JP2010060389A - Particle analyzer, data analyzer, x-ray analyzer, particle analysis method and computer program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a particle analyzer capable of suitably categorizing particles in a sample based on detection results of characteristic X-rays, and discriminating and displaying the particles, and to provide a data analyzer, X-ray analyzer, particle analysis method, and computer program. <P>SOLUTION: The X-ray analyzer generates an SEM image of a sample S by an SEM image processing part 14, detects a plurality of particles in the sample S based on the SEM image, irradiates each particle with electron beams (radiation beams) from an electron gun 11, and detects characteristic X-rays generated from each particle by an X-ray detector 21. An X-ray data analysis part 23 determines the amounts of a plurality of elements contained in each particle from the detection result of the characteristic X-rays. A particle analyzer 3 categorizes a large number of particles contained in the sample S into a plurality of groups in accordance with the composition by performing principal component analysis on the amounts of the plurality of elements in each particle and performing hierarchical cluster analysis using a plurality of principal component scores obtained by the principal component analysis. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to X-ray analysis that performs elemental analysis using characteristic X-rays, and more specifically, particle analysis that appropriately classifies particles in a sample detected by automatic particle analysis based on the results of elemental analysis results. The present invention relates to an apparatus, a data analysis apparatus, an X-ray analysis apparatus, a particle analysis method, and a computer program.

  X-ray analysis is an analysis method in which a substance such as an electron beam or X-ray is irradiated on a substance, characteristic X-rays generated from the substance are detected, and elements contained in the substance are identified from the spectrum of the characteristic X-rays. In addition, the sample is irradiated by irradiating the sample with a radiation beam in the form of a beam, and each point on the sample is analyzed for characteristic X-rays when irradiated with the beam. An elemental distribution is also obtained. Patent Document 1 discloses a technique for measuring the distribution of substances contained in a sample by scanning the sample with an electron beam.

  When the radiation applied to the substance is X-rays, the characteristic X-rays are called fluorescent X-rays, and the X-ray analyzer is provided as a fluorescent X-ray analyzer. Some X-ray analyzers that use an electron beam as a radiation for irradiating a substance are attached to an electron microscope and share an electron beam source with the electron microscope. The X-ray analyzer includes an energy dispersive X-ray analyzer that detects characteristic X-rays with a semiconductor detector, generates a current proportional to the energy of the characteristic X-ray, and selects and measures the generated current; There is a wavelength dispersive X-ray analyzer that measures characteristic X-rays after spectroscopic analysis. EDS or EDX is used as an abbreviation for Energy Dispersive X-ray Spectrometer, and WDS or WDX is used as an abbreviation for Wavelength Dispersive X-ray Spectrometer. In the following, EDS and WDS are used.

An apparatus that combines a scanning electron microscope (SEM) and EDS includes automatic particle analysis as a method of performing measurement by combining SEM and EDS. In automatic particle analysis, an SEM image of a sample is once acquired, particles having different brightness from the surroundings are detected in the SEM image, each particle is irradiated again with an electron beam, and characteristic X-rays from each particle are detected. Qualitative and quantitative analysis of elements contained in particles. The substance of the particle is a fine particle adhering to the surface of the sample or a content in the sample. As a result of the automatic particle analysis, for each particle, characteristic data such as the shape (aspect ratio) of the particle, the spectrum of characteristic X-rays generated from the particle, the type and content of the element contained in the particle, and the like are obtained.
JP-A-8-86762

  In automatic particle analysis, a large number of particles are usually detected in a sample, and a plurality of elements are identified as elements contained in each particle. Here, if the characteristic data of interest such as the area of the particle or the content of the specific element is limited, it is possible to identify and display the particles so that the particles can be classified and determined based on the value of the characteristic data. However, there are many cases where it is not determined which characteristic data should be focused on among a plurality of characteristic data, and when the sample contains many kinds of elements, many characteristic data are slightly different among particles. The result is obtained, and it is difficult to determine what criteria should be used for classification and identification display. Therefore, in practice, it is difficult to perform objectively meaningful classification / identification display on the particles in the sample from the result of the automatic particle analysis.

  The present invention has been made in view of such circumstances, and the object of the present invention is to perform multivariate analysis on the detection result of characteristic X-rays, so that the result of automatic particle analysis An object of the present invention is to provide a particle analysis device, a data analysis device, an X-ray analysis device, a particle analysis method, and a computer program that can appropriately classify and identify particles.

  The particle analysis apparatus according to the present invention analyzes a plurality of particles contained in a sample based on the result of analyzing characteristic X-rays generated from each particle by irradiating each of the plurality of particles contained in the sample with a radiation beam. In the particle analysis device to analyze, for each of a plurality of particles, the content of a plurality of elements obtained by analyzing the characteristic X-rays or the count number of the characteristic X-ray spectrum corresponding to each of the plurality of elements is stored. And a classifying unit that classifies the plurality of particles into a plurality of groups by performing cluster analysis based on the content or the count number of each of the plurality of particles stored in the storage unit. It is characterized by providing.

