A material analysis method, system and application based on crystal structure database

Abstract

A material analysis method, system, computer-readable storage medium and application based on a crystal structure database. The above-mentioned material analysis method includes comparing the experimental spectrum information detected by the sample to be tested with the theoretical spectrum information calculated from the material structure data in the crystal structure database, and obtaining the crystallographic information and phase composition of the sample to be tested through intelligent analysis; the crystallographic information includes space group, unit cell parameters, and specific coordinates of atoms in the unit cell; the crystal structure database has material structure data obtained by experimental measurement and / or theoretical prediction, including chemical formula, space group, unit cell parameters, and specific coordinates of atoms in the unit cell. The above-mentioned method adopts a new paradigm for material gene big data research and realizes automated crystallographic information and phase composition analysis through the crystal structure database. It is a new material analysis method; based on this method, large-scale, high-throughput and high-precision material analysis can be achieved, laying the foundation for building a material science spectrum data mining and analysis platform.

Description

Technical Field

[0001] The present application relates to the field of material analysis technology, and in particular to a material analysis method, system and application based on a crystal structure database. Background Art

[0002] Based on the fundamental principles of the interaction between matter and waves, researchers have developed structural analysis methods such as X-ray diffraction, neutron diffraction, electron diffraction, infrared spectroscopy, and Raman spectroscopy. By analyzing the resulting spectral data, they can obtain information such as the sample's phase composition, lattice parameters, and residual stress. This process involves the construction, analysis, and retrieval of a materials database.

[0003] Currently, all openly available diffraction databases are experimentally derived, subject to experimental errors and lacking regularity in data structure distribution, which hinders retrieval speed and accuracy. Diffraction databases derived from reverse theoretical calculations based on crystal structure databases are free of experimental errors, offer high data accuracy, and boast a massive data volume. They also directly link diffraction pattern data to crystal structure, representing a novel approach to building diffraction databases. Software such as MDI Jade and QualX, which accomplish the critical process of deriving search and matching conclusions from databases, are both developed internationally and are expensive to use. They also suffer from complex and cumbersome retrieval procedures, rigid functionality, and unsightly interfaces. To break the existing data paradigm, overcome technological monopolies, forge new paths, mitigate bottlenecks, and address deficiencies in existing data composition, analysis methods, and software technologies, there is an urgent need to develop new materials analysis technologies with independent intellectual property rights. Summary of the Invention

[0004] The purpose of this application is to provide a new material analysis method, system and application based on crystal structure database.

[0005] This application adopts the following technical solutions:

[0006] One aspect of the present application discloses a material analysis method based on a crystal structure database, comprising comparing experimental spectrum information obtained from the detection of a sample to be tested with theoretical spectrum information calculated from material structure data in the crystal structure database, and obtaining crystallographic information and related phase composition information of the sample to be tested through intelligent analysis; wherein the crystallographic information includes at least one of a space group, unit cell parameters, and specific coordinates of atoms in the unit cell; the crystal structure database has material structure data obtained by experimental measurement and / or theoretical prediction, and the material structure data includes a chemical formula, a space group, unit cell parameters, and specific coordinates of atoms in the unit cell.

[0007] It should be noted that the material analysis method of the present application uses known structural data to calculate and obtain theoretical spectrum information, and then compares the spectrum information of the sample to be tested obtained experimentally with the theoretical spectrum information, and obtains the crystallographic information and related phase composition information of the sample to be tested based on the intelligent analysis of the comparison results. Among them, the known structural data is obtained through experimental measurement or theoretical prediction, and is verified known material structure data; the theoretical spectrum information is obtained by calculation based on the known material structure data, and is a spectrum information simulated based on theoretical calculations. The present application uses the theoretical spectrum information obtained by theoretical calculation as a spectrum database for comparison, and performs comparison and analysis on the sample to be tested, which is a new paradigm for material analysis; it solves the defects of the existing material analysis method based on the experimental data of the standard for comparison, such as experimental errors and poor regularity of data structure distribution, and also solves the problems caused by this. In the material analysis method based on the crystal structure database of the present application, all spectrum information is obtained by calculation based on known material structure data. The spectrum information obtained by these calculations is highly accurate, does not have experimental errors, and has good distribution regularity, which is convenient for retrieval. Moreover, since it is obtained through calculation, it can accurately summarize all phase characteristics of crystal materials, greatly improving the accuracy of material analysis.

[0008] In one implementation of the present application, intelligent analysis includes:

[0009] (1) When the experimental spectrum information of the sample to be tested matches the theoretical spectrum information in the crystal structure database, the crystallographic information and related phase composition information of the sample to be tested are directly obtained based on the matching results;

[0010] (2) If the experimental spectrum information of the sample to be tested does not match the theoretical spectrum information of the crystal structure database,

[0011] 1) searching a crystal structure database for theoretical spectrum information of a structure isomorphic with the experimental spectrum information, refining the theoretical spectrum information of the isomorphic structure so that the refined theoretical spectrum information matches the experimental spectrum information, and obtaining crystallographic information and related phase composition information of the sample to be tested based on the matching results; in this application, refining the spectrum information or refining the spectrum refers to adjusting various parameters corresponding to the original spectrum, such as replacing known elements in the original spectrum to change the peak size, shape, position, etc. of the spectrum curve;

[0012] 2) For the case where no theoretical map information of the isomorphic structure is retrieved, a positive space search, simulated annealing and genetic algorithm are used to collaboratively build a structural model for the sample to be tested, and the theoretical map information of the sample to be tested is calculated based on the built structural model. The theoretical map information of the sample to be tested is refined so that the experimental map information obtained by the test sample is matched with the refined theoretical map information of the sample to be tested, and the crystallographic information and related phase composition information of the sample to be tested are obtained based on the matching result. It can be understood that the constructed structural model, the theoretical map information of the sample to be tested calculated based on the constructed structural model, and the refinement of the theoretical map information of the sample to be tested are all known information; therefore, if the experimental map information matches the refined theoretical map information of the sample to be tested, the crystallographic information and related phase composition information of the sample to be tested can be obtained. The corresponding theoretical map information of the sample to be tested, the obtained crystallographic information and related phase composition information of the sample to be tested, etc. can all be added to the crystal structure database to achieve iterative update of the database.

[0013] It should be noted that the matching of the present application refers to, for example, one implementation of the present application, the structural matching degree value of the experimental spectrum information of the sample to be tested and the theoretical spectrum information of the crystal structure database is calculated by the quality factor matching formula. The larger the matching degree value, the higher the matching degree, and vice versa. When the matching degree value is lower than 0.01, it is considered that the sample to be tested has not been matched in the crystal structure database. It can be understood that the judgment criteria or thresholds of the specific matching degree value can be adjusted and set according to needs. The calculation and threshold of the matching value of different spectrum types are different and are not specifically limited here.

[0014] In one implementation of the present application, a method for retrieving theoretical spectrum information of isomorphic structures includes retrieving theoretical spectrum information of isomorphic structures in a crystal structure database based on a peak intensity ratio relationship contained in experimental spectrum information.

[0015] It is understood that crystal structures that are isomorphic to the phase corresponding to the sample to be tested should have similar peak intensity ratios corresponding to different crystal plane indices; therefore, this application searches for isomorphism based on this peak intensity ratio. Based on the theoretical spectrum information corresponding to the isomorphic structure, refinement can obtain the crystallographic information and related phase information corresponding to the sample to be tested.

[0016] In one implementation of the present application, the intelligent analysis also includes, when the experimental spectrum information of the sample to be tested does not match the theoretical spectrum information of the crystal structure database, supplementing the obtained crystallographic information and related phase composition information of the sample to be tested into the crystal structure database to achieve iterative update of the crystal structure database.

[0017] For example, in one implementation of the present application, traditional crystallographic methods and machine learning methods are combined to realize automatic indexing of new structures, full spectrum fitting, structure refinement, and rapid identification of crystallographic information corresponding to sample spectra; rapid refinement obtains crystallographic information corresponding to new structures and updates it to the crystal structure database, realizing dynamic iterative upgrade of the database.

[0018] In one implementation of the present application, both the experimental spectrum information and the theoretical spectrum information include at least one of time / space spectrum information and energy spectrum information; wherein the time / space spectrum information includes at least one of an X-ray diffraction spectrum, a neutron diffraction spectrum, and an electron diffraction spectrum; and the energy spectrum information includes at least one of an emission spectrum, an absorption spectrum, and a scattering spectrum.

[0019] Preferably, the emission spectrum includes the emission spectrum (fluorescence spectrum) of an atom, a molecule or a material.

[0020] Preferably, the absorption spectrum includes the absorption spectrum of atoms, molecules or materials.

[0021] Preferably, the scattering spectrum is Raman spectroscopy.

[0022] In one implementation of the present application, the calculation method of the theoretical map information includes but is not limited to:

[0023] (1) Calculate time / space spectrum information based on the powder diffraction intensity integral formula;

[0024] (2) Calculating energy spectrum information by simulating crystal structure based on quantum chemical methods. Specifically, obtaining spectral information of emission spectrum, absorption spectrum and scattering spectrum by simulating atomic nuclear vibration of crystal structure based on quantum chemical methods. Obtaining energy spectrum information of emission spectrum, absorption spectrum and scattering spectrum by simulating electronic structure corresponding to crystal structure based on quantum chemical methods.

