Methods for analyzing metal-containing samples

Abstract

To accurately and efficiently analyze information on metal particles contained in a metal-containing sample.SOLUTION: An analysis method includes: a preparation step of preparing a metal-containing sample; an image acquisition step of imaging the metal-containing sample by multiple different sample amounts by varying an imaging distance with an X-ray CT device, and acquiring multiple CT images with different imaging visual fields depending on the imaging distance; an analysis step of acquiring multiple particle size distributions from each of the CT images; a normalization step of normalizing the multiple particle size distributions and acquiring multiple normalized particle size distributions; an extraction and integration step of comparing the two normalized particle size distributions with each other, setting a threshold of the particle size correlated with the number of metal particles, extracting one below the threshold from the one normalized particle size distribution and extracting one above the threshold from the other normalized particle size distribution for each of the multiple normalized particle size distributions; and an evaluation step of evaluating the metal particles contained in the metal-containing sample based on the extracted normalized particle size distributions.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a method for analyzing a metal-containing sample.

Background Art

[0002] As a metal-containing sample, for example, there are reduction products obtained when Ni oxide ore such as saprolite is subjected to reduction treatment to smelt an Fe-Ni alloy. This reduction product contains metals (metal particles) such as Fe-Ni alloy and slag (non-metallic substance), and the metals are scattered as fine particles in the slag.

[0003] When recovering fine metal particles scattered in the slag, it is desirable to aggregate and enlarge the fine metal particles scattered in the slag in order to improve smelting efficiency. Therefore, it is useful to obtain information such as the size, shape, distribution state, and existing ratio of the metal particles contained in the metal-containing sample. As a method for measuring the proportion of metal components, a method of analyzing shape parameters such as particle volume, equivalent sphere diameter, and sphericity of particles using, for example, an automatic mineral analyzer (so-called MLA) and an X-ray CT apparatus has been proposed (for example, Patent Document 1). There is also a method of measuring the content of metal components by performing chemical analysis on a metal-containing sample.

Prior Art Literature

Patent Literature

[0004]

Patent Document 1

Summary of the Invention

Problem to be Solved by the Invention

[0005] In the method described above, which combines MLA and X-ray CT equipment, the mineral species data identified by MLA is linked with the brightness data obtained from X-ray CT measurements. This allows X-ray CT measurements to provide information such as the volume, equivalent spherical diameter, and sphericity of specific mineral particles in the sample. While chemical analysis methods can measure the metal content, they cannot provide information on the size, shape, and distribution of metal particles. Therefore, the method described above, which combines MLA and X-ray CT equipment, can be applied to collect information on the size, shape, distribution, and relative abundance of metal particles in a metal-containing sample.

[0006] However, with X-ray CT measurements, if minute metal particles are present, they may not be accurately detected depending on the resolution or pixel size. Also, if there are only a few coarse metal particles, the numerical values ​​indicating the proportion of presence, such as volume ratio, will vary greatly depending on the sample amount and sampling accuracy. In other words, X-ray CT measurements have had challenges in reducing the detectable particle size and reducing variability. As a result, it has sometimes been difficult to accurately and efficiently analyze information about metal particles in metal-containing samples, such as their size, shape, quantity, and distribution.

[0007] This invention has been made in view of the above problems, and aims to provide a technology for accurately and efficiently analyzing information on metal particles contained in a metal-containing sample. [Means for solving the problem]

[0008] A first aspect of the present invention is: Preparation steps for preparing a metal-containing sample, which is composed of metal particles present in a nonmetallic substance, The image acquisition step involves using an X-ray CT scanner to image the metal-containing sample with different sample quantities by varying the imaging distance from the X-ray source to the metal-containing sample, thereby acquiring multiple CT images having different imaging fields depending on the imaging distance. An analysis step of obtaining multiple particle size distributions from each of the multiple CT images, which show the relationship between the particle size and number of metal particles contained in the metal-containing sample, A normalization step to obtain multiple normalized particle size distributions by normalizing the sample amount which changes according to the scanning distance when acquiring the CT image, An extraction and integration step is performed for each of the multiple normalized particle size distributions, which involves comparing one normalized particle size distribution obtained at a predetermined shooting distance with other normalized particle size distributions obtained at a longer shooting distance, setting a threshold for particle size that correlates with the number of metal particles, extracting normalized particle size distributions below the threshold from the one normalized particle size distribution, and extracting normalized particle size distributions above the threshold from the other normalized particle size distributions. The system includes an evaluation step of evaluating the metal particles contained in the metal-containing sample based on the extracted normalized particle size distribution. A method for analyzing metal-containing samples is provided.