  The particle analysis apparatus according to the present invention obtains a plurality of principal component scores for each particle by performing principal component analysis on the content or the count number relating to each of the plurality of particles stored in the storage unit. The apparatus further comprises principal component analysis means, wherein the classification means is configured to perform cluster analysis on a plurality of principal component scores relating to each of the plurality of particles obtained by the principal component analysis means. To do.

  A data analysis apparatus according to the present invention is a data analysis apparatus that analyzes a detection result of characteristic X-rays generated from each particle by irradiating each of a plurality of particles included in a sample with a radiation beam. By analyzing the detection result of the characteristic X-rays generated from each particle, the content of a plurality of elements contained in each of the plurality of particles contained in the sample, or the characteristic X-ray spectrum corresponding to each of the plurality of elements A means for obtaining a count number, and a particle analyzer according to the present invention for performing a process of analyzing a plurality of particles based on the content or the count number obtained by the means are provided.

  An X-ray analyzer according to the present invention is an X-ray analyzer that analyzes a sample using characteristic X-rays generated from the sample by irradiating the sample with radiation, and an irradiation unit that irradiates the sample with a radiation beam; Means for acquiring an image of the sample, means for detecting a plurality of particles corresponding to a closed region having different brightness from the surroundings in the image acquired by the means, and for each particle detected by the means, A means for irradiating a radiation beam from the irradiation means, a means for detecting characteristic X-rays generated from each particle by irradiation of the radiation beam, and a data analysis apparatus according to the present invention for analyzing a detection result of characteristic X-rays by the means It is characterized by providing.

  A particle analysis method according to the present invention is a computer including a storage unit and a calculation unit based on a result of analyzing characteristic X-rays generated from each particle by irradiating each of a plurality of particles contained in a sample with a radiation beam. In the particle analysis method for analyzing a plurality of particles contained in a sample, the content of a plurality of elements obtained by analyzing characteristic X-rays for each of the plurality of particles is stored in a storage unit, By performing principal component analysis on the content of a plurality of elements related to each of a plurality of particles stored in the storage unit or the count number of the characteristic X-ray spectrum corresponding to each of the plurality of elements in the calculation unit A plurality of principal component scores of a plurality of principal components are obtained for each particle, and a cluster analysis is performed on the plurality of principal component scores for each of the obtained plurality of particles in the calculation unit, whereby a plurality of particles are duplicated. Characterized in that it classified to the group.

  The computer program according to the present invention allows a computer to analyze a characteristic X-ray generated from each particle by irradiating each of the plurality of particles contained in the sample with a radiation beam, and to obtain a plurality of elements in each particle. In a computer program for executing a process of analyzing a plurality of particles contained in a sample based on the content, the content of a plurality of elements related to each of the plurality of particles or a characteristic corresponding to each of the plurality of elements By performing principal component analysis on the count number of the X-ray spectrum, obtaining a principal component score of a plurality of principal components for each particle, and a plurality of principal component scores for each of the obtained plurality of particles And performing a process including a step of classifying a plurality of particles into a plurality of groups by performing cluster analysis. That.

  In the present invention, a characteristic X-ray generated from each particle is detected by irradiating each of a plurality of particles included in the sample with a radiation beam, and a plurality of particles contained in each particle are analyzed by analyzing the characteristic X-rays. Classify multiple particles into multiple groups by calculating elemental content or counts of characteristic X-rays corresponding to each element and performing cluster analysis based on elemental content or counts corresponding to the element To do. Thereby, it is possible to classify a plurality of particles in a sample without having noticed characteristic data or a clear standard.

  In the present invention, the principal component analysis is performed on the content of a plurality of elements contained in the plurality of particles in the sample or the characteristic X-ray count corresponding to each element, and the principal component analysis is performed. Cluster analysis is performed using a plurality of principal component scores related to each particle. Performing principal component analysis leads to principal components that can represent the difference in particle composition more significantly than the content of each element as a component to be used in cluster analysis, and to reduce the number of independent components Can do.

  In the present invention, by performing a cluster analysis based on the content of a plurality of elements contained in a plurality of particles in the sample, each of a large number of particles contained in the sample contains many types of elements, Even when the criteria for classifying / identifying and displaying the particles are unknown, it is possible to appropriately classify a large number of particles into a plurality of groups in which particles having similar compositions are collected in one group. In materials such as aluminum alloys, according to the present invention, it is possible to obtain a characteristic diagram showing the distribution of various compounds unevenly distributed in the material or the results of cluster analysis. By analyzing the above, the present invention has an excellent effect, such as being able to obtain appropriate production conditions.