[0025] It should be noted that, in principle, all theoretical calculations or simulation map calculation methods can be used in this application and are not specifically limited here.

[0026] In one implementation of the present application, the material analysis method of the present application also includes collaboratively comparing at least two experimental spectrum information of the sample to be tested when it is difficult to perform accurate structural analysis on one experimental spectrum information of the sample to be tested, and obtaining the crystallographic information and related phase composition information of the sample to be tested through intelligent analysis.

[0027] Preferably, in the material analysis method of the present application, the intelligent analysis further includes quantitatively calculating the content percentage of each phase in the sample to be tested based on the phase composition information of the sample to be tested.

[0028] It should be noted that the key to this application is to use experimental spectrum information and theoretical spectrum information for comparison, so as to obtain the crystallographic information of the sample to be tested and the related phase composition information. The specific spectrum information to be used can be determined according to the specific experimental situation; for example, if the X-ray diffraction spectrum of the sample to be tested is measured, it is compared with the theoretical X-ray diffraction spectrum obtained by calculation from the corresponding crystal structure database; if the X-ray diffraction spectrum and atomic emission spectrum of the sample to be tested are measured at the same time, they are respectively compared with the theoretical X-ray diffraction spectrum and the theoretical atomic emission spectrum obtained by calculation from the corresponding crystal structure database. It can be understood that, in principle, the more experimental spectrum information obtained from the experiment of the sample to be tested, the more comparisons there are in the crystal structure database, but the corresponding results will be more accurate. Therefore, in one implementation of the present application, it is preferred to use at least two types of experimental spectrum information for comparison.

[0029] It should also be noted that the material analysis method of the present application can not only obtain the crystallographic information and related phase composition information of the sample to be tested, but also can quantitatively calculate the components of each phase according to needs. For example, taking Al2O3 as an example, if the sample contains multiple structural phases, the RIR value corresponding to the standard Al2O3 can be theoretically calculated to quantitatively calculate the percentage of the structure matching the characteristic peak of the sample spectrum in the crystal structure database, thereby realizing quantitative analysis of the content of each phase in the sample.

[0030] In one implementation of the present application, in a crystal structure database, the material structure data obtained by experimental measurement refers to the material structure data obtained by experimental characterization of a sample with a known structure; the material structure data obtained by theoretical prediction is calculated by at least one of the following calculation methods and obtained after eliminating repeated structures in the crystal structure database based on a graph theory method;

[0031] The aforementioned calculation methods include:

[0032] Method 1) The material structure data obtained by experimental characterization of samples with known structures are replaced with elements and then relaxed using density functional theory (DFT) to obtain new materials and their material structure data;

[0033] Method 2) Performing a structure search based on a particle swarm algorithm (PSO) and / or a genetic algorithm on the material structure data obtained by experimental characterization of samples with known structures to obtain new materials and their material structure data;

[0034] Method 3) is to extract features from the material structure data obtained by experimental characterization of samples with known structures, and reverse design is performed based on the extracted features to obtain new materials and their material structure data.

[0035] It should be noted that the graph theory method in this application, namely the method described in the applicant's previous patent application 201910390012.2, "A Method for Describing Material Crystal Structure," can be used to construct and obtain material structure data based on graph theory methods. It is understood that this new graph theory-based crystal structure description can be easily implemented using computer language, thereby achieving the purpose of deduplication of isomorphic structures; therefore, all technical information related to material crystal structure description in patent application 201910390012.2 is incorporated into this application for the purpose of explaining and illustrating the crystal structure database and its preparation method.

[0036] In one implementation of the present application, in the intelligent analysis of X-ray diffraction patterns (XRD diffraction patterns), the degree of matching of the diffraction peak positions and peak intensities corresponding to the experimental X-ray diffraction pattern information of the sample to be tested and the theoretical X-ray diffraction pattern information calculated based on the crystal structure database can be determined. Specifically, the structural matching degree value of the experimental pattern of the sample to be tested and the theoretical pattern is calculated by the quality factor matching formula. The larger the matching degree value, the higher the matching degree, and vice versa. When the matching degree value is higher than a certain value, for example, greater than or equal to 0.01, it is considered that the structure in the corresponding crystal structure database belongs to a phase in the sample to be tested. When the matching degree value is lower than a certain degree, such as lower than 0.01, it is considered that the sample to be tested has not been matched in the crystal structure database. Further isomorphism judgment is required. If there is an isomorphism match, analysis is performed according to the isomorphism match. If there is no isomorphism match, further analysis is required by combining traditional crystallographic methods, machine learning methods, and existing structures in the crystal database.

[0037] Specifically, the material analysis method based on X-ray diffraction patterns includes the following steps:

[0038] The experimental X-ray diffraction pattern information obtained from the test sample is compared with the theoretical X-ray diffraction pattern information calculated from the structural data of each material in the crystal structure database. If they match, the crystallographic information and related phase composition information of the test sample are directly obtained based on the matching results; if they do not match, the following operations are performed:

[0039] 1) Diffraction pattern indexing

[0040] The indexing method of the X-ray diffraction pattern of polycrystalline powder is to find the solution that satisfies the following equation within the range of experimental error, solve the lattice parameters including a, b, c and alpha, beta, gamma from the interplanar spacing corresponding to the diffraction peak, and determine the crystal plane index of the diffraction peak at the same time.

[0041] 2) Space group determination

[0042] Based on the systematic extinction caused by the presence of centered lattices, helical axes, and slip planes in the crystal, the lattice type and X-ray diffraction group of the crystal are determined after indexing; the space group is determined in combination with the statistical laws of diffraction intensity.

[0043] 3) Structural model construction

[0044] Relying on the support of a large amount of structural data in the crystal structure database, a search for isomorphic structures is first performed. Since a large number of crystal structures have been studied and entered into the database, the crystal structure of a new compound should first be checked to see if it is isomorphic with a compound of a known structure. Once the chemical formula, lattice parameters and space group of the new compound are determined, substances with similar chemical formulas, lattice parameters and the same space group can be searched in the structure database based on the number of atoms in the unit cell and the chemical properties of the elements, and the diffraction patterns of the two can be compared. The specific comparison method is to search for isomorphic structures in the crystal structure database based on the peak intensity information contained in the spectrum information and the crystal plane index information of the corresponding peak intensity obtained by indexing. The crystal structure that is isomorphic with the corresponding substance of the experimental sample should have a similar proportional relationship between the peak intensities corresponding to different crystal plane indices. If a compound with a similar diffraction pattern and a known crystal structure can be found, that is, theoretical spectrum information of an isomorphic structure, the distribution and rough position of atoms in the new compound can be determined. Rietveld refinement can then be used to match the refined theoretical spectrum information with the experimental spectrum information. Based on the matching results, the crystallographic information of the sample to be tested and related phase composition information can be obtained.

[0045] For the samples to be tested for which no isomorphic structure has been retrieved in the crystal structure database, traditional crystallographic methods and machine learning methods are combined, and positive space search, simulated annealing, genetic algorithms and other methods are used to collaboratively build a structural model. New theoretical map information is calculated based on the built structural model, and the new theoretical map information is refined so that the refined theoretical map information matches the experimental map information. Based on the matching results, the crystallographic information and related phase composition information of the sample to be tested are obtained.

[0046] In one implementation of the present application, the pattern information refinement or pattern refinement specifically employs Rietveld refinement, i.e., Rietveld structure refinement. This involves adjusting various parameters through fitting methods such as the least squares method to alter the peak size, shape, and position of the curve of the theoretical XRD diffraction pattern calculated based on the crystal structure database, so that the theoretical curve coincides with, or matches, the experimental X-ray diffraction pattern curve of the sample to be tested, thereby obtaining the desired experimental results, such as specific atomic positions, bond length and bond angle information, and more accurate lattice constants. Crystal structures obtained through refinement that are not included in the database will be stored in the crystal structure database, enabling the upgrade of the crystal structure database.

[0047] It can be understood that the crystal structure database of the present application is an open database, which can be continuously updated according to actual detection conditions, and more newly developed new materials can be added thereto to meet the needs of new material analysis.

[0048] In one implementation of the present application, the X-ray diffraction pattern theoretical spectrum information is obtained from the known crystal structure according to the XRD diffraction intensity integral formula of the powder sample. The powder diffraction intensity integral formula is shown in Formula 1:

[0049] Formula 1

[0050]

[0051] Where I0 is the incident beam intensity, λ is the wavelength, e is the charge of the electron, and m e is the mass of the electron, r is the distance from the scattered electron to the detector, c is the speed of light, M hkl and F hkl are the multiplicity and structure factor of the hkl reflection, V α is the unit cell volume of the α phase, θ and θ m are the diffraction angles of the hkl reflection and the monochromator, B is the average atomic displacement parameter (ADP), and W α and ρ α are the weight fraction and density of the α phase, μ* m is the mass absorption coefficient of the entire sample.