[0009] A second aspect of the present invention is, in the first aspect, In the extraction and integration process, the threshold is set from a range of at least three times the resolution corresponding to the shooting distance when acquiring the other normalized particle size distribution.

[0010] A third aspect of the present invention is, in the first aspect, In the extraction and integration step, the threshold is set from a particle size range in which the ratio of the number of particles in one normalized particle size distribution to the other normalized particle size distribution is 0.5 times or more and 1.5 times or less.

[0011] A fourth aspect of the present invention is, in any of the first to third aspects, In the image acquisition process, the ratio of the sample quantities of the two samples taken at close shooting distances is set such that the sample quantity taken at a relatively long shooting distance is 3 to 100 times the sample quantity taken at a relatively short shooting distance.

[0012] A fifth aspect of the present invention is, in any of the first to third aspects, In the normalization step, when A is the amount of a sample in a CT image captured at the maximum imaging distance, and B is the amount of the sample in CT images captured at other imaging distances, the particle size distribution obtained at the other imaging distances is normalized by multiplying the number by A / B.

Effects of the Invention

[0013] According to the present invention, information of metal particles contained in a metal-containing sample can be analyzed accurately and efficiently.

Brief Description of Drawings

[0014] [Figure 1] FIG. 1 is a flowchart showing a method for analyzing a metal-containing sample according to an embodiment of the present invention. [Figure 2] FIG. 2 is a diagram for explaining an imaging field of view when a CT image is captured using an X-ray CT apparatus. [Figure 3] FIG. 3 shows a particle size distribution obtained in an analysis step according to an embodiment of the present invention. [Figure 4] FIG. 4 shows a normalized particle size distribution obtained in a normalization step according to an embodiment of the present invention. [Figure 5] FIG. 5 shows a normalized particle size distribution obtained for Sample 1. [Figure 6] FIG. 6 shows a normalized particle size distribution obtained for Sample 2.

Mode for Carrying Out the Invention

[0015] <One Embodiment of the Present Invention> Hereinafter, a method for analyzing a metal-containing sample according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a flowchart showing the method for analyzing a metal-containing sample according to an embodiment of the present invention. FIG. 2 is a diagram for explaining an imaging field of view when a CT image is captured using an X-ray CT apparatus.

[0016] (Preparation Step S1) First, as shown in FIG. 1, prepare a metal-containing sample to be evaluated.

[0017] The metal-containing sample is constituted by metal particles present (or scattered) in a non-metallic substance. The metal-containing sample is, for example, a reduction-treated product obtained by the reduction treatment of Ni oxide ore such as saprolite. The metal particles contained in the reduction-treated product are metals such as Fe-Ni alloys. Further, the non-metallic substance contained in the reduction-treated product is slag, and the compound forms thereof are, for example, Fe₂O₃, Al₂O₃, CaO, MgO, SiO₂ and the like.

[0018] The shape of the metal-containing sample may be massive or powdery, as long as it does not interfere with X-ray CT measurement. However, a powdery form is preferable in order to fill the specified sample container and obtain the specified imaging field of view. The particle size when the metal-containing sample is formed into a powder varies depending on the production process, origin, and source of the metal-containing sample, as well as the information desired to be obtained by X-ray CT measurement. For example, when the metal-containing sample is a reduction-treated product obtained by the reduction treatment of Ni oxide ore such as saprolite, and it is desired to obtain information such as the size, shape, distribution state, and existing ratio of fine metal particles scattered in slag, necessary information cannot be obtained if the sample is pulverized so finely that the size and shape of the metal particles cannot be distinguished. In such a case, a sample obtained by crushing air-cooled slag to a degree of loosening, with a maximum size of about 6 mm, is used. For example, in the case of granular slag produced by the granulated shot method, that is, a method of dropping a melt into a jet water stream, crushing or pulverization may not be necessary in some cases. When the shape of the metal-containing sample is powdery, if there are too many voids, the air layer becomes a transmission layer that may interfere with precise measurement, so it is also necessary to adjust the apparent density by tapping. In other words, the particle size of the metal-containing sample is preferably 6 mm or less.