Hereinafter, the present invention will be specifically described with reference to the drawings showing embodiments thereof.
FIG. 1 is a block diagram showing the configuration of the X-ray analyzer of the present invention. The X-ray analyzer of the present invention is a device combining SEM and EDS. The X-ray analyzer includes an electron gun (irradiation means) 11 that irradiates a planar sample S with an electron beam (radiation beam), an electron beam scanning coil 12 that determines the direction of the electron beam, an electron gun 11 and an electron beam scan. And an SEM control unit 15 that controls the operation of the coil 12. The electron gun 11 emits an electron beam according to a control signal from the SEM control unit 15, and the electron beam scanning coil 12 sequentially changes the direction of the electron beam, so that the electron beam scans the sample S while scanning the sample S. Irradiated. The X-ray analyzer includes an electron detector 13 that detects reflected electrons or secondary electrons generated when the sample S is irradiated with an electron beam, and the electron detector 13 generates an SEM image. It is connected to the. In synchronization with the scanning of the sample S by the electron beam, the electron detector 13 sequentially inputs a signal indicating the detected electron intensity to the SEM image processing unit 14. The SEM image processing unit 14 generates an SEM image of the sample S in which the luminance of the pixel corresponding to the point on the sample S is a value corresponding to the electron intensity detected by the electron detector 13 based on the sequentially input signals. Generate the process. The SEM image processing unit 14 includes a display unit such as a liquid crystal display, and can display the generated SEM image on the display unit. The SEM image processing unit 14 performs processing for detecting particles contained in the sample S by performing image processing on the SEM image, and the SEM control unit 15 irradiates each detected particle with an electron beam. To perform the process.

  The X-ray analyzer includes an X-ray detector 21 that detects characteristic X-rays generated from the sample S when the sample S is irradiated with an electron beam. The X-ray detector 21 is connected to a multi-channel analyzer (MCA) 22. It is connected. The X-ray detector 21 has a configuration using a semiconductor detection element such as a Si element as a detection element, generates a pulse current having a current value proportional to the detected characteristic X-ray energy, and generates the generated pulse current. Input to MCA22. The MCA 22 selects the pulse current from the X-ray detector 21 according to the current value, and counts the pulse current of each current value. By the processing of the MCA 22, the relationship between the characteristic X-ray energy and the count number, that is, the characteristic X-ray spectrum is obtained. The MCA 22 is connected to the X-ray data analysis unit 23, and inputs a characteristic X-ray spectrum as a detection result of the characteristic X-rays to the X-ray data analysis unit 23. The X-ray data analysis unit 23 performs qualitative / quantitative analysis of the element that generated the characteristic X-ray based on the spectrum of the characteristic X-ray input from the MCA 22. The X-ray analysis apparatus further includes the particle analysis apparatus 3 of the present invention, and the particle analysis apparatus 3 is connected to the SEM image processing unit 14 and the X-ray data analysis unit 23. The combination of the particle analyzer 3 and the X-ray data analyzer 23 corresponds to the data analyzer of the present invention.

  FIG. 2 is a block diagram showing the configuration of the particle analyzing apparatus 3 of the present invention. The particle analysis device 3 is configured using a general-purpose computer such as a personal computer (PC), a CPU (arithmetic unit) 31 that performs computation, a RAM 32 that stores temporary information generated along with the computation, A drive unit 33 such as a CD-ROM drive for reading information from a recording medium D such as an optical disk and a storage unit 34 such as a hard disk are provided. The CPU 31 causes the drive unit 33 to read the computer program 30 of the present invention from the recording medium D, and stores the read computer program 30 in the storage unit 34. The CPU 31 loads the computer program 30 from the storage unit 34 to the RAM 32 as necessary, and executes processing necessary for the particle analyzer 3 according to the loaded computer program 30. Further, the particle analysis device 3 includes an interface unit 35 connected to the X-ray data analysis unit 23 and the SEM image processing unit 14. Furthermore, the particle analysis device 3 includes an input unit 36 such as a keyboard or a pointing device for inputting information such as various processing instructions operated by the user, and a display unit 37 such as a liquid crystal display for displaying various information. It has.

  The computer program 30 may be downloaded from an external server device (not shown) connected to the particle analysis device 3 via a communication network (not shown) and stored in the storage unit 34. . Further, the particle analysis device 3 may have a form in which a recording unit such as a ROM in which the computer program 30 is recorded is provided inside instead of receiving the computer program 30 from the outside.