[0052] In one implementation of the present application, the percentage of each phase in the sample to be tested is quantitatively calculated. Specifically, the data of each diffraction peak in the phase to be tested is used for quantitative analysis. That is, the RIR value of each diffraction peak of the phase to be tested is calculated based on the diffraction intensity at the corresponding position in the corundum diffraction pattern. The mass fraction of the phase to be tested is obtained by combining the intensity ratio of different diffraction peaks in the diffraction pattern of the entire phase to be tested. The details are as follows:

[0053] For phase x in a mixture of X phases, its mass fraction is

[0054]

[0055] in

[0056]

[0057]

[0058] M is the number of diffraction peaks of phase x; m is the mth diffraction peak in descending order of intensity in the diffraction pattern of phase x; I m is the diffraction intensity of the mth diffraction peak; It is the mth diffraction peak in the standard corundum diffraction pattern arranged in descending order of intensity.

[0059] In one implementation of the present application, the material analysis method of the present application also includes performing spectrum preprocessing on the experimental spectrum obtained from the test sample before comparing it with the theoretical spectrum information of the crystal structure database. The spectrum preprocessing includes eliminating stray peaks, subtracting background, and smoothing curves on the experimental spectrum information of the test sample.

[0060] It should be noted that the purpose of spectral preprocessing is to remove the noise or background signals in the experimental spectral information of the sample to be tested; or to facilitate subsequent comparison through curve smoothing, thereby improving the speed and accuracy of comparative analysis.

[0061] In one implementation of the present application, the material analysis method of the present application also includes a peak search step before comparing with the theoretical spectrum information of the crystal structure database, and the peak search step includes finding the characteristic peak information in the experimental spectrum information of the sample to be tested; when performing the comparison, the peak intensity recognition threshold and the peak spacing recognition threshold are set according to the characteristic peak information of the experimental spectrum of the sample to be tested, so as to determine the comparison range, thereby improving the spectrum peak search accuracy.

[0062] It should be noted that the purpose of the peak-finding step is to roughly determine the comparison range through pre-set parameters, thereby improving the efficiency and accuracy of the comparison. The peak position and peak intensity data entered during the search are determined by the peak-finding results, so the peak-finding precision setting will affect the final search accuracy: selecting too dense, small peaks or too sparse peaks will dilute or miss key sample information, resulting in inaccurate search results.

[0063] Another aspect of the present application discloses a material analysis system based on a crystal structure database, including a crystal structure database, a theoretical spectrum information calculation module, a data preprocessing module, a comparison module and a result output module; the crystal structure database has material structure data obtained by experimental measurement and / or theoretical prediction, and the material structure data includes a chemical formula, a space group, a unit cell parameter, and the specific coordinates of atoms in the unit cell; the theoretical spectrum information calculation module includes a module for calculating theoretical spectrum information based on the structural data of each material in the crystal structure database; the data preprocessing module includes a module for obtaining experimental spectrum information obtained by detecting the sample to be tested and denoising the experimental spectrum information; the comparison module includes a module for comparing the results of the data preprocessing module with the theoretical spectrum information calculated by the theoretical spectrum information calculation module; the result output module includes a module for outputting the crystallographic information and related phase composition information of the sample to be tested through intelligent analysis based on the results of the comparison module, wherein the crystallographic information includes at least one of the space group, unit cell parameters, and the specific coordinates of atoms in the unit cell.

[0064] It should be noted that, in the material analysis system based on the crystal structure database of the present application, the theoretical spectrum information calculation module is mainly used to calculate the theoretical spectrum information based on the structural data of each material in the crystal structure database, wherein the specific type of theoretical spectrum to be calculated can be determined according to the experimental spectrum obtained by the test sample detection, and is not specifically limited here. It can be understood that various types of theoretical spectrum information can also be pre-calculated and stored in the database, and can be directly called when used; this can save the theoretical spectrum information calculation steps and improve the system efficiency; however, this requires a large amount of storage space. Therefore, it is possible to choose to design a theoretical spectrum information calculation module or pre-store the calculated theoretical spectrum information according to the needs. In the material analysis system of the present application, the data pre-processing module performs denoising on the experimental spectrum information, which is actually the spectrum pre-processing of the experimental spectrum information obtained by the test sample detection, specifically including eliminating the miscellaneous peaks, subtracting the background and smoothing the curve of the experimental spectrum information of the test sample to facilitate subsequent comparison, thereby improving the speed and accuracy of the comparative analysis. Furthermore, the data preprocessing module may also include, for example, identifying characteristic peak information in the experimental spectrum information; so that when performing comparison, the peak intensity recognition threshold and the peak spacing recognition threshold are set according to the characteristic peak information of the experimental spectrum of the sample to be tested, so as to determine the comparison range, thereby improving the spectrum peak search accuracy.

[0065] It should also be noted that the crystal structure database-based material analysis system of this application actually implements the various steps of the crystal structure database-based material analysis method of this application through various modules to achieve automated material analysis. Therefore, the functions and effects of each module in the material analysis system of this application can refer to the material analysis method of this application. For example, the intelligent analysis of the result output module can refer to the material analysis method of the present application. When the experimental spectrum information matches the theoretical spectrum information, the crystallographic information and related phase composition information of the sample to be tested are directly obtained based on the matching results; when there is no match, the isomorphic structure is retrieved. If there is an isomorphic structure, the theoretical spectrum information of the isomorphic structure is refined to obtain the crystallographic information and related phase composition information of the sample to be tested; if there is no isomorphic structure, the traditional crystallographic method and machine learning method are combined to use positive space search, simulated annealing and genetic algorithm to collaboratively build a structural model for the sample to be tested, and the theoretical spectrum information of the sample to be tested is calculated based on the constructed structural model. The theoretical spectrum information of the sample to be tested is refined so that the experimental spectrum information obtained by the test of the sample to be tested matches the refined theoretical spectrum information of the sample to be tested, and the crystallographic information and related phase composition information of the sample to be tested are obtained based on the matching results. For example, the material analysis method of this application can be referred to for methods of searching for isomorphic structures, iterative updates of crystal structure databases, types of spectral information, calculation methods for theoretical spectral information, quantitative calculation of the percentage content of each phase in a sample to be tested, methods for obtaining material structure data from experimental measurements and theoretical predictions in crystal structure databases, etc. The types of spectral information include, for example, time / space spectral information and energy spectral information.

[0066] In one implementation of the present application, the data preprocessing module also includes a module for identifying the contents of files of different formats of the experimental spectrum information of the sample to be tested, and reading the relevant information in the file based on the file identification result for subsequent comparison.

[0067] In one implementation of the present application, the material analysis system of the present application also includes a crystal structure database retrieval module, which includes a function for extracting and presenting the entire crystal structure database. This module provides users with a method for directly accessing database information. All data of the corresponding structure in the crystal structure database, including the chemical formula, space group, unit cell parameters, and the specific coordinates of atoms in the unit cell, can be found by any of the ICSD numbers, chemical formulas, and constituent elements. In the present application, unit cell parameters mainly include lattice vectors and unit cell volumes.

[0068] In one implementation of the present application, the material analysis system further includes a spectrum calculation module. This module includes spectrum calculation for a given structure and broadening calculation for spectra containing only peak position and peak intensity data. This module provides intuitive peak shape visualization, allowing users to more easily compare with their own experimental data.

[0069] Preferably, the broadening calculation includes four broadening methods for the intensity values: Gaussian, Lorentzian, Voigt, and convolution, with the overlapping portions accumulated. Different experimental diffraction conditions and sample conditions will produce different peak shapes. This step allows users to fine-tune the standard peak shape based on their own experimental spectrum to obtain the optimal comparison mode.

[0070] In one implementation of the present application, the material analysis system further includes a single peak search module. The single peak search module is configured to search and present a crystal structure database based on mouse selection, manual peak search, or input of specific peak position, peak intensity, and precision data. In this step, a user can search for a specific peak and obtain all structural information containing that peak in the database.

[0071] It can be understood that the material parsing method based on the crystal structure database of the present application, all or part of its functions can be implemented by hardware or by computer program. When implemented by a computer program, the program can be stored in a computer-readable storage medium, and the storage medium may include: read-only memory, random access memory, disk, CD, hard disk, etc., and the program is executed by a computer to implement the material parsing method of the present application. For example, the program is stored in the memory of the device, and when the program in the memory is executed by the processor, the method of the present application can be implemented. When all or part of the functions in the method of the present application are implemented by a computer program, the program can also be stored in a storage medium such as a server, another computer, disk, CD, flash drive or mobile hard disk, and saved in the memory of the local device by downloading or copying, or the system of the local device is updated, and then when the processor executes the program in the memory, all or part of the functions of the material parsing method based on the crystal structure database of the present application can be implemented.

[0072] Therefore, another aspect of the present application also discloses a material analysis system based on a crystal structure database, which includes a memory and a processor; wherein the memory is used to store programs; and the processor is used to implement the material analysis method based on the crystal structure database of the present application by executing the programs stored in the memory.

[0073] Another aspect of the present application also discloses a computer-readable storage medium, including a program stored therein, which can be executed by a processor to implement the material analysis method based on the crystal structure database of the present application.

[0074] Another aspect of the present application discloses the application of the material analysis method of the present application, or the material analysis system of the present application, or the computer-readable storage medium of the present application in the research and development of new materials, analysis of the structure of unknown materials, or analysis of the composition of unknown materials.

