JP2014181951A - Method for discriminating foreign substances in metal - Google Patents

Abstract

PROBLEM TO BE SOLVED: To identify and discriminate foreign substances such as inclusion and deposition in a metal in a large amount and in a short time.SOLUTION: Elements and intensities thereof included in characteristic X-ray information obtained from inclusion or deposition in a metal are acquired. An intensity correlation between one element except elements O, S, N, C included in the characteristic X-ray information and the elements O, S, N, C is obtained, and a correlation between both of the elements is obtained except data equal to or less than a specified value of the one element. A prescribed correlation range is applied to a characteristic X-ray intensity ratio table of oxide, sulfide, nitride, and carbide, and it is determined whether the correlation between both of the elements satisfies the prescribed correlation range to the characteristic X-ray intensity ratio table of the oxide, sulfide, nitride, and carbide of the one element. When the prescribed correlation is satisfied, the inclusion or the deposition is identified understanding that there is any of the oxide, sulfide, nitride, and carbide of the one element or a composite thereof. Such an identification is repeated for the elements (except the elements O, S, N, C) included in the characteristic X-ray information except the one element.

Description

  The present invention provides composition information obtained by a physical analyzer using characteristic X-rays capable of morphological observation and composition analysis such as scanning electron microscope (SEM) and electron probe microanalyzer, and particle size information and position information by image analysis. The present invention relates to a method for accurately deriving the size and number distribution of each type of inclusions and precipitates present in a metal by analysis.

  In recent years, with the establishment of clean steel manufacturing technology, there is a strong demand for technology for rapid and large-scale processing with improved evaluation accuracy for the evaluation of cleanliness of manufactured steel. It is coming.

  In particular, rapid evaluation of the cleanability of continuous casting (hereinafter also referred to as continuous casting), which is the raw material of various steel materials, will further increase the needs in the future to prevent the occurrence of a large number of product defects. Can be considered.

  By the way, the internal defects of the slab are roughly classified into hollow defects called pinholes and non-metallic inclusions (hereinafter also simply referred to as inclusions). Among these, non-metallic inclusions include spherical inclusions and massive inclusions due to the generation factors. In addition, there are intragranular precipitates and grain boundary precipitates (hereinafter simply referred to as precipitates) having different generation factors from nonmetallic inclusions.

  For example, Ar gas is blown into a dip tube that injects molten steel from a tundish into a mold to prevent clogging of alumina, but if this blowing amount is not appropriate, the Ar gas is brought into the mold. There is. This Ar gas is the main cause of the pinhole.

  Further, in continuous casting, it is common to provide lubrication by interposing mold powder between the mold and the slab so as not to be seized. However, when the molten metal surface fluctuation in the mold is large, unmelted mold powder is trapped by the meniscus of the slab (the end of the slab at the initial stage of solidification). This mold powder has been found to be the cause of the formation of massive inclusions.

  In general hot metal refining, a large amount of oxygen is blown into molten steel in a converter to reduce C. At this time, various non-metallic inclusions are generated at the same time. Usually, non-metallic inclusions have a lower specific gravity than molten steel, so they float on the top of the molten steel pan before continuous casting, but may be brought to the tundish without being able to float. The non-metallic inclusions are alumina-based oxides that are networked, which is the main cause of the formation of alumina clusters.

  On the other hand, in the process of solidifying steel in a continuous casting machine, a phenomenon occurs in which elements such as Mn, Al, S, Nb, and Ti contained in the molten steel form compounds and precipitate at grain boundaries. These grain boundary precipitates are known to weaken the grain boundary strength depending on the composition of precipitation.

  In order to effectively remove or detoxify these grain boundary precipitates and the non-metallic inclusions described above, it is important to accurately grasp the number distribution by size according to form and composition.

  Conventionally, the following several cleanliness evaluation methods have been used for these requirements, but each has its own problems.

  For example, the total oxygen analysis method and the sand alumina analysis method allow rapid and large-scale analysis, but cannot determine the particle size distribution, shape, and type necessary for inclusion evaluation of clean steel.

  On the other hand, in the microscopic method and the accuracy method such as the electrolytic slime extraction method that directly observes the inclusion extracted by electrolytic melting of the sample cut out from the slab using a stereomicroscope or SEM, the shape and type can be distinguished. Is possible. However, these methods are difficult to automate sample processing and result calculation. Therefore, since these methods are man-made operations, the workload of the operator is large, and it takes time for the determination. Therefore, it is unsuitable for analyzing a large amount of inclusions using a statistical method, and only a small number of samples can be evaluated, and it is difficult to evaluate the entire slab.

