evolution mechanism.

[0053] This invention employs the VASP software package based on density functional theory (DFT) for computational simulation. The interaction between ions and electrons is described using the projection-enhanced wave (PAW) pseudopotential method, while the exchange-correlation effect between electrons is handled using the PBE (Perdew-Burke-Ernzerhof) functional based on the generalized gradient approximation (GGA). The valence electron configurations of each element are as follows: hydrogen is 1s... 1 Titanium is 3p 6 3D 3 4s 1 Vanadium is 3p 6 3D 4 4s 1 Niobium is 4p 6 4d 5 5s 1 Zirconium is 4s 2 4p 6 4d 3 5s 1In the calculations, the plane wave cutoff energy was set to 500 eV, and the Brillouin zone integral used a 2×2×2 Γ-center k-point grid. The energy convergence criterion for structural optimization was set to 10 eV. -4 eV, force convergence criterion set to 10 -3 eV / Å is used to ensure the accuracy and reliability of the simulation results.

[0054] Furthermore, in one embodiment of the present invention, for Ti 50 Zr 18 Nb 15 V 12 Al5 was heat-treated to study changes in its short-range chemically ordered structure, which was then characterized using nuclear magnetic resonance (NMR). The specific experiments were as follows: Aging treatment: Ti with the same composition 50 Zr 18 Nb 15 V 12 Al5 samples were annealed at 300°C for 1 day, 3 days and 5 days respectively, and then cooled in the furnace to form different types of short-range chemical structures.

[0055] Hydrogenation treatment: The hydrogenation reaction was carried out using a self-made Siefert-type apparatus. First, the treated high-entropy alloy Ti... 50 Zr 18 Nb 15 V 12 The Al5 sample was evacuated at 573 K for 1 hour, then hydrogenated in 7N hydrogen gas and annealed at 873 K and 3 MPa hydrogen pressure for 20 minutes. The hydrogenated sample was then ground into powder and sieved through a 200-mesh sieve for use in proton nuclear magnetic resonance spectroscopy.

[0056] 7N is a common purity representation method in the ultra-high purity gas industry. The letter N represents "Nine" and the number 7 indicates that the purity contains seven consecutive 9s. Therefore, 7N purity hydrogen means that the purity of hydrogen reaches 99.99999%, and the corresponding total impurity content is ≤0.01 ppm (ppm is one part per million, and 0.01 ppm means that impurities account for only 0.01 parts per million parts of gas).

[0057] Nuclear magnetic resonance (NMR) characterization: 2g of the sample powder to be tested was encapsulated in a rotor adapted to the NMR instrument for MAS experiments. The pulse was set to solid-state echo pulse, the rotation speed was set to 18 kHz, the 90° pulse duration was set to 3 μs, and the 1H NMR spectrum of the sample was measured at room temperature.

[0058] First-principles simulations: Subsequently, simulations of the experimentally determined NMR Knight shift were performed on the system. The simulations were conducted using the VASP software package based on density functional theory (DFT). The interaction between ions and electrons was described using the projection-enhanced wave (PAW) pseudopotential method, while the exchange correlation effect between electrons was calculated using the Perdew-Ernzerhof (PE) functional based on the generalized gradient approximation (GGA). The kinetic energy cutoff was set to 500 eV, and a 2×2×2 gamma k mesh was used in the Brillouin zone. The energy convergence criterion was set to 1×10⁻⁶. -4 eV / atom, force convergence criterion is 1×10⁻⁶ -3 To ensure the accuracy and reliability of the simulation results, the following steps are taken: eV / A Construct a 2x2x2 primitive face-centered cubic supercell structure and insert hydrogen atoms into all tetrahedral interstitial positions to ensure bulk system stability. Calculate the Knight's shift value corresponding to each hydrogen atom and record the types of atoms at the tetrahedral interstitial vertices surrounding that hydrogen atom, forming a statistical table of different chemical shift values ​​corresponding to different chemical environments.