[0075] It can be understood that the material analysis method and system of the present application can analyze unknown materials, obtain their crystallographic information and related phase composition information, and even perform quantitative analysis of the phase composition of unknown materials; therefore, it can be used for new material research and development, unknown material structure analysis, or unknown material composition analysis.

[0076] The beneficial effects of this application are:

[0077] The material analysis method of this application calculates theoretical map information based on material structure data such as specific atomic positions and unit cell configurations provided by structural data in a crystal structure database. By comparing the experimental map information of the test sample with the theoretical map information and performing intelligent analysis, the crystallographic information and related phase composition information of the test sample are obtained. This realizes material analysis driven by material gene big data and represents a new paradigm for material analysis methods. By comparing the theoretical map information obtained by calculation, the material analysis method of this application can achieve large-scale, high-throughput, and high-precision material analysis, laying the foundation for building a materials science map data mining and analysis platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 This is the interface diagram after importing the file in the embodiment of the present application;

[0079] Figure 2 This is an element search interface diagram in the embodiment of the present application;

[0080] Figure 3 This is the interface presentation diagram after the search is completed in the embodiment of the present application;

[0081] Figure 4 It is a report diagram in the examples of this application;

[0082] Figure 5 This is a comparison chart of the most recently opened projects in the embodiment of this application;

[0083] Figure 6 This is a diagram of a database search pattern in an embodiment of the present application;

[0084] Figure 7 This is a diagram of searching all information in the database in the embodiment of this application;

[0085] Figure 8 This is a module diagram for calculating diffraction patterns in an embodiment of the present application;

[0086] Figure 9It is a single peak search pattern diagram in the embodiment of this application. DETAILED DESCRIPTION

[0087] Existing material analysis methods generally rely on comparative analysis of new materials based on actual test results from standard samples or other known materials. This data analysis paradigm is largely monopolized by foreign countries. Furthermore, it suffers from limitations such as experimental errors, poor data structure and distribution regularity, slow retrieval speed, and poor accuracy. Furthermore, high user fees, cumbersome and complex search procedures, rigid functionality, and unsightly interfaces have long been criticized by the industry.

[0088] Professor Pan Feng, one of the inventors of this application, and his team developed a new paradigm for structural chemistry and materials genetics research based on graph theory. They defined the nearest neighbor atoms bonded to a central atom as structural primitives, then abstracted atoms or structural primitives into points in graph theory. Combined with computer intelligent computing, they achieved a technological breakthrough in accurately determining structural isomorphism and quantitatively describing the heterogeneity and evolutionary relationships between structures. Based on graph theory's crystal structure and material genetic expression methods, Professor Pan Feng and his team geometrically topologically topologically represented spatial atomic / molecular structures, greatly improving the efficiency and accuracy of intelligent material identification and constructing a high-quality crystal structure database with over 600,000 structures. The graph theory and material crystal structure description methods are referenced in patent application 201910390012.2. The crystal structure descriptions of material structure data in this application's crystal structure database all refer to that patent application, and all relevant content in that patent application is incorporated into this application.

[0089] Based on the above research, this application creatively proposes a new material analysis method and system, which uses graph theory methods to achieve high-quality classification and cleaning of the database, reversely calculates to obtain standard theoretical spectrum information, and combines computer high-throughput intelligent comparison and intelligent analysis to achieve intelligent detection and analysis of the materials to be tested. The first generation of the system has begun to be put into use and provide services to a large number of scientific researchers.

[0090] Specifically, the material analysis method of the present application is based on a crystal structure database, including comparing the experimental spectrum information obtained from the detection of the sample to be tested with the theoretical spectrum information calculated from the structural data of each material in the crystal structure database, and obtaining the crystallographic information and related phase composition information of the sample to be tested through intelligent analysis; the crystallographic information includes at least one of the space group, unit cell parameters, and specific coordinates of atoms in the unit cell; the crystal structure database has material structure data obtained by experimental measurement and / or theoretical prediction, and the material structure data includes chemical formula, space group, unit cell parameters, and specific coordinates of atoms in the unit cell.

[0091] Among them, intelligent analysis can be divided into three situations:

[0092] When the experimental spectra information of the sample to be tested matches the theoretical spectra information, the crystallographic information and related phase composition information of the sample to be tested are directly obtained based on the matching results. It can be understood that if the experimental spectra information matches the theoretical spectra information, it means that the sample to be tested is the same material in the crystal structure database corresponding to the theoretical spectra information. Therefore, the crystallographic information and related phase composition information of the sample to be tested can be directly output, including the chemical formula, space group, unit cell parameters, and the specific coordinates of atoms in the unit cell.

[0093] If the experimental spectrum information of the sample to be tested does not match the theoretical spectrum information, but the theoretical spectrum information of the isomorphic structure of the experimental spectrum information can be retrieved, the theoretical spectrum information of the isomorphic structure is refined so that the refined theoretical spectrum information matches the experimental spectrum information, and the crystallographic information and related phase composition information of the sample to be tested are obtained based on the matching results. In this case, the isomorphic method is actually used to analyze the sample to be tested; how to judge isomorphism and how to resolve isomorphism can refer to the existing technology; however, in one implementation of the present application, specifically based on the peak intensity ratio relationship contained in the experimental spectrum information, the theoretical spectrum information of the isomorphic structure is retrieved in the crystal structure database; and by refining the theoretical spectrum information of the isomorphic structure, the refined theoretical spectrum information matches the experimental spectrum information.

[0094] The experimental spectrum information of the test sample does not match the theoretical spectrum information, and the theoretical spectrum information of the isomorphic structure of the experimental spectrum information cannot be retrieved, that is, there is no isomorphism. This application combines traditional crystallographic methods and machine learning methods, and uses positive space search, simulated annealing and genetic algorithms to collaboratively build a structural model for the test sample. The theoretical spectrum information of the test sample is calculated based on the built structural model, and the theoretical spectrum information of the test sample is refined so that the experimental spectrum information obtained by the test sample is matched with the refined theoretical spectrum information of the test sample, and the crystallographic information and related phase composition information of the test sample are obtained based on the matching results.

[0095] In a further improvement scheme of the present application, in the case where the experimental spectrum information does not match the theoretical spectrum information, the obtained crystallographic information of the sample to be tested and the related phase composition information can be supplemented into the crystal structure database to achieve iterative update of the crystal structure database.

[0096] In this application, there are two ways to obtain material structure data from the crystal structure database. One is the material structure data obtained by experimental measurement, that is, the material structure data obtained by experimental characterization of samples with known structures; the other is the material structure data obtained by theoretical prediction.

[0097] The material structure data obtained by theoretical prediction is calculated by at least one of the following methods and obtained by crystal structure deduplication based on graph theory methods:

[0098] Method 1) The material structure data obtained by experimental characterization of samples with known structures are replaced with elements and then the structure is relaxed using the DFT method to obtain new materials and their material structure data;

[0099] Method 2) Performing structural search based on PSO and / or genetic algorithm on material structure data obtained by experimental characterization of samples with known structures to obtain new materials and their material structure data;

[0100] Method 3) is to extract features from the material structure data obtained by experimental characterization of samples with known structures, and reverse design is performed based on the extracted features to obtain new materials and their material structure data.

[0101] In this application, the spectrum information, namely experimental spectrum information and theoretical spectrum information, includes time / space spectrum information and energy spectrum information; time / space spectrum information such as X-ray diffraction spectrum, neutron diffraction spectrum, electron diffraction spectrum; energy spectrum information includes emission spectrum, absorption spectrum and scattering spectrum; emission spectrum such as atomic emission spectrum and / or infrared emission spectrum; absorption spectrum such as infrared absorption spectrum and / or ultraviolet absorption spectrum; scattering spectrum such as Raman spectrum.

[0102] The method of calculating theoretical spectrum information from material structure data in this application includes:

[0103] (1) Calculate time / space spectrum information based on the powder diffraction intensity integral formula;

[0104] (2) Calculating energy spectrum information by simulating crystal structure based on quantum chemical methods. Specifically, obtaining spectral information of emission spectrum, absorption spectrum and scattering spectrum by simulating atomic nuclear vibration of crystal structure based on quantum chemical methods. Obtaining energy spectrum information of emission spectrum, absorption spectrum and scattering spectrum by simulating electronic structure corresponding to crystal structure based on quantum chemical methods.

[0105] It is understood that in order to improve the accuracy of the analysis of the sample material to be tested, the present application can further use at least two experimental spectrum information of the sample to be tested for comparison to obtain the crystallographic information and related phase composition information of the sample to be tested.

[0106] In addition, the material analysis method of the present application can also quantitatively calculate the content percentage of each phase in the sample to be tested based on the phase composition information of the sample to be tested.

[0107] Based on the material analysis method of the present application, the present application further proposes a material analysis system based on a crystal structure database, including a crystal structure database, a theoretical spectrum information calculation module, a data preprocessing module, a comparison module and a result output module; the crystal structure database has material structure data obtained by experimental measurement and / or theoretical prediction, and the material structure data includes a chemical formula, a space group, a unit cell parameter, and the specific coordinates of atoms in the unit cell; the theoretical spectrum information calculation module includes a module for calculating theoretical spectrum information based on the structural data of each material in the crystal structure database; the data preprocessing module includes a module for obtaining experimental spectrum information obtained by detecting the sample to be tested and denoising the experimental spectrum information; the comparison module includes a module for comparing the results of the data preprocessing module with the theoretical spectrum information calculated by the theoretical spectrum information calculation module; the result output module includes a module for outputting the crystallographic information and related phase composition information of the sample to be tested through intelligent analysis based on the results of the comparison module, wherein the crystallographic information includes at least one of the space group, unit cell parameters, and the specific coordinates of atoms in the unit cell.