  Next, processing performed by the X-ray analysis apparatus having the above configuration will be described. FIG. 3 is a flowchart showing a procedure of processing performed by the X-ray analysis apparatus. First, the X-ray analyzer performs processing for acquiring an SEM image of the sample S (S1). In step S1, the SEM control unit 15 controls the operation of the electron gun 11 and the electron beam scanning coil 12, thereby irradiating the sample S with the electron beam while scanning the sample S. Backscattered electrons or secondary electrons are detected, and the detection result is input to the SEM image processing unit 14. The SEM image processing unit 14 generates a two-dimensional SEM image based on the detection result from the electron detector 13. To do.

  Next, the SEM image processing unit 14 performs a process of detecting a plurality of particles included in the sample S based on the generated SEM image (S2). In step S2, the SEM image processing unit 14 binarizes the SEM image with a predetermined threshold as a boundary, and extracts a closed region in which pixels having luminance equal to or higher than the threshold are accumulated in the binarized SEM image. , Detecting a particle which is a part on the sample S corresponding to the closed region. In addition, by designating a predetermined range for the luminance in the SEM image and extracting the closed region where the pixels having the luminance within the predetermined range are collected, particles that are parts on the sample S corresponding to the closed region are extracted. You may make it detect. Further, the SEM image processing unit 14 adds a particle number to each of the detected plurality of particles, measures the shape of each particle, and stores the shape and position information of each particle in association with the particle number.

  FIG. 4 is a schematic diagram illustrating an example of a binarized SEM image. The part indicated by P in FIG. 4 is a particle, and corresponds to a part where pixels having higher luminance than other parts in the SEM image are integrated. The substance of the particle is a fine particle adhering to the surface of the sample S, an aggregate of impurities contained in the sample S, or a portion of a phase different from the base material in the sample S. These particles have higher luminance than the surrounding base material in the SEM image, and the particles can be detected by extracting a closed region with high luminance. Note that the SEM image processing unit 14 stores a luminance threshold value for binarization in advance. The SEM image processing unit 14 may be configured such that the threshold value can be changed according to an instruction received from the user. Further, the SEM image processing unit 14 may be configured to detect particles corresponding to a closed region having lower luminance than the surrounding base material.

  Next, the X-ray analyzer irradiates each of the detected plurality of particles with an electron beam to detect characteristic X-rays generated from each particle (S3). In step S3, the SEM image processing unit 14 inputs position information indicating the position of each detected particle on the sample S to the SEM control unit 15, and the SEM control unit 15 scans the electron gun 11 and the electron beam according to the position information. By controlling the operation of the coil 12, each of a plurality of particles included in the sample S is irradiated with an electron beam in order. At the time of electron beam irradiation, the SEM control unit 15 may scan the inside of the particle with the electron beam, or may irradiate a representative point in the particle with the electron beam. The X-ray detector 21 detects characteristic X-rays generated from each particle by electron beam irradiation, and the MCA 22 inputs the spectrum of the characteristic X-rays generated from each particle to the X-ray data analysis unit 23.

  FIG. 5 is a characteristic diagram showing an example of a spectrum of characteristic X-rays. The horizontal axis in the figure indicates the energy of characteristic X-rays detected by the X-ray detector 21. The vertical axis in the figure represents the count number, and corresponds to the characteristic X-ray intensity of each energy detected by the X-ray detector 21. The energy value corresponding to the peak included in the spectrum is a value unique to the element that has emitted characteristic X-rays, and the element included in the particle can be identified by obtaining the energy value corresponding to the peak.

  Next, the X-ray data analysis unit 23 performs elemental analysis for identifying the elements contained in each particle and calculating the content of each element by analyzing the spectrum of characteristic X-rays generated from each particle. (S4). The X-ray data analysis unit 23 stores in advance a database that records standard data of characteristic X-rays for each element. In step S4, the X-ray data analysis unit 23 compares the characteristic X-ray standard data in the database with the obtained characteristic X-ray spectrum, and the energy value corresponding to the peak included in the characteristic X-ray spectrum. Based on the above, the elements contained in each particle are identified. Further, the X-ray data analysis unit 23 calculates the content of each element contained in each particle in mass% based on the count number of the peak corresponding to each element. The X-ray data analysis unit 23 inputs an analysis result including the content of each element contained in each particle to the particle analysis device 3. The processing in steps S2 to S4 corresponds to automatic particle analysis processing. Note that the analysis result input to the particle analyzer 3 may be a peak count corresponding to each element.