[0059] The Knight's displacements for each of the five elements—Ti, Zr, V, Nb, and Al—are labeled a, b, c, d, and e, respectively. The table above then transforms this into a system of 64 five-element linear equations. Statistical analysis of this data using machine learning allows for a clear determination of the Knight's displacement values ​​for each element in this high-entropy alloy system and their corresponding values ​​within the system. 1 The relative positions on the H spectrum can be used to quickly and accurately identify which chemical short-range ordered structure the peak in the proton NMR spectrum corresponds to, and how it changes.

[0060] Figure 4 For Ti 50 Zr 18 Nb 15 V 12 Measurements were taken at 18 kHz for as-cast Al5 high-entropy alloy samples, homogenized annealed samples, and samples aged for 1, 3, and 5 days. 1 The H spectrum shows that "Aged 5d" indicates a sample annealed at 300℃ for 5 days followed by furnace cooling, "Aged 3d" indicates annealed at 300℃ for 3 days followed by furnace cooling, and "Aged 1d" indicates annealed at 300℃ for 1 day followed by furnace cooling. "WQ" indicates a sample homogenized at 1000℃ for 2 hours followed by water quenching. "As-cast" represents Ti. 50 Zr 18 Nb 15 V 12Al5 high-entropy alloy as-cast sample. As shown in the figure, there are several impurity peaks next to the main peak in the NMR spectrum of the as-cast sample, indicating that the Al-containing high-entropy alloy system originally contained a small amount of inhomogeneous structure.

[0061] WQ represents a sample homogenized and annealed at 1000℃ for 2 hours followed by water quenching. Observation of its curves reveals that after homogenization and annealing, a distinct peak representing a chemically short-range ordered structure appears to the left of the main peak. With increasing aging time, this peak gradually shifts to the right, indicating that the chemically short-range ordered structure in this system becomes increasingly apparent with homogenization and annealing, and then changes with the duration of low-temperature aging. Simulation calculations and statistical results show that the peak corresponding to the non-uniform structure displayed in the as-cast state corresponds to an Al-rich structure. The peak appearing after homogenization and quenching represents a Zr-rich short-range ordered structure. With increasing annealing days, this structure changes due to elemental interactions, gradually transforming into a Ti-rich structure.

[0062] Specifically, the NMR spectrum of the as-cast sample exhibits multiple heterogeneous peaks, mainly because Al and the five components Ti, Zr, Nb, and V all have negative enthalpy of mixing. During non-equilibrium solidification, the strong interaction between Al and different metal atoms induces the formation of various types of CSROs, resulting in hydrogen atoms being in a highly heterogeneous local coordination environment, which directly reflects the compositional segregation and structural inhomogeneity of the as-cast system. The NMR spectrum of the homogenized annealed + water-quenched state (WQ sample) shows a single main peak (Peak 2) accompanied by a characteristic secondary peak (Peak 1). In this state, after high-temperature homogenization annealing and water quenching, the system completes atomic rearrangement and segregation elimination under thermodynamic drive. Combining simulation results with mixing enthalpy analysis, Peak 1 corresponds to a Zr-rich CSRO. This is because the strong negative mixing enthalpy of Zr and Al preferentially promotes the formation of a locally ordered Zr-rich structure, becoming the dominant CSRO type outside of the disordered solid solution, manifested as a single secondary resonance peak in the spectrum. During the isothermal aging stage, Peak 1 gradually shifts to the right, indicating that with prolonged aging time, atoms undergo further short-range diffusion and rearrangement. Simulation results show that this shift corresponds to a gradual transformation of the CSRO from Zr-rich to Ti-rich, perfectly matching the rightward shift trend of Peak 1 in the experimental spectrum, confirming that the CSRO undergoes a dynamic evolution from Zr-rich to Ti-rich during aging.