[0019] (Image acquisition step S2) Subsequently, the above-described metal-containing sample is accommodated in a container (capsule) for X-ray CT measurement. The capsule containing the metal-containing sample is introduced into an X-ray CT apparatus.

[0020] An X-ray CT (Computed Tomography) system irradiates a sample with X-rays while rotating it 360°, and uses the difference in X-ray absorption by different materials in each part of the sample to non-destructively evaluate the internal structure of a material in three dimensions. As shown in Figure 2, the X-ray CT system 1 is composed of an X-ray source 10, a detector 20, and a stage (not shown) on which a capsule is placed. The X-ray source 10 is configured to irradiate by spreading X-rays in a cone shape, for example, using the cone beam method. The detector 20 can be, for example, a flat panel detector, and a transmission image can be obtained by detecting X-rays that penetrate the metal-containing sample. The stage is placed between the X-ray source 10 and the detector 20, is rotatable, and is configured to move between the X-ray source 10 and the detector 20. In the X-ray CT system 1, after moving the stage to a predetermined position, X-rays are irradiated onto the metal-containing sample in the capsule, and a transmission image is obtained by detecting the X-rays that penetrate the metal-containing sample. Then, the capsule is rotated and irradiated with X-rays to acquire multiple transmission images, and finally, a three-dimensional CT image is obtained from these multiple transmission images.

[0021] In this embodiment, the imaging distance from the X-ray source to the capsule (metal-containing sample) is changed, and imaging is performed with multiple sample amounts of different sizes, acquiring multiple CT images with different imaging fields depending on the imaging distance. In this embodiment, as shown in Figure 2, the case in which imaging is performed at three imaging distances d1 to d3 and three CT images P1 to P3 are acquired will be explained as an example. Note that d1 <d2<d3となる。

[0022] The CT image P1 obtained at scanning distance d1 has the shortest scanning distance and the highest magnification. Therefore, the scanning field of view (FOV) of CT image P1 is small, and the amount of sample observed is the smallest. On the other hand, because the magnification of CT image P1 is high, its minimum pixel size is small, and its resolution is the highest. In other words, a CT image obtained at a short scanning distance d1 can display a small amount of sample with high resolution. Here, FOV refers to the size of the scanning area observed at a given distance, and is shown, for example, by the diameter of the scanning area displayed as a cylinder in Figure 2. Sample amount refers to the weight or volume of a portion of the sample contained in the capsule that is imaged by X-ray transmission. Sample amount fluctuates in accordance with the change in the X-ray transmission area as the scanning distance changes.

[0023] In contrast, as the scanning distance increases to d2 and d3, the magnification decreases. Therefore, the FOV of CT images P2 and P3 increases with scanning distance, and the amount of sample observed also increases. On the other hand, the resolution decreases in proportion to the scanning distance, with the lowest resolution occurring at scanning distance d3. In other words, while increasing the scanning distance lowers the resolution, it allows for a larger amount of sample to be observed.

[0024] According to the inventors' research, CT images P1 acquired at a short scanning distance d1 have high resolution, allowing even fine metal particles contained in a metal-containing sample to be imaged. However, the particle size of metal particles in a metal-containing sample can vary, and if the amount of sample scanned is too small, it may not be possible to accurately evaluate the overall particle size of metal particles in the metal-containing sample. For example, if a single coarse metal particle is present in the CT image P1, it may be impossible to determine whether this metal particle is uniformly distributed throughout the entire metal sample or is localized, making it difficult to accurately evaluate the particle size distribution of the metal particles. To improve evaluation accuracy, it is necessary to increase the amount of sample, which requires repeatedly acquiring CT images P1 at a short scanning distance d1. In this case, the evaluation time increases, and work efficiency is significantly reduced.

[0025] On the other hand, in the case of CT image P3, which is acquired at a long scanning distance d3, for example, the sample volume is large, so it is not necessary to repeatedly acquire CT images as in the case of CT image P1, thus suppressing a decrease in work efficiency. However, the resolution of CT image P3 is low, and fine metal particles that are below the resolution level among the metal particles contained in the metal-containing sample cannot be observed. For this reason, neither CT image P1 nor CT image P3 can accurately evaluate the metal particles contained in the metal-containing sample across a wide range of particle sizes.