  The particle analysis device 3 receives the analysis result input from the X-ray data analysis unit 23 by the interface unit 35, and the CPU 31 stores the received analysis result in the storage unit 34 (S5). Further, the SEM image processing unit 14 inputs the shape information indicating the shape of each particle and the position information of each particle to the particle analysis device 3, and the CPU 31 stores the shape information and the position information together with the analysis result in the storage unit 34. Let

  FIG. 6 is a chart illustrating an example of analysis results stored in the storage unit 34. FIG. 6 shows an example in which the number of particles is L (L is a natural number) and the number of elements analyzed by the X-ray data analysis unit 23 is M (M is a natural number). Each of the plurality of particles is distinguished by a particle number, and an aspect ratio and an equivalent circle diameter indicating the shape of the particle are recorded in association with each particle number. In association with each particle number, the content of each element from element 1 to element M contained in the particle is recorded in units of mass%.

  Next, the CPU 31 of the particle analyzer 3 performs principal component analysis on the content of each element related to each particle stored in the storage unit 34 in accordance with the computer program 30 loaded in the RAM 32 (S6). The principal component analysis is a process of generating new components by synthesizing M elements that are contents of M kinds of elements. In step S6, the CPU 31 calculates the average value of the content of each element over all the particles, calculates the standard deviation of the content of each element, and calculates the content of each element related to each particle as (deviation / Standard deviation). By this treatment, the content of the low-concentration element and the content of the high-concentration element can be evaluated equally. The CPU 31 further performs a principal component analysis on the content of M elements related to each of the L particles converted to (deviation / standard deviation), thereby obtaining a first principal component score for each particle. To the N-th principal component score. Here, N is a natural number equal to or less than M. In addition, CPU31 may perform the process which performs a principal component analysis, without converting content of each element into (deviation / standard deviation). In this case, the evaluation according to the difference in content of each element can be performed.

  FIG. 7 is a chart showing an example of the result of the principal component analysis. In association with each particle number, principal component scores from the first principal component score to the Nth principal component score are obtained. Since the obtained main component score is an amount obtained from the content of the element in the particle, it is a component representing the composition of the particle. Since one of the purposes of the principal component analysis in step S6 is to reduce the components rather than the content of M elements, N is set to a value smaller than M. In this way, by performing principal component analysis, a principal component that can more significantly represent the difference in particle composition is derived from the content of a plurality of elements, and the number of independent components that represent the particle composition is reduced. be able to. The value of N may be determined in advance. Further, the value of N may be determined by obtaining a principal component having an eigenvalue of a predetermined value or more, such as selecting a principal component having an eigenvalue of 1 or more as the principal component. Further, the value of N may be determined by obtaining a principal component having a contribution ratio equal to or greater than a predetermined value as the principal component. Further, as the main component, the value of N may be determined by obtaining the main component until the cumulative contribution rate from the first main component reaches a predetermined value.

  Next, the CPU 31 of the particle analyzer 3 performs cluster analysis on the N principal component scores relating to the L particles according to the computer program 30 loaded in the RAM 32 (S7). In step S7, the particle analyzer 3 uses hierarchical cluster analysis as a cluster analysis technique. In the hierarchical cluster analysis, each of L particles is represented as a point on an N-dimensional space with N principal component scores as coordinates on independent coordinate axes, and in the initial state, L particles to which each particle belongs. , Calculate the distance between the clusters in the N-dimensional space, merge the clusters with the nearest distance into one cluster, calculate the distance and merge the clusters until the number of clusters becomes 1. repeat. The particle analysis apparatus 3 of the present invention uses the Ward method as an algorithm for hierarchical cluster analysis. In step S7, the CPU 31 performs a hierarchical cluster analysis to generate a cluster tree diagram that records the process of sequentially merging clusters that are close to each other in the N-dimensional space, and stores the generated cluster tree diagram in the storage unit. 34 is stored.

  FIG. 8 is a schematic diagram showing an example of a cluster tree diagram. The vertical axis in the figure indicates the particle number, and the horizontal axis in the figure corresponds to the distance between two clusters merged into one cluster. In the cluster dendrogram, particle numbers whose lines are connected to each other indicate that particles represented by N-dimensional points are included in one cluster. A plurality of particles included in one cluster are particles whose N principal component scores are closer to each other than other particles not included in the cluster, and the principal component scores are components representing the composition of the particles. Therefore, a plurality of particles included in one cluster are particles having a composition close to each other. That is, a cluster obtained by cluster analysis corresponds to a group in which a plurality of particles are classified by collecting particles having similar compositions. For example, if you specify the number of clusters or the distance between clusters on the cluster dendrogram, multiple clusters each containing one or more particles are specified, and multiple groups that classify multiple particles according to composition Is obtained. Therefore, a plurality of particles included in the sample S can be classified into a plurality of groups according to the composition by the cluster analysis process in step S7. The particle analyzer 3 records the cluster tree diagram by recording the particle numbers of the particles included in each cluster at each stage where the clusters are sequentially merged, for example. Note that the CPU 31 may perform a process of stopping the cluster analysis when the number of clusters reaches a predetermined number in step S7. After the process of step S7 is completed, the X-ray analyzer ends the process.