  In order to solve these problems and to perform a cleanness evaluation test quickly and in large quantities, a fully automatic flaw detection system that introduces a signal analysis function and a flaw detection function using a neutral network to a high-frequency ultrasonic C-scope flaw detector This is proposed in Patent Document 1. However, in the flaw detection system proposed in Patent Document 1, pinholes and inclusions can be recognized, but the type of inclusion cannot be identified. Hereinafter, high-frequency ultrasonic C-scope flaw detection is simply referred to as high-frequency ultrasonic flaw detection.

  Further, Patent Document 2 proposes a technique for discriminating defect information obtained by high-frequency ultrasonic flaw detection according to the form of inclusions using a technique applying an extreme value statistical method. However, although the method proposed in Patent Document 2 can determine the number-specific number distribution after discriminating inclusions by form, it is impossible to grasp the composition of inclusions.

  Furthermore, Patent Document 3 and Patent Document 4 propose methods for obtaining the form and average composition of inclusions using X-rays. However, the technique proposed in Patent Document 3 only classifies the detected inclusions according to the form, and cannot be identified in detail. Moreover, the method proposed in Patent Document 4 only shows the average composition of inclusions, and has not yet been identified.

Japanese Patent Laid-Open No. 10-62357 JP 2006-184101 A Special table 2012-507002 gazette JP 2004-191060 A

  The problem to be solved by the present invention is that in the case of the conventional method, it is possible to grasp the foreign matter such as inclusions and precipitates existing in the metal quickly and in large quantities, and the number distribution by size according to form and composition. It is a point that cannot be done.

  In order to solve the above-mentioned problems, the present invention is a method for identifying and discriminating inclusions and precipitates present in a metal, for example, measurement of non-metallic inclusions and grain boundary precipitates for evaluating the cleanliness of steel. The present invention proposes a method for discriminating foreign substances in metals that enables rapid and large-scale evaluation by type.

  The identification refers to determining what kind of inclusion or precipitate is what is confirmed (analyzed) what kind of element is contained in the inclusion or precipitate.

The foreign matter discrimination method in the metal of the present invention,
The most important feature is to identify inclusions or precipitates contained in the metal material by the following steps (1) to (7).

(1) An element included in the characteristic X-ray information obtained from the individual inclusions or precipitates by irradiating the individual inclusions or precipitates with an electron beam using a scanning electron microscope or an electron beam microanalyzer The process of acquiring strength.
(2) One element excluding O, S, N, and C elements and each element of O, S, N, and C included in the characteristic X-ray information obtained from the individual inclusions or precipitates The process of calculating the intensity correlation.
(3) In the intensity correlation, except for data less than a predetermined value of the one element included in the characteristic X-ray information, the correlation between the one element and each element of O, S, N, and C The process to seek.
(4) A step of giving a predetermined correlation range to the table of characteristic X-ray intensity ratios of oxides, sulfides, nitrides, and carbides.
(5) The characteristic X-ray intensity ratio of the oxide, sulfide, nitride and carbide of the one element is the correlation between the one element and each of O, S, N and C in the process of (3) Determining whether or not a predetermined correlation range is satisfied in the table.
(6) When the predetermined correlation is satisfied, an inclusion or a precipitate is identified as any one of the oxide, sulfide, nitride, carbide, or composite of the one element. Process.
(7) The calculation is performed by removing the intensity of the one element and the corresponding O, S, N, and C from the process of (1), and excluding the one element, the elements included in the characteristic X-ray information (O, Steps (2) to (6) are repeated for (except S, N, and C elements).

  According to the method of the present invention described above, a large amount of inclusions and precipitates present in the metal are analyzed in a short time from inclusions and precipitates that have been originally combined, and the analyzed inclusions and precipitates are analyzed. It becomes possible to identify the type.

  In the present invention, the scanning electron microscope or the electron beam microanalyzer further comprises a backscattered electron detector or an angle selective backscattered electron detector. By analyzing the image data of the detector and the image data, inclusions or precipitates, By measuring the size or arrangement or size and position and identifying the composition of inclusions or precipitates from the characteristic X-ray information, it is possible to quickly evaluate the grain boundary precipitates.

  In the present invention, a large amount of non-metallic inclusions and precipitates present in the metal material can be quickly identified and the particle size distribution can be evaluated for each type of inclusion and precipitate in the measurement material, This can contribute to preventing the occurrence of product defects.