[0026] For example, in a CT image with a 30mm field of view (FOV), the observed sample volume is 20g, whereas in a CT image with a 7mm FOV, the observed sample volume is 0.2g. To obtain the same sample volume as with a 30mm FOV, 100 CT images would be required when using a 7mm FOV. Furthermore, with a 30mm FOV, the resolution is approximately 90μm, and metal particles smaller than 90μm may not be detectable.

[0027] Thus, with only one CT image acquired at a predetermined scanning distance, the amount of sample may be small depending on the scanning distance, significantly reducing work efficiency, or the resolution may be low, making accurate evaluation impossible.

[0028] Therefore, in this embodiment, three CT images with different imaging fields are acquired by changing the imaging distance. Subsequently, analysis step S3, normalization step S4, extraction and integration step S5, and evaluation step S6 are performed on these CT images.

[0029] In the image acquisition step S2, the multiple shooting distances should be appropriately changed according to the range of particle sizes that the metal particles contained in the metal-containing sample can take. From the viewpoint of performing the extraction and integration step S5 described later with high accuracy, it is preferable that the ratio of sample amounts of two samples taken at close shooting distances be such that the sample amount at the relatively long shooting distance is 3 to 100 times the sample amount at the relatively short shooting distance, and more preferably 3 to 25 times. For example, the sample amount at shooting distance d2 should be 3 to 100 times the sample amount at shooting distance d1, and the sample amount at shooting distance d3 should be 3 to 100 times the sample amount at shooting distance d2. From the viewpoint of performing the extraction and integration step S5 described later with higher accuracy, it is preferable that the sample amount at the longest shooting distance be 3 to 100 times the sample amount at the shortest shooting distance among the sample amounts at multiple shooting distances.

[0030] Furthermore, in the image acquisition process S2, the shortest scanning distance is not particularly limited as long as fine metal particles contained in the metal-containing sample can be detected on the CT image, and should be appropriately changed according to the size of the fine metal particles to be detected. While a smaller detection limit is generally better, it cannot be determined definitively as it depends on the performance of the equipment. Generally, the scanning distance should be set to achieve a resolution that can detect fine metal particles with a particle size of 7 μm or larger.

[0031] The magnification ratio refers to the ratio of the distance from the focal point of the X-ray source 10 to the distance from the focal point of the X-ray source 10 to the distance from the capsule (metal-containing sample). The minimum pixel size is the value obtained by dividing the pixel size originating from the detector 20 by the magnification ratio.

[0032] (Analysis process S3) Next, each of the three acquired CT images is analyzed to obtain multiple particle size distributions that show the relationship between the particle size and number of metal particles contained in the metal-containing sample.

[0033] In CT images, the amount of X-ray absorption differs depending on the material composition of the metal particles and non-metallic substances in a metal-containing sample. In a metal-containing sample, components with higher density, thickness, and atomic number have lower transmitted X-ray intensity, while components with lower density, thickness, and atomic number have higher transmitted X-ray intensity. In CT images, metal particles in a metal-containing sample are displayed with high brightness because they are denser than non-metallic substances, such as slag. Slag, being relatively low in density, is displayed with low brightness.

[0034] In this embodiment, voxels that are displayed with high brightness, for example, are extracted from the CT image, and the particle size and number of metal particles are determined to obtain the particle size distribution. In this embodiment, the particle size distribution is obtained for each of three CT images having different imaging fields.

[0035] Specifically, each particle size distribution displays the number of metal particles of particle sizes x1 to x11, as shown in Figure 3. Figure 3 shows the particle size distribution obtained in analysis step S3 in one embodiment of the present invention, where the horizontal axis represents the particle size of the metal particles and the vertical axis represents the number of metal particles. Since the sample amount differs depending on the scanning distance of the three CT images, the particle size distribution obtained from each CT image is based on information from different sample amounts. For example, the particle size distribution obtained from CT image P3, which has a large sample amount, tends to show a higher number of particles of the same size compared to CT image P1, which has a small sample amount.

[0036] (Standardization process S4) Next, the three particle size distributions are normalized by the sample quantity.