[0108] The present application further proposes a material analysis system based on a crystal structure database, comprising a memory and a processor; the memory is used to store programs; the processor is used to implement the following method by executing the programs stored in the memory: comprising comparing the experimental spectrum information obtained from the detection of the sample to be tested with the theoretical spectrum information calculated from the structural data of each material in the crystal structure database, and obtaining the crystallographic information and related phase composition information of the sample to be tested through intelligent analysis; wherein the crystallographic information includes at least one of the space group, unit cell parameters, and specific coordinates of atoms in the unit cell; the crystal structure database has material structure data obtained by experimental measurement and / or theoretical prediction, and the material structure data includes chemical formula, space group, unit cell parameters, and specific coordinates of atoms in the unit cell.

[0109] The present application further proposes a computer-readable storage medium, including a program stored therein, which can be executed by a processor to implement the following method: including comparing the experimental spectrum information obtained from the detection of the sample to be tested with the theoretical spectrum information calculated from the structural data of each material in the crystal structure database, and obtaining the crystallographic information and related phase composition information of the sample to be tested through intelligent analysis; wherein the crystallographic information includes at least one of the space group, unit cell parameters, and specific coordinates of atoms in the unit cell; the crystal structure database has material structure data obtained by experimental measurement and / or theoretical prediction, and the material structure data includes chemical formula, space group, unit cell parameters, and specific coordinates of atoms in the unit cell.

[0110] The material analysis method and system of this application address the cumbersome use and high cost of existing material analysis systems, breaking the existing data paradigm and technological monopoly, creating a new era for material analysis and addressing deficiencies in existing data composition, analysis methods, and software technology. Furthermore, the theoretical atlas information used in this material analysis method is calculated from material structure data in a crystal structure database, avoiding the experimental errors and poor regularity of data structure distribution that are common in existing crystallographic databases, which rely on experimentally obtained data.

[0111] Furthermore, the material analysis system of this application is neat, beautiful, and easy to use, representing a new "Unnamed Material Analysis System" with complete independent intellectual property rights. Even the various interfaces and presentation methods of this application's material analysis system, including icons, login screen, file import screen, element search screen, post-search screen, report charts, recently opened project comparison, and database search mode, have been meticulously designed, making the search process simple and easy to use, with a neat and beautiful interface, fast search speed, and more user-friendly functions.

[0112] In short, the material analysis system of this application comprehensively considers various factors such as the current status of research and development at home and abroad, the positioning and strength of schools and colleges, and the needs of scientific researchers. It is a structural analysis system that is particularly suitable for use by domestic scientific researchers.

[0113] The present invention is further described in detail below through specific examples and drawings. The following examples are only used to further illustrate the present invention and should not be construed as limiting the present invention.

[0114] Example

[0115] The material analysis method based on the crystal structure database in this example includes comparing the experimental spectrum information obtained from the detection of the sample to be tested with the theoretical spectrum information calculated from the structural data of each material in the crystal structure database, and obtaining the crystallographic information and related phase composition information of the sample to be tested through intelligent analysis; wherein the crystallographic information includes at least one of the space group, unit cell parameters, and specific coordinates of atoms in the unit cell; the crystal structure database has material structure data obtained by experimental measurement and / or theoretical prediction, and the material structure data includes chemical formula, space group, unit cell parameters, and specific coordinates of atoms in the unit cell.

[0116] In this example, intelligent analysis includes:

[0117] (1) When the experimental spectrum information of the sample to be tested matches the theoretical spectrum information in the crystal structure database, the crystallographic information and related phase composition information of the sample to be tested are directly obtained based on the matching results;

[0118] (2) If the experimental spectrum information of the sample to be tested does not match the theoretical spectrum information of the crystal structure database,

[0119] 1) searching a crystal structure database for theoretical spectrum information of a structure isomorphic with the experimental spectrum information, refining the theoretical spectrum information of the isomorphic structure so that the refined theoretical spectrum information matches the experimental spectrum information, and obtaining crystallographic information and related phase composition information of the sample to be tested based on the matching results; in this application, refining the spectrum information or refining the spectrum refers to adjusting various parameters corresponding to the original spectrum, such as replacing known elements in the original spectrum to change the peak size, shape, position, etc. of the spectrum curve;

[0120] 2) In the case where no theoretical spectrum information of the isomorphic structure is retrieved, traditional crystallographic methods and machine learning methods are combined to collaboratively build a structural model of the sample to be tested using positive space search, simulated annealing, and genetic algorithm. The theoretical spectrum information of the sample to be tested is calculated based on the built structural model, and the theoretical spectrum information of the sample to be tested is refined so that the experimental spectrum information obtained from the detection of the sample to be tested matches the refined theoretical spectrum information of the sample to be tested. Based on the matching results, the crystallographic information of the sample to be tested and related phase composition information are obtained.

[0121] The experimental spectrum information and theoretical spectrum information can be at least one of time / space spectrum information and energy spectrum information. Time / space spectrum information includes, for example, X-ray diffraction patterns, neutron diffraction patterns, and electron diffraction patterns. Energy spectrum information includes emission spectra, absorption spectra, and scattering spectra. Emission spectra include, for example, atomic emission spectra and / or infrared emission spectra, absorption spectra include, for example, infrared absorption spectra and / or ultraviolet absorption spectra, and scattering spectra include, for example, Raman spectra.

[0122] The calculation method of the theoretical spectrum information in this example includes:

[0123] (1) Calculate time / space spectrum information based on the powder diffraction intensity integral formula;

[0124] (2) Calculating energy spectrum information by simulating crystal structure based on quantum chemical methods. Specifically, obtaining spectral information of emission spectrum, absorption spectrum and scattering spectrum by simulating atomic nuclear vibration of crystal structure based on quantum chemical methods. Obtaining energy spectrum information of emission spectrum, absorption spectrum and scattering spectrum by simulating electronic structure corresponding to crystal structure based on quantum chemical methods.

[0125] In the crystal structure database of this example, the material structure data obtained by experimental measurement refers to the material structure data obtained by experimental characterization of samples with known structures; the material structure data obtained by theoretical prediction is calculated by at least one of the following calculation methods and obtained through crystal structure deduplication based on graph theory methods.

[0126] The calculation methods for theoretically predicted material structure data include:

[0127] Method 1) The material structure data obtained by experimental characterization of samples with known structures are replaced with elements and then the structure is relaxed using the DFT method to obtain new materials and their material structure data;

[0128] Method 2) Performing structural search based on PSO and / or genetic algorithm on material structure data obtained by experimental characterization of samples with known structures to obtain new materials and their material structure data;

[0129] Method 3) is to extract features from the material structure data obtained by experimental characterization of samples with known structures, and reverse design is performed based on the extracted features to obtain new materials and their material structure data.

[0130] In an improved approach to this example, the experimental spectrum obtained from the test sample is pre-processed before comparison with the theoretical spectrum information in the crystal structure database. The spectrum pre-processing includes removing impurities, subtracting background, and smoothing the curve of the experimental spectrum information of the test sample. Furthermore, before comparison with the theoretical spectrum information in the crystal structure database, a peak search step is performed. The peak search step includes finding characteristic peak information in the experimental spectrum information of the test sample. During the comparison, the peak intensity recognition threshold and the peak spacing recognition threshold are set based on the characteristic peak information of the experimental spectrum of the test sample to determine the comparison range, thereby improving the accuracy of spectrum peak search.

[0131] The material analysis method in this example can use one experimental spectrum obtained from the test sample for comparative analysis, or can use two or more experimental spectrums for comparative analysis to obtain more accurate crystallographic information and related phase composition information of the test sample.

[0132] Furthermore, the material analysis method of this example can also quantitatively calculate the content percentage of each phase in the sample to be tested based on the phase composition information of the sample to be tested.

[0133] Taking the X-ray diffraction pattern (XRD diffraction pattern) as an example, the degree of match between the diffraction peak position and peak intensity corresponding to the experimental X-ray diffraction pattern information of the sample to be tested and the theoretical X-ray diffraction pattern information calculated based on the crystal structure database can be used to determine the degree of match between the experimental X-ray diffraction pattern information of the sample to be tested and the corresponding structure in the theoretical X-ray diffraction pattern of the crystal structure database. Specifically, the structural matching degree value of the experimental X-ray diffraction pattern of the sample to be tested and the theoretical X-ray diffraction pattern is calculated by the quality factor matching formula. The larger the matching degree value, the higher the matching degree, and vice versa. When the matching degree value is higher than a certain value, for example, greater than or equal to 0.01, it is considered that the structure in the corresponding crystal structure database belongs to a phase in the sample to be tested. When the matching degree value is lower than a certain degree, for example, lower than 0.01, it is considered that the sample to be tested has not been matched in the crystal structure database. Further isomorphism judgment is required. If there is an isomorphism match, analysis is performed according to the isomorphism match. If there is no isomorphism match, further analysis is required by combining traditional crystallographic methods, machine learning methods and existing structures in the crystal database.