  Also, the correspondence between the image of the grain boundary position of the crystal grains such as the backscattered electron (BSE) image and the angle selective backscattered electron (AsB) image, and the determined position information of each inclusion or crystal By confirming, the grain boundary precipitates can be quickly evaluated.

It is the figure which showed an example of the identification process of a composite inclusion and a precipitate. It is a distribution map reflecting the position information of particles analyzed as TiN. It is an image of the crystal grain of the measurement range by an AsB image. It is the figure which showed the result of having overlapped FIG. 2 and FIG. 3 and having counted the intragranular precipitate. It is the figure which showed the result of carrying out the particle size analysis of the intragranular precipitate and grain boundary precipitate in TiN. It is the figure which showed an example of the element pair by which the correlation was acquired by statistical processing based on characteristic X-ray information. 6A and 6B are diagrams showing correlations with other element pairs in the particles classified in FIG. 6, wherein FIG. 6A shows the correlation between Mn and S, and FIG. 6B shows the correlation between Ti and N. FIG. is there. It is a view showing a change in the hit rate of the inclusions and precipitates to be detected by the threshold R _t of the correlation coefficient.

  In general, inclusions and precipitates are rarely present alone in steel, and a plurality of inclusions are often mixed.

  The inclusions and precipitates in these steels can be identified mainly by SEM and the attached energy dispersive X-ray analyzer (EDX) and electron microanalyzer (EPMA). However, in particular, identification of complex inclusions and precipitates is analyzed based on the positional relationship and concentration profile of each element detected by element mapping, and thus it takes a long time to identify one particle. .

  By the way, when point analysis is performed on particles using EDX or the like, it is difficult to identify which other element O, S, N, etc. are combined with. Further, when a plurality of inclusions and precipitates are mixed, it is difficult to identify complex inclusions and precipitates only by point analysis.

  On the other hand, in the evaluation of the grain boundary precipitates, the measurer judged whether or not each particle was at the grain boundary from the SEM image.

  In addition, it is possible to clarify the positional relationship between the grain boundaries and inclusions and precipitates using the electron backscatter diffraction method (scanning electron microscope-crystal orientation analysis: EBSD). Both methods take a long time.

  In other words, in the conventional particle measurement, the measurer mainly displays a list of measurement particle information such as particle size and composition analysis results, and discriminates only particles that meet a predetermined condition. There was a drawback that the setting of became subjective. In addition, particles that do not meet the set conditions have to be judged again by the measurer, and a huge amount of work is required.

  The present invention provides a threshold value in the characteristic X-ray information for the purpose of improving the above-mentioned conventional defects and identifying and discriminating a large amount of foreign substances such as inclusions and precipitates existing in a metal material in a short time. This is achieved by excluding uncorrelated particles.

  That is, in the present invention, based on the composition information detected from a plurality of particles, the binding element is specified by utilizing the correlation of the composition ratio, and from the inclusions and precipitates that exist as a single element to the composite elements Discriminate.

  Further, by statistically processing the composition information of each particle, the composition determination of a large amount of particles and the inclusions and precipitates are identified in a short time. In practice, it is roughly classified by using inclusions and precipitates composed of elements that are correlated by statistical processing, and inclusions and precipitation are determined using thresholds determined in advance from the constituent element ratios. Discriminate objects finely.

Hereinafter, modes for carrying out the invention discovered by the inventors will be described below.
Use a sample whose surface is polished with diamond and mirror finished. Further, an appropriate etching process is performed so that the crystal grain boundaries on the observation surface and the inclusion particles and precipitate particles to be detected can be detected. By performing this process, it becomes possible to obtain the correlation between the grain boundary precipitates and the grain boundaries or metal grains.

  The analysis sample is observed at an arbitrary magnification, and an image analysis application capable of acquiring particle size information and position information of inclusion particles and precipitate particles is used in the following procedure.

(A) By the observation, an image such as a BSE image or an AsB image having different contrasts depending on the existing elements is acquired. In addition, inclusions and precipitates that are imaged with different contrasts are subjected to image processing to obtain particle shape information such as particle size.

(B) Inclusion particles and precipitate particles are basically based on identifying inclusions and precipitates from characteristic X-ray information in the analysis region using EDX.

(C) Each element is statistically processed based on characteristic X-ray information detected from a plurality of particles, and plotted on a graph with element pairs exhibiting a correlation R ≧ 0.4, for example. Since a plurality of pieces of information such as spectrum intensity and quantitative values (mass%, atomic%) are obtained from the characteristic X-ray detection result, information obtained from the characteristic X-ray detection result is referred to as characteristic X-ray information.