[0037] In this embodiment, in order to eliminate differences in sample amount for each particle size distribution, multiple particle size distributions are normalized by the sample amount. This results in multiple normalized particle size distributions from which differences in sample amount have been eliminated. Each normalized particle size distribution has the same sample amount due to normalization, thus eliminating differences in sample amount.

[0038] The method of normalization is not particularly limited, but when the sample amount in the CT image acquired at the maximum scanning distance is A, and the sample amount in the CT images acquired at other scanning distances is B, it is preferable to normalize the particle size distribution obtained at other scanning distances by multiplying the number of particles by A / B. In this embodiment, when the sample amount in the CT image P3 acquired at the maximum scanning distance d3 is A, the sample amount in the CT image P1 is B1, and the sample amount in the CT image P2 is B2, it is preferable to multiply the number of particles in the particle size distribution obtained from the CT image P1 by A / B1, and the number of particles in the particle size distribution obtained from the CT image P2 by A / B2. This makes it possible to correct the particle size distribution obtained from the CT images P1 and P2, which have relatively small sample amounts, so that they have the same sample amount as the CT image P3.

[0039] The normalized particle size distribution based on each CT image is shown in Figure 4, for example. Figure 4 shows the normalized particle size distribution obtained in the normalization process in one embodiment of the present invention, where the horizontal axis represents the particle size of the metal particles and the vertical axis represents the number of metal particles.

[0040] (Extraction / integration step S5) Next, we identify and extract the particle size range that can ensure accuracy in counting from the three normalized particle size distributions. In normalized particle size distributions, the resolution decreases as the scanning distance increases, so as the sample amount increases (CT images P2 and P3), the accuracy in counting tends to decrease in the range of small particle sizes. On the other hand, as the scanning distance decreases and the sample amount decreases, the variability of large particles increases, so as the sample amount decreases (CT images P2 and P1), it becomes more difficult to ensure accuracy in counting in the range of large particle sizes. Therefore, we identify and extract the particle size range that can ensure accuracy in counting from each normalized particle size distribution.

[0041] A method for determining the particle size range in which the accuracy of the number of particles can be ensured involves comparing one normalized particle size distribution obtained at a predetermined imaging distance with other normalized particle size distributions obtained at longer imaging distances, and setting a threshold particle size for which the number of metal particles correlates. In this embodiment, the normalized particle size distributions obtained at the closest imaging distances are compared, and the correlation of the number distribution at each particle size is confirmed.

[0042] Specifically, first, the normalized particle size distributions obtained from the CT image P1 with the shortest scanning distance and the CT image P2 with the closest scanning distance are compared. Then, the particle size range with small discrepancies in the number of particles is identified. For example, in Figure 4, since there is a correlation in the number of particles between CT image P1 and CT image P2 when the particle size is x6 or greater, particle size x6 is set as the threshold T1 at which the number of metal particles correlates between CT image P1 and CT image P2. In the range of particle size less than x6 (x5 or less), the normalized particle size distribution obtained from CT image P1 reflects the particle size distribution of metal particles with higher accuracy, and in the range of particle size x6 or greater, the normalized particle size distribution obtained from CT image P2 reflects the particle size distribution of metal particles with higher accuracy. Next, the same operation is performed for the normalized particle size distributions obtained from CT image P3 and the CT image P2 with the closest scanning distance. In Figure 4, a correlation is observed between the number of particles in CT image P2 and CT image P3 for particle sizes x9 and above. Therefore, particle size x9 is set as the threshold T2 for which the number of metal particles correlates between CT image P2 and CT image P3. In the range where the particle size is less than x9 (x8 or less), the normalized particle size distribution obtained from CT image P2 reflects the particle size distribution of metal particles with higher accuracy. In the range where the particle size is x9 and above, the normalized particle size distribution obtained from CT image P3 reflects the particle size distribution of metal particles with higher accuracy.

[0043] In this way, after comparing each normalized particle size distribution and setting a particle size threshold that can ensure accuracy in counting for each normalized particle size distribution, the count distribution is extracted from each normalized particle size distribution.