[0063] Figure 5 For Ti 50 Nb 17 Zr 13 Measurements were taken at 18 kHz for as-cast and homogenized annealed V8Al8Cr4 high-entropy alloy samples. 1The H-spectrum shows that, even in the as-cast state, the sample exhibits a small impurity peak to the left of the main peak. However, after homogenization annealing, the sample gradually becomes more homogeneous due to the interactions between elements in the system, corresponding to the nuclear magnetic resonance (NMR) spectrum. 1 The disappearance of impurity peaks in the H spectrum.

[0064] Figure 6 For Ti 30 V 30 Nb 20 Zr 20 The high-entropy alloy sample was measured at a rotation speed of 18 kHz after homogenization annealing. 1 The H-ray spectra show that Al has been removed from this system, as some scholars believe that Al is the main element that promotes the formation of a chemically ordered short-range structure in this system. After homogenization annealing, this sample did not exhibit a bimodal structure similar to sample 1, indicating that this system does not contain a chemically ordered short-range structure. This conclusion is clearly reflected in the H-ray NMR spectrum.

[0065] To verify the scientific validity and correctness of the method of this invention, Ti was used. 50 Zr 18 Nb 15 V 12 The chemical short-range structure of Al5 alloy was characterized by transmission electron microscopy and compared with the nuclear magnetic resonance characterization method.

[0066] Figure 7a A high-angle annular dark-field transmission electron microscope atomic image of a T50 high-entropy alloy sample aged for 5 days

[110] , in which, Figure 7a The upper right inset corresponds to the Fast Fourier Transform (FFT) graph. Figure 7b for Figure 7a The image corresponds to the inverse Fourier transform image after extracting additional diffraction spots. Typical CSRO domains are marked by yellow dashed circles. Figure 8a A high-angle annular dark-field transmission electron microscope atomic image of the

[110] zone axis of a T50 high-entropy alloy sample aged for 7 days, wherein, Figure 8a The upper right inset corresponds to the Fast Fourier Transform (FFT) graph. Figure 8b for Figure 8a The image corresponds to the inverse Fourier transform image after extracting additional diffraction spots. Typical CSRO domains are marked by yellow dashed circles. Figure 7a and Figure 8a The atomic structure morphology of T50 high entropy alloy under the high angle annular dark field (HAADF) under the zone axis

[110] after isothermal aging at 300 ℃ for 5 days and 7 days, respectively, and the corresponding fast Fourier transform (FFT) and inverse Fourier transform (IFFT) analysis results are shown, which confirm the existence of CSRO in the aged samples and provide experimental support for subsequent nuclear magnetic resonance characterization.

[0067] Figure 7a neutralization Figure 8a The HAADF atomic images show clear BCC lattice fringes, indicating that the alloy maintains a stable single-phase BCC matrix structure in both aging states. Figure 7a neutralization Figure 8a In the FFT patterns embedded in the upper right corner, the red dashed circles mark the intrinsic Bragg diffraction spots of the BCC matrix, while the yellow dashed circles mark the additional diffraction spots induced by the CSRO structure. The appearance of these additional diffraction spots directly confirms that aging treatment can induce short-range chemical ordering at the atomic scale, and that their intensity increases significantly with the extension of aging time. Figure 7b and Figure 8b The images shown are IFFT images obtained after extracting additional diffraction spots. The pink signal points correspond to the spatial distribution of the CSRO region, and the yellow dashed circles mark typical CSRO domains.

[0068] The comparison shows that the number, size, and distribution density of CSRO regions in the 7-day aged sample are significantly higher than those in the 5-day aged sample. At 5 days of aging, CSRO domains are sparser and smaller, only sporadically distributed in local areas. This phenomenon indicates that as the aging time increases from 5 days to 7 days, the short-range diffusion of atoms driven by thermodynamics becomes more complete, and atomic pairs undergo further local rearrangement, leading to an increase in the nucleation density and domain size of CSRO domains, ultimately resulting in a continuous increase in the area fraction of LCO regions.