(D) For the purpose of improving the accuracy of correlation, a threshold value is set based on characteristic X-ray information, and particles having no correlation are excluded. This threshold value varies depending on the chemical composition of the measurement sample and the measurement conditions, but the threshold value can be set within the range shown in Table 1 below. In addition, when there are inclusions and precipitates for which sufficient correlation cannot be obtained due to the small number of detections, elements having a high ratio of composition information to all detected elements are extracted, and the characteristics are within the ranges shown in Table 1 below. By providing an X-ray intensity threshold, uncorrelated particles can be excluded.

(E) The composition is identified from the characteristic X-ray information ratio (mass ratio: mass%, atomic mass ratio: atomic%, characteristic X-ray intensity), particle shape, etc. of the elements in which the correlation is observed. Table 2 below shows the case where the characteristic X-ray intensity, mass%, and atomic% are used as criteria for identifying main inclusions and precipitates. That is, the composition ratio is determined according to the criteria shown in Table 2 below. Next, the quantitative inclusion amount and the precipitate amount are calculated for each type. The characteristic X-ray intensity includes the total amount (Count) in which characteristic X-rays are detected and the amount detected per second (Count per second: CPS). Tables 1 and 2 indicate the total amount.

(F) The complex inclusion system is identified by performing the same process step by step in the order shown in FIG.

(G) The positional information (see FIG. 2) of each identified inclusion and each precipitate is compared with the information on the crystal grains obtained in the BSE image or AsB image (see FIG. 3) and shown in FIG. A grain boundary precipitate counting process and a grain size analysis process as shown in FIG. 5 are performed.

  According to the method of the present invention carried out in the above procedures (a) to (g), it is possible to evaluate the particle size distribution of each inclusion and each precipitate and to quickly quantitatively evaluate the grain boundary precipitate. In particular, by setting a threshold value for characteristic X-ray information and excluding uncorrelated particles, it is possible to quickly evaluate by type in measuring non-metallic inclusions and grain boundary precipitates, which was not possible with conventional methods. .

  Inclusions or precipitates are often present in a composite of oxides, sulfides, nitrides, and the like, and when the correlation coefficient R is set high, the composite particles are overlooked. On the other hand, if the correlation coefficient R is set to a low value such as less than 0.4, excessive data unrelated to inclusions or precipitates increases, making rapid evaluation difficult.

It shows the change in the hit rate of the inclusions and precipitates to be detected by the threshold R _t of the correlation coefficient in FIG. As shown in FIG. 8, select the 0.4 value as the threshold R _t of the correlation coefficient, the correlation coefficient R R _t or more, when a 1.0 or less, the hit rate is 100%. On the other hand, when a value greater than 0.4 is selected as the correlation coefficient threshold value R _t and the correlation coefficient R is set to R _t or more and 1.0 or less, the hit rate decreases from 100%.

Also, select the upper limit 0.4 the following values to the threshold R _t of the correlation coefficient, below its R _t correlation coefficient R, 0 or more to the hit rate becomes 0%. On the other hand, select the 0.4 value greater than the threshold value R _t in the upper limit of the correlation coefficient, below its R _t correlation coefficient R, the hit rate when the zero or increases from 0%, the correlation coefficient R At 1.0, the hit rate is 100%.

From the above, it directed to a data correlation coefficient is 1.0 or less, and the threshold value R _t in the lower limit of the correlation coefficient is 0.4. Thereby, inclusions or precipitates can be analyzed without omission with a hit rate of 100%. Therefore, the effect of the method of the present invention becomes more prominent by selecting an element pair whose correlation coefficient R that achieves a hit rate of 100% satisfies 0.4 ≦ R ≦ 1.0.

Next, an example in which the present invention is applied to a steel material will be specifically described.
The sample used was a diamond whose surface was mirror-polished and appropriately etched. The measurement was performed using a field emission scanning electron microscope (FE-SEM) having an analysis application capable of acquiring particle size information and position information by image analysis of inclusion particles and precipitate particles. As the measurement conditions, an electric current and a diaphragm obtained by 10000 cps with Cu-Kα ray at an acceleration voltage of 12 KV were selected, and the EDX spectrum was calibrated. The magnification was 2000 times, and the captured image was adjusted so that the inclusions and precipitates to be detected can be easily detected by image processing. Further, an element detected by EDX analysis was selected, and a particle having a particle size of 0.01 to 10 μm ² was measured for a measurement field of view 3 mm ² .