(Example)
Next, examples of analyzing actual samples using the present invention will be described. As the sample S, an aluminum alloy was used. More specifically, the sample S is an Al—Si—Cu-based ADC 12 known as an aluminum alloy for die casting. This aluminum alloy is widely used for automobile parts, machine parts, household appliances, and the like. When the Si content is too large, gloss is not easily produced, and a color close to brown tends to be obtained. Moreover, mechanical property improves by containing Cu, and machinability improves by containing Pb. Moreover, by containing Mn, an intermetallic compound of Al—Si—Fe—Mn is generated, and the elongation, impact strength, and machinability are reduced. As described above, since the characteristics of the aluminum alloy change depending on the element added, the concentration control of the element is important from the viewpoint of managing the quality of the product.

  The sample S was formed by embedding an aluminum alloy ingot in a resin, and cutting and then polishing the surface. This sample S was analyzed by the X-ray analyzer of the present invention. By the automatic particle analysis in steps S2 to S4, 875 particles are detected from a region of 254 × 190 μm, and the elements contained in the particles are C, O, Mg, Al, Si, V, Cr, Mn, Fe, Elements such as Ni, Cu, Zn, Sn, and Pb were detected. Among these, Al is a main component element of the sample S, and the other is an additive element. The detected particles are intermetallic compounds unevenly distributed in the aluminum alloy, and are considered to be portions having a composition different from that of the surroundings. Among the additive elements, the principal component analysis process of step S6 is performed using the contents of 11 kinds of elements such as Mg, Si, V, Cr, Mn, Fe, Ni, Cu, Zn, Sn, and Pb having a large content. went. That is, in this embodiment, L = 875 and M = 11.

  FIG. 9 is a chart showing the results of principal component analysis in the example. In the figure, the results obtained from the first principal component to the sixth principal component are shown. In this example, the sixth principal component having a cumulative contribution rate of slightly over 70% was obtained and used as a variable. Moreover, in FIG. 9, the ratio in which content of each element is contained in each main component is shown. The first main component is considered to be an index showing a high Cr—Mn—Fe ratio and a Cr—Mn—Fe rich composition. In addition, the second main component is considered to be an index showing a high Mg—Cu ratio and a Mg—Cu rich composition, and the third main component has a high Sn, Pb ratio and a Sn—Pb rich composition. It is considered to be an index indicating Using the principal component score of each particle corresponding to the first to sixth principal components, the cluster analysis in step S7 was performed. That is, in this embodiment, N = 6.

  FIG. 10 is a schematic diagram showing a cluster tree diagram showing the results of cluster analysis in the example. The vertical axis in the figure indicates the particle number, and the horizontal axis corresponds to the distance between two clusters merged into one cluster. As indicated by a broken line in the figure, in this example, data at a stage where the number of clusters was 9 was extracted as a processing result from the result of cluster analysis. That is, according to the treatment of the present invention, 875 particles were classified into 9 groups according to the difference in composition.

  FIG. 11 is a chart showing the classification results in the examples. Each cluster is indicated by a cluster number, and the number of particles contained in each cluster is indicated. Moreover, the average value of the aspect ratio and equivalent circle diameter of the particles classified into each cluster is shown. Furthermore, the average value of the content of each element contained in the particles classified into each cluster is shown.

  FIG. 12 is a characteristic diagram showing the results of cluster analysis in the example as a three-dimensional distribution. FIG. 12 is a diagram in which particles represented by six-dimensional points whose coordinates are determined by six principal component scores are projected onto a three-dimensional space. The X axis, Y axis, and Z axis in the figure correspond to the first principal component, the second principal component, and the third principal component, respectively. The shape of the coordinate point representing each particle is different depending on the cluster into which each particle is classified. By obtaining a distribution diagram of clusters and particles as shown in FIG. 12, it can be confirmed that there is a bias between the distribution state of each cluster and the distribution state of each particle.

  FIG. 13 is a characteristic diagram showing the results of cluster analysis in the example in a two-dimensional distribution. FIG. 13 is a diagram in which particles represented by six-dimensional points whose coordinates are determined by six principal component scores are projected on a plane with the axes of the first principal component and the third principal component as the XY axes. The horizontal axis in the figure indicates the first main component, and the right side indicates that the particles are rich in Cr—Mn—Fe. The vertical axis in the figure indicates the third main component, and the higher the value is, the more the particles are Sn-Pb rich. As shown in the figure, 875 particles are classified into 9 clusters, each of which has a similar composition. The cluster 2 is a cluster composed of one particle although it is buried in another cluster in the figure. Particles classified as cluster 5 have an average elemental composition, particles classified as clusters 4 and 8 are Sn-Pb rich, and particles classified as clusters 6 and 9 are Cr-Mn-Fe rich. Yes, particles classified as cluster 2 are V-rich, particles classified as cluster 3 are Ni-rich, and particles classified as cluster 5 are Si-rich.