[0069] It is important to note that while transmission electron microscopy (TEM) and related techniques can reveal the presence and abundance of CSRO, their spatial resolution and elemental resolution remain significantly limited. Firstly, TEM characterization can only indirectly reflect the area fraction of LCO through diffraction spot intensity and IFFT signal density, failing to accurately represent the structure and evolutionary trends of CSRO. Secondly, TEM observation relies on local regions, making it difficult to characterize the overall CSRO in a sample and capture subtle changes in elemental coordination environments at the atomic scale. Therefore, to deeply analyze the specific structural evolution and class changes of CSRO during aging, it is necessary to combine high-resolution magic-angle rotating NMR and first-principles calculations with more refined characterization and simulation methods to reveal the dynamic evolution mechanism of CSRO.

[0070] The first-principles simulation of this invention specifically refers to first-principles calculation, which is a theoretical simulation method based on the fundamental laws of quantum mechanics. Its core lies in directly solving the Schrödinger equation or its equivalent form for many-body systems. The most significant feature of this method is that it "starts from scratch," requiring only the atomic types and spatial coordinates of the constituent materials as input, without relying on any empirical parameters or fitting potential functions from experiments. By calculating the electron density distribution and total energy of the system, first-principles calculations can accurately predict the thermodynamic stability, crystal structure, electronic band structure, and mechanical properties of materials at the atomic and electronic scales. This is an important means of deeply revealing the essential connection between the local chemical environment and macroscopic properties in high-entropy alloys.

[0071] The VASP software package refers to VASP, a high-performance computer program for atomic-scale material modeling. Based on the principles of quantum mechanics, it can predict material properties by simply inputting the types and positions of atoms, without relying on any experimental parameters.

[0072] This invention provides a nuclear magnetic resonance (NMR) characterization method for chemically short-range ordered structures in high-entropy alloys. The method involves first hydrogenating the sample, then acquiring the 1H NMR spectrum of the high-entropy alloy sample. Based on the NMR spectrum, simulation calculations are performed to establish a correspondence between characteristic peak positions in the spectrum and the atomic configurations of specific chemically short-range ordered structures. Based on this correspondence, the type of chemically short-range ordered structure represented by the characteristic peak position is determined. The innovation of this invention lies in using hydrogen as a probe to indirectly and rapidly locate the chemical environment of chemically short-range ordered structures in high-entropy alloys. Transmission electron microscopy (TEM) characterization of chemically short-range ordered structures is very complex, and while NMR can characterize metal elements in high-entropy alloys, its direct characterization signal for metal elements is not strong. The NMR spectra of metal elements often exhibit a large, broad peak, making it impossible to distinguish the presence of chemically short-range ordered structures even with MAS experiments. This invention proposes using hydrogen, with its very strong NMR signal, as a probe to indirectly reflect the surrounding chemically short-range ordered structures through the 1H NMR spectrum of high-entropy alloys. This method characterizes the entire sample, extending beyond a small region observed by TEM to represent the structure of the entire sample, fundamentally improving accuracy. The sample preparation process is simple and quick, eliminating sample pretreatment time. Furthermore, the acquisition of 1H NMR spectra is typically completed in less than one minute, compared to at least 3 hours for transmission electron microscopy characterization and sample pretreatment. This invention significantly improves characterization efficiency. The method utilizes first-principles calculations to analyze the aforementioned 1H NMR spectra, clearly identifying the positions of each short-range ordered structure corresponding to its peaks. Compared to characterization methods like transmission electron microscopy, NMR characterization is simpler in sample preparation and faster in detection time. The combination of NMR and simulation calculations greatly improves experimental efficiency and provides clearer analysis of the types of short-range ordered chemical structures.