[0044] Furthermore, in the extraction and integration step S5, it is preferable to set the particle size threshold based on the resolution, i.e., the pixel size, corresponding to the scanning distance of the CT image. Specifically, for one normalized particle size distribution acquired at a predetermined scanning distance, and other normalized particle size distributions acquired at longer scanning distances, it is preferable to set the particle size threshold from a range of three times or more the resolution (pixel size) of the other normalized particle size distributions. For example, in the case of Figure 4, for the normalized particle size distributions acquired from CT image P1 and CT image P2, it is preferable to set the threshold from a range of three times or more the resolution (pixel size) of CT image P2. Also, for the normalized particle size distributions acquired from CT image P2 and CT image P3, it is preferable to set the threshold from a range of three times or more the resolution (pixel size) of CT image P3. According to the inventors' studies, it was found that even metal particles with a particle size greater than or equal to the resolution (pixel size) could not be detected with high accuracy. On the other hand, it was found that particles with a particle size of three times or more the resolution (pixel size) could be detected with high accuracy.

[0045] Furthermore, in the extraction and integration process S5, when setting the particle size threshold for which the number of metal particles correlates, it is preferable to set the threshold from a particle size range in which the ratio of the number of particles between one predetermined normalized particle size distribution and another normalized particle size distribution with a longer imaging distance is between 0.5 and 1.5 times. If the ratio of the number of particles is within the above range, the deviation of the normalized particle size distributions is small, and the accuracy of the number of particles can be ensured in a range above the threshold.

[0046] (Evaluation step S6) Next, the metal particles contained in the metal-containing sample are evaluated based on the extracted and integrated normalized particle size distribution. This evaluation can be performed, for example, by determining the weight ratio of metal particles in the sample based on the particle size and number of metal particles. In addition to the weight ratio, the shape and distribution of the metal particles can also be evaluated and analyzed based on the particle size threshold that correlates with the set number of metal particles.

[0047] Based on the above, it is possible to evaluate the information of metal particles contained in a metal-containing sample.

[0048] <Effects according to this embodiment> This embodiment provides one or more of the following effects.

[0049] As a comparative method, for example, when evaluating metal particles contained in a metal-containing sample based on CT images acquired at a single scanning distance, the following challenges arise.

[0050] For example, when evaluating metal particles based on CT images acquired at a short scanning distance, high resolution is achieved, making it possible to detect even fine metal particles. However, because the sample volume is small, if even one coarse metal particle is present, it may be difficult to determine whether the distribution reflects the true particle size distribution of the metal-containing sample. In other words, the accuracy of counting large metal particles may not be guaranteed. To overcome this, it is conceivable to increase the sample volume by performing evaluations multiple times at the same scanning distance, but this increases the evaluation time and reduces work efficiency.

[0051] Conversely, when obtaining particle size distribution based on CT images acquired at long scanning distances and evaluating metal particles from that particle size distribution, the large sample volume makes it possible to obtain the true particle size distribution of the metal-containing sample in the coarse particle size range. However, because the resolution of the CT images is low, it may not be possible to accurately detect fine metal particles, and therefore the accuracy of counting small metal particles may not be guaranteed.

[0052] Thus, when acquiring CT images at a single scanning distance and evaluating metal-containing samples, the resolution of the CT image and the sample volume are inversely proportional, making it difficult to accurately and quickly acquire a particle size distribution with a wide range of particle sizes.

[0053] (a) In this embodiment, first, in the image acquisition step S2 and the analysis step S3, particle size distributions are obtained from three CT images P1 to P3 taken at three different shooting distances (d1 to d3). Subsequently, since these particle size distributions are information based on the sample amount corresponding to the shooting distance, each particle size distribution is normalized by the sample amount in the normalization step S4. This obtains normalized particle size distributions in which the sample amount is normalized to be the same. Each normalized particle size distribution has a different particle size range in which accuracy in the number of metal particles can be ensured due to the different shooting distances. Therefore, in the extraction and integration step S5, the three normalized particle size distributions are compared with the normalized particle size distributions with the closest shooting distances, and a threshold particle size for which the number of metal particles correlates is set. For example, as shown in Figure 4, the normalized particle size distribution at shooting distance d1 and the normalized particle size distribution at shooting distance d2 are compared to identify the particle size range for which the number of metal particles correlates, and the threshold T1 is set. Similarly, the normalized particle size distribution at imaging distance d2 and the normalized particle size distribution at imaging distance d3 are compared to set a particle size threshold T2. Next, from each normalized particle size distribution, the particle size range that ensures accuracy in the number of metal particles is extracted. Specifically, from the normalized particle size distribution at imaging distance d1, the range between threshold T1 and threshold T2 is extracted from the normalized particle size distribution at imaging distance d2, and the range greater than or equal to threshold T2 is extracted from the normalized particle size distribution at imaging distance d3. By extracting from each normalized particle size distribution in this way, the accuracy of the number of metal particles at each particle size can be maintained with high precision, so that in evaluation step S6, fine to coarse metal particles contained in the metal-containing sample can be evaluated with high precision. Moreover, even though CT images are acquired three times, the largest amount of information can be obtained from the sample corresponding to the longest imaging distance d3. In other words, it is possible to increase the sample amount for accurate evaluation while simultaneously shortening the evaluation time.