  As described above in detail, in the present invention, a principal component analysis is performed on the contents of a plurality of elements contained in a large number of particles contained in the sample S, and a plurality of particles related to each particle obtained by the principal component analysis are analyzed. By performing a hierarchical cluster analysis using the principal component scores, a large number of particles contained in the sample S can be classified into a plurality of groups according to the composition. Even when a large number of particles included in the sample S contain many types of elements and the criteria for classifying the particles are unknown, a plurality of particles having similar compositions in one group are collected according to the present invention. It becomes possible to appropriately classify a large number of particles into the groups. In materials such as aluminum alloys, the present invention makes it possible to obtain distributions of various compounds unevenly distributed in the materials and characteristic diagrams showing cluster distributions. Various compounds unevenly distributed in the material affect the properties of the material such as strength or machinability. By analyzing the material according to the present invention while changing the manufacturing conditions of the material such as the temperature conditions, it is possible to know the change in the distribution of the compound or the change in the characteristic diagram showing the cluster distribution according to the manufacturing conditions, and the material It is also possible to determine the production conditions of the material from which a characteristic diagram showing the distribution or cluster distribution of the compound that provides the characteristics can be obtained.

  In this embodiment, the Ward method is used as the hierarchical cluster analysis algorithm. However, the present invention is not limited to this, and other methods are used as the hierarchical cluster analysis algorithm. It may be in the form. Further, in the present embodiment, the form of performing the hierarchical cluster analysis is shown as the cluster analysis method, but the present invention is not limited to this, and the form of performing the cluster analysis by a method other than the hierarchical cluster analysis is shown. There may be. Further, in the present embodiment, a mode in which cluster analysis is performed after principal component analysis is performed on the content of each element in each particle is shown, but the present invention is not limited to this, and each element is not limited to this. Alternatively, the cluster analysis may be performed directly on the content of. Further, in the present embodiment, an embodiment has been shown in which a process for classifying particles based on the content of each element in each particle is shown, but the present invention is not limited to this, and the characteristic X corresponding to each element The form which performs the process which classifies particle | grains based on the count number of a line spectrum may be sufficient.

  In the present embodiment, the X-ray analyzer of the present invention is shown in the form in which the EDS is attached to the SEM. However, the present invention is not limited to this, and the X-ray analyzer of the present invention is not a SEM but a transmission electron. The form which attached EDS to the microscope (TEM: Transmission Electron Microscope) may be sufficient. In the case of this form, the X-ray analyzer acquires a TEM image of the sample S and performs a process of detecting particles in the sample S by performing image processing on the acquired TEM image. Further, the X-ray analyzer of the present invention may be in the form of using X-rays instead of electron beams as a radiation beam. In this embodiment, the X-ray analyzer includes an X-ray irradiation unit and an optical microscope, acquires an optical image obtained by photographing the sample S using the optical microscope, and performs image processing on the acquired optical image. Thus, the present invention is implemented by detecting particles in the sample S and irradiating each particle with an X-ray beam. In the present embodiment, an EDS is used as an X-ray analyzer, but the present invention is not limited to this, and an embodiment using WDS as an X-ray analyzer may be used.

  Further, in the present embodiment, the particle analysis device 3 is configured using a PC, and an embodiment is shown in which processing is executed after the results of elemental analysis are stored in the internal storage unit 34. However, the present invention is not limited to this. Instead, the particle analysis apparatus 3 may be configured to store the result of elemental analysis in a storage unit outside the PC and execute processing based on the stored contents of the storage unit. In the present embodiment, the configuration in which the X-ray data analysis unit 23 and the particle analysis device 3 are separated is shown, but the configuration of the present invention is not limited to this. For example, the configuration of the present invention may be a configuration in which the functions of the X-ray data analysis unit 23 and the particle analysis device 3 are realized by a data analysis device realized by a single computer. The configuration of the present invention may be a configuration in which the function of the MCA 22 is further integrated with the data analysis apparatus. In the present embodiment, the particle analysis device 3 or the data analysis device of the present invention has been incorporated into an X-ray analysis device using an SEM. However, the configuration of the present invention is not limited to this. The particle analysis device 3 or the data analysis device of the present invention may have a configuration independent of the SEM. In the case of this form, the particle analysis device 3 or the data analysis device inputs measurement data by SEM and EDS using a communication network or a recording medium, and performs a process of analyzing the input measurement data.