[0134] Specifically, the material analysis method based on X-ray diffraction patterns includes the following steps:

[0135] The experimental X-ray diffraction pattern information obtained from the test sample is compared with the theoretical X-ray diffraction pattern information calculated from the structural data of each material in the crystal structure database. If they match, the crystallographic information and related phase composition information of the test sample are directly obtained based on the matching results; if they do not match, the following operations are performed:

[0136] 1) Diffraction pattern indexing

[0137] The indexing method of the X-ray diffraction pattern of polycrystalline powder is to find the solution that satisfies the following equation within the range of experimental error, solve the lattice parameters including a, b, c and alpha, beta, gamma from the interplanar spacing corresponding to the diffraction peak, and determine the crystal plane index of the diffraction peak at the same time.

[0138] 2) Space group determination

[0139] Based on the systematic extinction caused by the presence of centered lattices, helical axes, and slip planes in the crystal, the lattice type and X-ray diffraction group of the crystal are determined after indexing; the space group is determined in combination with the statistical laws of diffraction intensity.

[0140] 3) Structural model construction

[0141] Relying on the support of a large amount of structural data in the crystal structure database, a search for isomorphic structures is first performed. Since a large number of crystal structures have been studied and entered into the database, the crystal structure of a new compound should first be checked to see if it is isomorphic with a compound of a known structure. Once the chemical formula, lattice parameters and space group of the new compound are determined, substances with similar chemical formulas, lattice parameters and the same space group can be searched in the structure database based on the number of atoms in the unit cell and the chemical properties of the elements, and the diffraction patterns of the two can be compared. The specific comparison method is to search for isomorphic structures in the crystal structure database based on the peak intensity information contained in the spectrum information and the crystal plane index information of the corresponding peak intensity obtained by indexing. The crystal structure that is isomorphic with the corresponding substance of the experimental sample should have a similar proportional relationship between the peak intensities corresponding to different crystal plane indices. If a compound with a similar diffraction pattern and a known crystal structure can be found, that is, theoretical spectrum information of an isomorphic structure, the distribution and rough position of atoms in the new compound can be determined. Rietveld refinement can then be used to match the refined theoretical spectrum information with the experimental spectrum information. Based on the matching results, the crystallographic information of the sample to be tested and related phase composition information can be obtained.

[0142] For the test samples for which no isomorphic structure has been retrieved in the crystal structure database, traditional crystallographic methods and machine learning methods are combined to collaboratively build a structural model of the test samples using positive space search, simulated annealing, genetic algorithms and other methods. The computational map information of the test samples is obtained based on the structural model, and the computational map information is refined so that the experimental map information obtained from the test sample detection matches the refined computational map information. Based on the matching results, the crystallographic information and related phase composition information of the test samples are obtained.

[0143] In one implementation of the present application, the pattern information refinement or pattern refinement specifically employs Rietveld refinement, i.e., Rietveld structure refinement. This involves adjusting various parameters through fitting methods such as the least squares method to alter the peak size, shape, and position of the curve of the theoretical XRD diffraction pattern calculated based on the crystal structure database, so that the theoretical curve coincides with, or matches, the experimental X-ray diffraction pattern curve of the sample to be tested, thereby obtaining the desired experimental results, such as specific atomic positions, bond length and bond angle information, and more accurate lattice constants. Crystal structures obtained through refinement that are not included in the database will be stored in the crystal structure database, enabling the upgrade of the crystal structure database.

[0144] This example uses the integrated intensity calculation method to calculate the theoretical X-ray diffraction pattern information of material structure data, specifically including obtaining the material diffraction pattern database based on the known crystal structure according to the powder sample XRD diffraction intensity integral intensity formula. The powder diffraction intensity integral formula is shown in Formula 1.

[0145] Formula 1

[0146]

[0147] Where I0 is the incident beam intensity, λ is the wavelength, e is the charge of the electron, and m e is the mass of the electron, r is the distance from the scattered electron to the detector, c is the speed of light, M hkl and F hkl are the multiplicity and structure factor of the hkl reflection, V α is the unit cell volume of the α phase, θ and θ m are the diffraction angles of the hkl reflection and the monochromator, B is the average atomic displacement parameter (ADP), and W α and ρ α are the weight fraction and density of the α phase, μ* m is the mass absorption coefficient of the entire sample.

[0148] Furthermore, this example can also quantitatively calculate the percentage of each phase in the sample to be tested. Specifically, the data of each diffraction peak in the phase to be tested is used for quantitative analysis. That is, the RIR value of each diffraction peak of the phase to be tested is calculated based on the diffraction intensity at the corresponding position in the corundum diffraction pattern, and the mass fraction of the phase to be tested is obtained by combining the intensity ratio of different diffraction peaks in the entire diffraction pattern of the phase to be tested. The details are as follows:

[0149] For phase x in a mixture of X phases, its mass fraction is

[0150]

[0151] in

[0152]

[0153]

[0154] M is the number of diffraction peaks of phase x; m is the mth diffraction peak in descending order of intensity in the diffraction pattern of phase x; I m is the diffraction intensity of the mth diffraction peak; It is the mth diffraction peak in the standard corundum diffraction pattern arranged in descending order of intensity.

[0155] Based on the above material analysis method, this example further developed a material analysis system based on the crystal structure database, including a crystal structure database, a theoretical spectrum information calculation module, a data preprocessing module, a comparison module and a result output module.

[0156] Among them, the crystal structure database has material structure data obtained by experimental measurement and / or theoretical prediction, and the material structure data includes chemical formula, space group, unit cell parameters, and specific coordinates of atoms in the unit cell; the theoretical spectrum information calculation module includes a module for calculating theoretical spectrum information based on the structural data of each material in the crystal structure database; the data preprocessing module includes a module for obtaining experimental spectrum information obtained by detecting the sample to be tested and denoising the experimental spectrum information; the comparison module includes a module for comparing the results of the data preprocessing module with the theoretical spectrum information calculated by the theoretical spectrum information calculation module; the result output module includes a module for outputting the crystallographic information and related phase composition information of the sample to be tested through intelligent analysis based on the results of the comparison module, wherein the crystallographic information includes at least one of the space group, unit cell parameters, and specific coordinates of atoms in the unit cell.

[0157] The material analysis system of this example actually implements the material analysis method of this example through various modules to achieve automated material analysis. Therefore, the functions and roles of each module in the material analysis system of this example can refer to the material analysis method of this example. For example, intelligent analysis, retrieval methods for isomorphic structures, iterative updates of crystal structure databases, types of atlas information, calculation methods for theoretical atlas information, quantitative calculation of the percentage content of each phase in the sample to be tested, methods for obtaining material structure data through experimental measurements and theoretical predictions in the crystal structure database, etc., can all refer to the material analysis method of this example.

[0158] Therefore, the data preprocessing module further includes a process for removing clutter peaks, subtracting background, and smoothing the curve of the acquired analytical data of the sample to be tested. Furthermore, the data preprocessing module also includes a process for finding characteristic peak information in the experimental spectrum information of the sample to be tested; when performing a comparison, the peak intensity recognition threshold and the peak spacing recognition threshold are set based on the characteristic peak information of the experimental spectrum of the sample to be tested, thereby determining the comparison range and improving the accuracy of spectrum peak finding.

[0159] Furthermore, the data preprocessing module also includes a module for identifying the contents of files of different formats of the experimental atlas information of the sample to be tested, and reading the relevant information in the file according to the file identification result for subsequent comparison.

[0160] Furthermore, the material analysis system of this example also includes a crystal structure database retrieval module, which includes a method for extracting and presenting the entire crystal structure database. This module provides users with a method for directly accessing database information. All data of the corresponding structure in the crystal structure database, including chemical formula, space group, unit cell parameters, and specific coordinates of atoms in the unit cell, can be found by any of the ICSD numbers, chemical formulas, and constituent elements. In this application, unit cell parameters mainly include lattice vectors and unit cell volumes.

[0161] Furthermore, the material analysis system of this example also includes a spectrum calculation module, which includes spectrum calculation for a given structure and broadening calculation for a spectrum containing only peak position and peak intensity data. This module can provide an intuitive peak shape visual effect, allowing users to compare with their own experimental data more conveniently. Among them, the broadening calculation includes four types of broadening of the intensity value: Gaussian, Lorentz, Voigt, and convolution, and the overlapping parts are accumulated. Different experimental diffraction conditions and sample conditions may produce different peak shapes. This step allows users to fine-tune the standard peak shape according to their own experimental spectrum to obtain the best comparison mode.

[0162] Furthermore, the material analysis system in this example also includes a single peak search module, which allows users to search and present the crystal structure database using mouse selection, manual peak search, or input of specific peak position, peak intensity, and precision data. In this step, users can search for a specific peak and obtain all structural information containing that peak in the database.