  The data shown in Table 3 below obtained by measurement was output to general-purpose data analysis software, and the characteristic X-ray information of each element was statistically processed.

In order to remove the part (particles) in which the correlation is not observed and plot the element pairs in which the correlation is observed in the graph, a threshold is set in the spectrum information as shown in FIG. In this way, by setting a threshold value and selecting only correlated particles, the composition of the particles is identified from the ratio of spectral data with the correlated element pairs. FIG. 6 shows an embodiment of the present invention identified as Al ₂ O ₃ from the correlation between Al and O.

The composite inclusions were similarly analyzed by analyzing the spectrum data. For example, when further statistical processing was performed in Al ₂ O ₃ classified in FIG. 6, a correlation was obtained between the element pairs of Mn and S and Ti and N shown in FIG.

A composite was identified by setting a threshold value in the data of FIG. FIG. 7 shows an embodiment of the present invention identified as MnS and TiN. From these FIGS. 6 and 7, it can be finally identified as a composite inclusion of Al ₂ O ₃ + MnS + TiN.

  As described above, according to the method of the present invention, it is possible to quickly identify a complex compound simply by repeating the analysis step by step in the order shown in FIG. 1 including single inclusions and precipitates. It becomes.

  On the other hand, the evaluation of grain boundary precipitates is performed by comparing the positional information of the identified inclusions and precipitates (see Fig. 2) with the information on the crystal grains obtained from the BSE image or AsB image (see Fig. 3). What is precipitated in the grain boundary and what is precipitated in the grain are distinguished and evaluated.

  Furthermore, using the area item of the particle shape information shown in Table 3, the particle size distribution of the intragranular precipitate and the grain boundary precipitate is evaluated as shown in FIG.

  Needless to say, the present invention is not limited to the above-described examples, and the embodiments may be appropriately changed within the scope of the technical idea described in the claims.

  For example, the invention according to claims 1 to 3 of the present application can be applied not only to the analysis of inclusions and precipitates contained in steel, but also to the analysis of inclusions and precipitates contained in electrolytic extraction residues.

Claims (4)

A method for discriminating foreign matter in a metal, characterized in that inclusions or precipitates contained in the metal material are identified by the following steps (1) to (7).
(1) An element included in the characteristic X-ray information obtained from the individual inclusions or precipitates by irradiating the individual inclusions or precipitates with an electron beam using a scanning electron microscope or an electron beam microanalyzer The process of acquiring strength.
(2) One element excluding O, S, N, and C elements and each element of O, S, N, and C included in the characteristic X-ray information obtained from the individual inclusions or precipitates The process of calculating the intensity correlation.
(3) In the intensity correlation, except for data less than a predetermined value of the one element included in the characteristic X-ray information, the correlation between the one element and each element of O, S, N, and C The process to seek.
(4) A step of giving a predetermined correlation range to the table of characteristic X-ray intensity ratios of oxides, sulfides, nitrides, and carbides.
(5) The characteristic X-ray intensity ratio of the oxide, sulfide, nitride and carbide of the one element is the correlation between the one element and each of O, S, N and C in the process of (3) Determining whether or not a predetermined correlation range is satisfied in the table.
(6) When the predetermined correlation is satisfied, an inclusion or a precipitate is identified as any one of the oxide, sulfide, nitride, carbide, or composite of the one element. Process.
(7) The calculation is performed by removing the intensity of the one element and the corresponding O, S, N, and C from the process of (1), and excluding the one element, the elements included in the characteristic X-ray information (O, Steps (2) to (6) are repeated for (except S, N, and C elements).   In the step (2), the element included in the characteristic X-ray information (the elements of O, S, N, and C are the elements included in the X-ray information obtained from the individual inclusions or precipitates are The method for discriminating foreign matter in metal according to claim 1, further comprising a step of selecting in order from the highest integrated intensity of (except). When the lower limit threshold value of the correlation coefficient is R _t , the predetermined correlation coefficient R is an element pair that satisfies R _t ≦ R ≦ 1.0 and R _t is 0.4 or less. The method for discriminating foreign matter in metal according to claim 1 or 2, wherein the foreign matter is discriminated.   The scanning electron microscope or electron beam microanalyzer further includes a backscattered electron detector or an angle selective backscattered electron detector, and the size and / or position of the inclusions or precipitates by image data of the detector and analysis of the image data. The metal foreign matter discrimination method according to claim 1, wherein the composition of the inclusions or precipitates is identified from the characteristic X-ray information.