It is a block diagram which shows the structure of the X-ray analyzer of this invention. It is a block diagram which shows the structure of the particle | grain analysis apparatus of this invention. It is a flowchart which shows the procedure of the process which an X-ray analyzer performs. It is a schematic diagram which shows the example of the binarized SEM image. It is a characteristic view which shows the example of the spectrum of characteristic X-ray. It is a graph which shows the example of the analysis result which a memory | storage part memorize | stores. It is a graph which shows the example of the result of a principal component analysis. It is a schematic diagram which shows the example of a cluster dendrogram. It is a graph which shows the result of the principal component analysis in an Example. It is a schematic diagram which shows the cluster dendrogram which shows the result of the cluster analysis in an Example. It is a chart which shows the classification result in an example. It is a characteristic view which shows the result of the cluster analysis in an Example by three-dimensional distribution. It is a characteristic view which shows the result of the cluster analysis in an Example by two-dimensional distribution.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Electron gun 12 Electron beam scanning coil 13 Electron detector 14 SEM image processing part 15 SEM control part 21 X-ray detector 22 MCA
23 X-ray data analysis unit 3 Particle analyzer 30 Computer program 31 CPU
34 Storage unit 35 Interface unit S Sample

Claims (6)

In a particle analysis apparatus for analyzing a plurality of particles contained in a sample based on a result of analyzing characteristic X-rays generated from each particle by irradiating each of the plurality of particles contained in the sample with a radiation beam,
Storage means for storing the content of a plurality of elements obtained by analyzing characteristic X-rays for each of a plurality of particles, or the count number of characteristic X-ray spectra corresponding to each of the plurality of elements;
Classifying means for classifying the plurality of particles into a plurality of groups by performing cluster analysis based on the content or the count number of each of the plurality of particles stored in the storage means, Particle analyzer. A principal component analysis means for obtaining a plurality of principal component scores for each particle by performing principal component analysis on the content or the count number of each of the plurality of particles stored in the storage means;
2. The particle according to claim 1, wherein the classifying unit is configured to perform cluster analysis on a plurality of principal component scores relating to each of the plurality of particles obtained by the principal component analyzing unit. Analysis device. In a data analysis apparatus for analyzing a detection result of characteristic X-rays generated from each particle by irradiating each of a plurality of particles contained in a sample with a radiation beam,
By analyzing the detection result of the characteristic X-rays generated from each particle by irradiation of the radiation beam, it corresponds to the content of a plurality of elements contained in each of the plurality of particles contained in the sample, or to each of the plurality of elements. Means for determining the count of the characteristic X-ray spectrum;
A data analysis apparatus comprising: the particle analysis apparatus according to claim 1 or 2 that performs a process of analyzing a plurality of particles based on the content or the count number obtained by the means. In an X-ray analyzer for analyzing a sample using characteristic X-rays generated from the sample by irradiating the sample with radiation,
An irradiation means for irradiating the sample with a radiation beam;
Means for obtaining an image of the sample;
Means for detecting from the sample a plurality of particles corresponding to a closed region having a brightness different from that of the surroundings in the image acquired by the means;
Means for irradiating each particle detected by the means with a radiation beam from the irradiation means;
Means for detecting characteristic X-rays generated from each particle by irradiation of a radiation beam;
An X-ray analysis apparatus comprising: the data analysis apparatus according to claim 3 that analyzes a detection result of characteristic X-rays by the means. Based on the result of analyzing characteristic X-rays generated from each particle by irradiating each of a plurality of particles contained in the sample with a radiation beam, a plurality of particles contained in the sample are included using a computer having a storage unit and a calculation unit. In the particle analysis method for analyzing particles of
For each of a plurality of particles, the content of a plurality of elements obtained by analyzing characteristic X-rays is stored in a storage unit,
By performing principal component analysis on the content of a plurality of elements related to each of a plurality of particles stored in the storage unit or the count number of the characteristic X-ray spectrum corresponding to each of the plurality of elements in the calculation unit , Find principal component scores of multiple principal components for each particle,
A particle analysis method characterized by classifying a plurality of particles into a plurality of groups by performing cluster analysis on a plurality of principal component scores relating to each of the obtained plurality of particles in an arithmetic unit. Included in the sample based on the content of multiple elements in each particle obtained by analyzing characteristic X-rays generated from each particle by irradiating each of the multiple particles contained in the sample with a radiation beam on a computer In a computer program for executing a process of analyzing a plurality of particles,
On the computer,
By performing principal component analysis on the content of a plurality of elements related to each of a plurality of particles or the count number of the characteristic X-ray spectrum corresponding to each of the plurality of elements, Obtaining a principal component score;
A computer program that executes a process including: classifying a plurality of particles into a plurality of groups by performing cluster analysis on a plurality of principal component scores relating to each of the plurality of particles obtained.

Images