[0163] The above material analysis system is based on the material diffraction spectrum database. In this case, the system name is: Unknown Material Analysis System, and the development language is: Python. The login interface of this system requires matching of username and password to log in; the main interface is built with the help of PyQt5, the logic and interface have been separated, and corresponding connections have been established; the import module adopts the method of identifying the file name + suffix related string, and uses regular expressions to filter the reading of text content to select specific information, such as the peak position and peak intensity values ​​contained in various support files; the scanning speed function is actually a point selection method, and the 2 / 3 / 4 times scanning speed selects one for every 2 / 3 / 4 data points of peak position and peak intensity respectively, thereby realizing the change of the spectrum scanning speed magnification; the background recognition function is to compare each data point in the spectrum with the adjacent Compare the data points, select the relatively smaller ones, connect the starting point and the end point, refer to the smaller point for data fitting, and make a baseline graph; the Kα2 subtraction function is similar to background identification. Each diffraction peak is composed of two characteristic peaks, Kα1 and Kα2. Find the weaker item of each diffraction peak and reduce it to zero or return it to near the baseline; smoothing is actually the use of linear / quadratic / cubic functions to smooth the curve, and the number of fittings refers to the number of times the function is used to process the data; the database interface function uses sqlite to access, modify, and extract the database; the retrieval report is an integration of the above functions and processing results.

[0164] For example, the pseudo code and formula for implementing the retrieval process are as follows:

[0165]

[0166]

[0167] FOM=NR(w θ FOM θ +(1-w θ )FOM I )N=number of matched_CP

[0168]

[0169]

[0170]

[0171] The formula for calculating the diffraction pattern is as follows:

[0172] Lorentzian peak shape function:

[0173] w=H / 2

[0174] Gaussian peak shape function:

[0175] I(2θ)=I max exp[-π(2θ-2θ0) 2 / b 2 ],β=0.5Γ(π / log e 2)1 / w Γ G =2.355σ

[0176] Pearson peak shape function:

[0177]

[0178] The material analysis system in this example is a new scientific system based on database and analysis software. On the one hand, it develops a new paradigm for structural chemistry and material gene research based on graph theory, defining the nearest neighbor atoms bonded to the central atom as structural primitives. Then, atoms or structural primitives are abstracted into points in graph theory. Combined with computer intelligent calculations, this system achieves a technological breakthrough in accurately judging structural isomorphism and quantitatively describing the heterogeneity and evolutionary relationship between structures. Based on the crystal structure and material gene expression method of graph theory, the spatial atomic / molecular structure is geometrically topologically transformed, greatly improving the efficiency and accuracy of intelligent material identification and constructing a high-quality crystal structure database with more than 600,000 structures (SCIENCE CHINAChemistry (2019) 62, 982). On the other hand, the unnamed material analysis system uses microscopic elements such as structural primitives and their connections and chemical bond interactions, combined with high-throughput computer intelligent comparison analysis, to achieve intelligent detection and analysis of materials.

[0179] In the implementation of the material analysis system function in this example, the deconstruction, query and extraction of the crystallographic database can refer to existing software such as jade and qualx.

[0180] Compared with existing software such as Jade and Qualx, on the one hand, these software are all developed by foreign personnel and require high fees to use; on the other hand, these software generally have shortcomings such as complex and cumbersome search procedures, rigid functions, and unsightly interfaces. The material analysis system in this example:

[0181] 1. All developed by the School of Advanced Materials, Peking University;

[0182] 2. Completely free services have been launched to the public;

[0183] 3. The search process is simple and easy to use, and the interface is beautiful and neat;

[0184] 4. Fast retrieval speed and user-friendly functions.

[0185] The material analysis system in this example takes into account the current status of domestic and international research and development, the school's positioning and strength, and the needs of scientific researchers. It is a structural analysis system that is very suitable for domestic scientific researchers to use free of charge.

[0186] Specific application demonstration:

[0187] 1. α-MnO2 Phase Search Example

[0188] This example uses the experimentally synthesized suspected α-MnO2 XRD test results in the example directory of the client program installation package as an example of phase retrieval.

[0189] like Figure 1 As shown, open α-MnO2.csv and enter the main interface. After subtracting the background, subtracting Kα2, and smoothing the curve, enter the following Figure 2The element search interface is shown. Preprocessing options such as background subtraction, Kα2 subtraction, and curve smoothing can be customized. In the "Element Analysis" dialog box, you can click on the periodic table to enter an element, or manually enter it and search. The default delta value is 0.3. You can also click Advanced Settings to enter the dialog box. 1. Number of Phases: Select the desired number of phases; 2. Delta: Influences matching accuracy; 3. W-Theta: The weight of theta's influence on the quality factor; 4. Thread Value: Use multi-threaded data processing. A higher thread count consumes more CPU resources and increases search speed; 5. Search Mode: Selectively search for the selected element in one of three modes: Contains Only, Contains All, and Contains Any. These modes provide a wider search scope, improved search accuracy, and slower search speed. These values ​​are preset by default in the software and can be modified as needed. In this example, the default values ​​are used and the "Contains Any" search range is selected. After setting the parameters, click on the periodic table and type in the two elements "Mn" and "O". Click search to enter the progress bar reading mode. After reading, the results will be displayed as follows: Figure 3 shown. Figure 3 The phase information in the database that matched is displayed below, sorted in descending order according to the quality factor (FOM, fitness), indicating that the matching degree decreases. Clicking on the phase can display the comparison chart mode of the two in the spectrum area for intuitive comparison. In addition, during the progress bar reading and retrieval, the user can use other functions of the software without interfering with each other (dual thread). The result shows K 1.33 Mn8O 16 is the most suitable phase. Click "Export Report" to get the results, such as Figure 4 The “Export Report” module contains the header information such as searcher, search unit, search time, search sample number, and remarks, which can be filled in by the user or the testing unit. Figure 4 Below it are listed all the physical phases of the test results and their ICSD numbers. Click on the component to draw a comparison between the sample diffraction pattern and the standard pattern of the test results in the main area. Press and hold "ctrl" to select multiple images, up to three standard images to compare with the original image. The drawing mode is divided into single image and multi-image. "Single image mode" means putting all the spectra in one coordinate system, and "multi-image mode" means that each spectrum has its own coordinate axis. Users can switch freely according to their preferences and the degree of result presentation.

[0190] 2. “Recently opened projects” display

[0191] like Figure 5As shown, the software records each new project opened in the "Recently Opened Projects" section in the lower left corner, allowing users to open it directly the next time they access the project. If a project already exists on the current screen, the software will ask whether to open a new screen or compare it with the existing project, allowing users to directly observe the project. Additionally, the "Data Overview" section on the left displays the data after each import or processing in a table, allowing users to directly view the processed peak position and intensity data.

[0192] 3. Database Access System Display

[0193] After opening the software, click "Database (D)" in the menu bar or click the data icon in the toolbar to enter the database access system. Taking MnO2 as an example, we will demonstrate in three modes: input chemical formula, input ICSD number, and input element search. Figure 6 As shown. Clicking on the search result will display the standard map in the database on the main interface. If you choose to search directly, all the information in the database will be displayed on the interface at a speed of 100 per second, as shown Figure 7 shown.

[0194] 4. Display of the Calculated Diffraction Pattern Module

[0195] After opening the software, click "Options (O) - Calculate Diffraction Pattern" in the menu bar or click the data icon in the toolbar to enter the diffraction pattern calculation mode. Click "Import" in the dialog box to store the data to be calculated into the system. Currently, files in txt / csv / cif / dat formats are supported. Its peak position and peak intensity information will be displayed in the "Crystallographic Data" column on the left. Then click "Calculate" to expand the peak shape with the default Lorentz peak shape. The default value of the half-height width is 0.3. Figure 8 As shown in the figure. Adjusting different peak shapes such as Lorentz, Gaussian, Pearson, and Voigt peak shapes and different parameters such as half-height width, peak shape correction factor, and convolution factor can produce different diffraction peak shapes. The core algorithm is the four peak shape formulas.

[0196] 5. Single-peak search module display

[0197] Take the suspected Zn3V3O8 sample imported into the software as an example. If there is a peak, usually a small peak, which cannot be found in the standard phase after detection, the single peak search function can be used to search all phases containing this peak in the database to find the second phase that may exist in the sample. Click "Identification (I) - Single Peak Search" in the menu bar or the data icon in the toolbar to enter the single peak search mode. Figure 9As shown. It mainly includes two modes: spectrum peak selection and manual peak search. In "spectrum peak selection", when the mouse moves on the imported spectrum, the coordinates of the mouse position will be displayed on the left. Move to the peak to be selected and click to input the peak position and peak intensity information into the system. Then you can set the accuracy. For example, the deviation of ±1° in position and the deviation of ±200 units in intensity (default value). For example, when 18.33° and 267.29 intensity values ​​are selected, the actual search range is 17.33~19.33° and 257.29~277.29 intensity values. Manual peak search means manually entering the peak position and peak intensity values.

[0198] The above content is a further detailed description of the present application in conjunction with specific implementation methods, and the specific implementation of the present application cannot be considered to be limited to these descriptions. For ordinary technicians in the technical field to which the present application belongs, several simple deductions or substitutions can be made without departing from the concept of the present application.