[0073] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for characterizing the short-range chemical order of high-entropy alloys, characterized in that, It includes the following steps: S1: The high-entropy alloy sample to be tested is subjected to hydrogen treatment to introduce sufficient hydrogen atoms into the crystal lattice structure of the high-entropy alloy, so that a hydrogen NMR spectrum signal can appear in the nuclear magnetic resonance experiment. S2: measuring the whole of the sample 1 H nuclear magnetic resonance spectrum, S3: to 1 The hydrogen nuclear magnetic resonance spectrum is analyzed by analyzing the nuclear magnetic resonance Knight shift represented by the abscissa. Specifically, the number of Knight shifts of each hydrogen atom in the environment of the surrounding elements is counted to form a relationship between all elements of the high-entropy alloy sample to be tested and the number of Knight shifts, and the number of Knight shifts corresponding to each element and its position in the 1 H nuclear magnetic resonance spectrum are analyzed according to the 1 The peaks in the H nuclear magnetic resonance spectrum represent the chemical short-range ordered structure of the corresponding elements.

2. The method as described in claim 1, characterized in that, In step S1, the hydrogenation treatment specifically involves holding the high-entropy alloy sample under test at a hydrogen pressure of at least 3 MPa and a high temperature of 500°C to 600°C for at least 20 minutes.

3. The method as described in claim 2, characterized in that, In step S1, hydrogenation is performed until the number of H atoms in the high-entropy alloy sample to be tested is no less than 1.69 times the total number of metal element atoms.

4. The method as described in claim 3, characterized in that, In step S2, the hydrogen-treated sample is ground into powder, and at least 1 g of powder sample is randomly selected for NMR detection. The pulse is set to solid-state echo pulse, the rotation speed is set to 18 kHz, and the 90° pulse duration is set to 3 μs. The overall sample density is measured at room temperature. 1 H nuclear magnetic resonance spectrum.

5. The method as described in claim 1 or 4, characterized in that, In step S3, a face-centered cubic supercell containing 32 metal atoms is constructed, and hydrogen atoms are filled into all the tetrahedral interstices to form a supercell structure containing 32 metal atoms and 64 hydrogen atoms. In the supercell structure, each hydrogen atom occupies a tetrahedral interstitial space, and is surrounded by four nearest-neighbor metal atoms. Calculate the coordination environment of these nearest-neighbor metal atoms. 1 The Knight shift value of H is obtained by statistically analyzing the Knight shift value generated by each hydrogen atom in its surrounding elemental environment.

6. The method as described in claim 5, characterized in that, Statistical analysis of the nearest-neighbor metal atom coordination of hydrogen atoms determined the types and numbers of the four nearest-neighbor metal atoms for each hydrogen atom. This four-coordinate metal atom configuration directly corresponds to a short-range chemical order structure at the atomic scale. By comparing the theoretical Knight shifts of different chemical short-range ordered structures with the corresponding hydrogen atoms, the contribution weights of each metal element to the Knight shifts in the ¹H NMR spectrum were quantitatively analyzed.

7. The method as described in claim 6, characterized in that, The contribution of each metal element to the Knight shift in the ¹H NMR spectrum is set as an unknown variable. Using the theoretical Knight shift as the response value, a system of five linear equations with 64 samples is constructed: y=Xx Where y is the theoretical Knight displacement, X is the metal atom coordination configuration matrix, and x is the contribution vector of the element to be determined to the Knight displacement.

8. The method as described in claim 7, characterized in that, The random sampling consensus algorithm is used to screen outliers and fit the model in the dataset. The minimum sample size n=6 is selected, and the probability convergence formula of the random sampling consensus algorithm is used for calculation. 1 (1 (1 ) n ) k =P in, denoted as the proportion of outliers, k as the number of iterations, P as the probability of successful model fit, and n as the minimum number of samples.

9. The method as described in claim 8, characterized in that, Set outlier ratio =30%, maximum number of iterations k=1000, ensure P>0.99999, use median absolute deviation to automatically estimate residual threshold to achieve adaptive discrimination of outliers.

Images