[0054] (b) In the extraction and integration step S5, it is preferable to compare one normalized particle size distribution acquired at a predetermined imaging distance with another normalized particle size distribution acquired at a longer and closest imaging distance, and to set the particle size threshold from a range of three times or more the resolution of the other normalized particle size distribution, i.e., the pixel size. In this embodiment, it is preferable to select the particle size threshold T1 for the normalized particle size distributions acquired at imaging distances d1 and d2 from a range of three times or more the resolution (pixel size) of the CT image P2 at imaging distance d2. Similarly, it is preferable to select the particle size threshold T2 for the normalized particle size distributions acquired at imaging distances d2 and d3 from a range of three times or more the resolution (pixel size) of the CT image P3 at imaging distance d3. This makes it possible to acquire information about the number of metal particles from each normalized particle size distribution with higher accuracy.

[0055] (c) In addition, in the extraction and integration step S5, it is preferable to set a threshold from the particle size range in which the ratio of the number of metal particles is between 0.5 and 1.5 times when comparing one normalized particle size distribution with other normalized particle size distributions. If the ratio of the number of particles is within the above range, the correlation of the number of particles between one normalized particle size distribution and other normalized particle size distributions will be higher, making it possible to obtain information about the number of metal particles with higher accuracy.

[0056] (d) In addition, in the image acquisition step S2, it is preferable that the ratio of the sample amount at a relatively long shooting distance to the sample amount at a relatively short shooting distance be 3 to 100 times. This allows for more reliable extraction of particle size ranges with guaranteed accuracy in number from each normalized particle size distribution in the extraction and integration step S5, and enables more accurate evaluation of metal particles in the evaluation step S6.

[0057] (e) In the normalization step S4, when the sample amount in the CT image acquired at the maximum scanning distance is A, and the sample amount in the CT image acquired at other scanning distances is B, it is preferable to normalize the particle size distribution obtained at other scanning distances by multiplying the number of particles by A / B. This allows for more accurate evaluation of metal particles in the evaluation step S6.

[0058] <Other Embodiments> Although embodiments of the present invention have been described above, the present invention is not limited in any way to the embodiments described above, and can be modified in various ways without departing from the spirit of the invention.

[0059] In the above embodiment, the case in which CT images are acquired at three different scanning distances was described. However, it is also possible to acquire CT images at two different scanning distances and perform the analysis steps S3 to evaluation steps S6 in the same manner as in the above embodiment. In this case, it is preferable to set the maximum scanning distance to 3 times or more and 100 times or less the minimum scanning distance. Note that four or more different scanning distances may be selected. There is no particular upper limit, but since the evaluation time increases as the scanning distance increases, it is best to set it appropriately within a range that does not cause a significant decrease in work efficiency. [Examples]

[0060] The present invention will be described below based on more detailed examples, but the present invention is not limited to these examples.

[0061] (Sample 1) In Sample 1, a crushed material was prepared as a metal-containing sample by crushing the reduction product obtained from the reduction treatment of Ni oxide ore (saprolite). This crushed material mainly consists of slag containing metal oxides, with the metal present in particulate form. This crushed material was crushed so that it could pass through a 2 mm sieve.