Claims

1. A material analysis method based on a crystal structure database, characterized by: This includes comparing the experimental spectrum information obtained from the test sample with the theoretical spectrum information calculated from the structural data of each material in the crystal structure database, and obtaining the crystallographic information and related phase composition information of the test sample through intelligent analysis; The crystallographic information includes at least one of a space group, unit cell parameters, and specific coordinates of atoms in the unit cell; The crystal structure database contains material structure data obtained by experimental measurement and / or theoretical prediction, wherein the material structure data includes chemical formula, space group, unit cell parameters, and specific coordinates of atoms in the unit cell; The intelligent analysis includes: (1) When the experimental spectrum information matches the theoretical spectrum information, the crystallographic information and related phase composition information of the sample to be tested are directly obtained based on the matching result; (2) In the case where the experimental map information does not match the theoretical map information, 1) searching a crystal structure database for theoretical map information of a structure isomorphic to the experimental map information, refining the theoretical map information of the isomorphic structure so that the refined theoretical map information matches the experimental map information, and obtaining crystallographic information and related phase composition information of the sample to be tested based on the matching results; 2) In the case where no theoretical spectrum information of the isomorphic structure is retrieved, the traditional crystallographic method and machine learning method are combined to collaboratively build a structural model of the sample to be tested using positive space search, simulated annealing and genetic algorithm. The theoretical spectrum information of the sample to be tested is calculated based on the built structural model, and the theoretical spectrum information of the sample to be tested is refined so that the experimental spectrum information obtained from the test of the sample to be tested matches the refined theoretical spectrum information of the sample to be tested. The crystallographic information and related phase composition information of the sample to be tested are obtained based on the matching results.

2. The material analysis method according to claim 1, wherein: The method for retrieving the theoretical spectrum information of the isomorphic structure includes retrieving the theoretical spectrum information of the isomorphic structure in the crystal structure database according to the peak intensity ratio relationship contained in the experimental spectrum information.

3. The material analysis method according to claim 1, wherein: The intelligent analysis also includes, when the experimental map information does not match the theoretical map information, supplementing the obtained crystallographic information and related phase composition information of the sample to be tested into the crystal structure database to achieve iterative updating of the crystal structure database.

4. The material analysis method according to claim 1, wherein: The experimental spectrum information and the theoretical spectrum information both include at least one of time / space spectrum information and energy spectrum information; The time / space pattern information includes at least one of an X-ray diffraction pattern, a neutron diffraction pattern, and an electron diffraction pattern; The energy spectrum information includes at least one of an emission spectrum, an absorption spectrum, and a scattering spectrum.

5. The material analysis method according to claim 4, characterized in that: The scattering spectrum is a Raman spectrum.

6. The material analysis method according to claim 4, characterized in that: The calculation method of the theoretical map information includes but is not limited to: (1) calculating the time / space pattern information based on the powder diffraction intensity integral formula; (2) Calculating the energy spectrum information based on the quantum chemical method to simulate the crystal structure. Specifically, obtaining the spectral information of the emission spectrum, absorption spectrum, and scattering spectrum by simulating the atomic nuclear vibration of the crystal structure based on the quantum chemical method. Obtaining the energy spectrum information of the emission spectrum, absorption spectrum, and scattering spectrum by simulating the electronic structure corresponding to the crystal structure based on the quantum chemical method.

7. The material analysis method according to claim 4, characterized in that: It also includes the case where it is difficult to perform accurate structural analysis on one experimental map information of the sample to be tested, collaboratively comparing at least two experimental map information of the sample to be tested, and obtaining the crystallographic information and related phase composition information of the sample to be tested through intelligent analysis.

8. The material analysis method according to claim 7, wherein: The intelligent analysis also includes quantitatively calculating the content percentage of each phase in the sample to be tested based on the phase composition information of the sample to be tested.

9. The material analysis method according to claim 7, wherein: The material analysis method also includes performing spectrum preprocessing on the experimental spectrum obtained from the test sample before comparing it with the theoretical spectrum information in the crystal structure database. The spectrum preprocessing includes eliminating impurity peaks, subtracting background, and smoothing the curve of the experimental spectrum information.

10. The material analysis method according to claim 7, wherein: The material analysis method also includes, before comparing with the theoretical spectrum information of the crystal structure database, performing a peak search step, the peak search step including finding characteristic peak information in the experimental spectrum information obtained by testing the sample to be tested; when performing the comparison, setting a peak intensity recognition threshold and a peak spacing recognition threshold according to the characteristic peak information.

11. The material analysis method according to claim 1, wherein: In the crystal structure database, the material structure data obtained by experimental measurement refers to the material structure data obtained by experimental characterization of samples with known structures; The material structure data obtained by theoretical prediction is calculated by at least one of the following calculation methods and obtained after eliminating repeated structures in the structure database based on graph theory methods; The calculation method includes: 1) After element substitution in the material structure data obtained by experimental characterization of samples with known structures, the new material and its material structure data are obtained by combining the density functional theory (DFT) method to relax the structure; 2) Conducting structure search based on particle swarm optimization (PSO) and / or genetic algorithm on the material structure data obtained by experimental characterization of samples with known structures to obtain new materials and their material structure data; 3) Extract features from the material structure data obtained by experimental characterization of samples with known structures, and perform reverse design based on the extracted features to obtain new materials and their material structure data.

12. A material analysis system based on a crystal structure database, characterized by: It includes crystal structure database, theoretical spectrum information calculation module, data preprocessing module, comparison module and result output module; The crystal structure database contains material structure data obtained by experimental measurement and / or theoretical prediction, wherein the material structure data includes chemical formula, space group, unit cell parameters, and specific coordinates of atoms in the unit cell; The theoretical spectrum information calculation module includes a module for calculating theoretical spectrum information based on the structural data of each material in the crystal structure database; The data preprocessing module includes a module for obtaining experimental atlas information obtained by testing the sample to be tested and performing denoising processing on the experimental atlas information; The comparison module includes a module for comparing the result of the data preprocessing module with the theoretical spectrum information calculated by the theoretical spectrum information calculation module; The result output module is configured to output crystallographic information and related phase composition information of the sample to be tested through intelligent analysis based on the results of the comparison module, wherein the crystallographic information includes at least one of space group, unit cell parameters, and specific coordinates of atoms in the unit cell; The intelligent analysis includes: (1) When the experimental spectrum information matches the theoretical spectrum information, the crystallographic information and related phase composition information of the sample to be tested are directly obtained based on the matching result; (2) In the case where the experimental map information does not match the theoretical map information, 1) searching a crystal structure database for theoretical map information of a structure isomorphic to the experimental map information, refining the theoretical map information of the isomorphic structure so that the refined theoretical map information matches the experimental map information, and obtaining crystallographic information and related phase composition information of the sample to be tested based on the matching results; 2) In the case where no theoretical spectrum information of the isomorphic structure is retrieved, the traditional crystallographic method and machine learning method are combined to collaboratively build a structural model of the sample to be tested using positive space search, simulated annealing and genetic algorithm. The theoretical spectrum information of the sample to be tested is calculated based on the built structural model, and the theoretical spectrum information of the sample to be tested is refined so that the experimental spectrum information obtained from the test of the sample to be tested matches the refined theoretical spectrum information of the sample to be tested. The crystallographic information and related phase composition information of the sample to be tested are obtained based on the matching results.

13. The material analysis system according to claim 12, characterized in that: The data preprocessing module also includes a peak finding step, which includes finding characteristic peak information in the experimental spectrum information obtained by detecting the sample to be tested; so that the comparison module can set the peak intensity recognition threshold and the peak distance recognition threshold according to the characteristic peak information when performing the comparison.

14. The material analysis system according to claim 13, characterized in that: The data preprocessing module also includes a module for identifying the contents of files of different formats of the experimental atlas information of the sample to be tested, and reading the relevant information in the file according to the file identification result for subsequent comparison.

15. The material analysis system according to claim 13, wherein: The material analysis system further includes a crystal structure database retrieval module, which is used to extract and present the crystal structure database as a whole.

16. The material analysis system according to claim 15, characterized in that: The overall extraction of the crystal structure database includes retrieving material structure data of the corresponding structure in the crystal structure database by at least one of ICSD number, chemical formula and constituent elements, wherein the material structure data includes chemical formula, space group, unit cell parameters, and specific coordinates of atoms in the unit cell.

17. The material analysis system according to claim 13, characterized in that: The material analysis system also includes a spectrum calculation module, which is used to calculate the spectrum of a given structure and to perform broadening calculations on a spectrum containing only peak position and peak intensity data.

18. The material analysis system according to claim 17, characterized in that: The stretching calculation includes four types of stretching of the intensity value: Gaussian, Lorentzian, Voigt, and convolution, and accumulating the overlapping parts.

19. The material analysis system according to claim 13, wherein: The material analysis system also includes a single peak search module, which is used to search and present the crystal structure database based on mouse peak selection, manual peak search, or input of specific peak position, peak intensity and accuracy data.

20. A material analysis system based on a crystal structure database, characterized by: including memory and processor; The memory is used to store programs; The processor is used to implement the material analysis method according to any one of claims 1 to 11 by executing the program stored in the memory.

21. A computer-readable storage medium, characterized in that: The invention comprises a program stored therein, wherein the program can be executed by a processor to implement the material analysis method according to any one of claims 1 to 11.

22. Application of the material analysis method according to any one of claims 1 to 11, or the material analysis system according to any one of claims 12 to 20, or the computer-readable storage medium according to claim 21 in new material research and development, unknown material structure analysis, or unknown material composition analysis.

Images