[0062] Next, the fragmented material was placed in a container (capsule) for X-ray CT measurement and introduced into an X-ray CT scanner (Rigaku Corporation's "CTLabHX130"). In Sample 1, the imaging distance between the X-ray source and the metal-containing sample was varied to obtain three CT images with different field-of-view (FOV) sizes of 7 mm, 20 mm, and 30 mm. Hereafter, an FOV of 7 mm will be referred to as "FOV7". The sample amounts were 0.2 g for FOV7, 5 g for FOV20, and 20 g for FOV30. The resolution, i.e., pixel size, of each CT image was 7.1 μm for FOV7, 20 μm for FOV20, and 30 μm for FOV30.

[0063] Next, each acquired CT image was analyzed to measure the particle size and number of metal particles contained in the metal-containing sample, thereby obtaining the particle size distribution. Subsequently, each particle size distribution was normalized to match the sample amount with an FOV of 30. Specifically, the number of particles in the FOV 7 particle size distribution was multiplied by 100 (20 / 0.2), and the number of particles in the FOV 20 particle size distribution was multiplied by 4 (20 / 5). This resulted in obtaining three normalized particle size distributions. The obtained normalized particle size distributions are shown in Figure 5. Figure 5 shows the normalized particle size distribution obtained in Sample 1, with the horizontal axis representing particle size [μm] and the vertical axis representing the logarithm of the number of particles.

[0064] Next, we compared each normalized particle size distribution and set a threshold for particle size at which the number of metal particles correlates. When comparing the normalized particle size distributions of FOV7 and FOV20, a high correlation in particle size was observed in the range of 63 μm or larger, so the threshold for FOV7 and FOV20 was set to 63 μm. Similarly, when comparing the normalized particle size distributions of FOV20 and FOV30, a high correlation in particle size was observed in the range of 90 μm or larger, so the threshold for FOV20 and FOV30 was set to 90 μm. The particle size threshold was set from a range of particle size that is three times or more the resolution at each FOV, i.e., the pixel size, and where the ratio of the number of particles in the normalized particle size distribution being compared is between 0.5 and 1.5 times.

[0065] Next, particle size distributions were extracted based on the particle size thresholds set for each normalized particle size distribution. Specifically, for the FOV7 normalized particle size distribution, the range of particle size from 8.1 μm to 53 μm was extracted; for the FOV20 normalized particle size distribution, the range of particle size from 63 μm to 75 μm was extracted; and for the FOV30 normalized particle size distribution, the range of particle size of 90 μm or larger was extracted, and these were then combined into one.

[0066] Next, the metal particles contained in the metal-containing sample were evaluated based on the extracted and integrated normalized particle size distribution. In Sample 1, the content (weight ratio) of metal particles in the metal-containing sample was analyzed as information on metal particles. Specifically, the weight ratio of metal in the metal-containing sample was evaluated based on the particle size and number of metal particles. As a result, it was confirmed that the metal content in the metal-containing sample was 10.7% by weight.

[0067] (Sample 2) In Sample 2, the same procedure as in Sample 1 was followed to evaluate the metal-containing sample, except that CT images were acquired by imaging at two points with FOV 7 and FOV 30. The normalized particle size distribution obtained in Sample 2 is shown in Figure 6. Figure 6 shows the normalized particle size distribution obtained in Sample 2, with the horizontal axis representing particle size [μm] and the vertical axis representing the logarithm of the number of particles. In Sample 2, the normalized particle size distributions of FOV 7 and FOV 30 were compared, and since there was a high correlation in the number of particles in the range of particle size 90 μm or larger, the threshold for FOV 7 and FOV 30 was set to 90 μm. Based on the extracted normalized particle size distribution, the metal sample was evaluated, and it was confirmed that the metal content in the metal sample in Sample 2 was 11.1% by weight.

[0068] (Sample 3) In Sample 3, the particle size distribution was obtained from CT images acquired at a single scanning distance to achieve an FOV of 7. The metal sample was evaluated using the same procedure as in Sample 1, except that normalization was omitted. As a result, it was confirmed that the metal content in the metal-containing sample in Sample 3 was 22.0% by weight.

[0069] (Sample 4) In Sample 4, the particle size distribution was obtained from CT images acquired at a single scanning distance to achieve an FOV of 30. The same procedure as in Sample 1 was followed to evaluate the metal sample, except that normalization was omitted. As a result, it was confirmed that the metal content in the metal-containing sample in Sample 4 was 7.5% by weight.

[0070] (Evaluation results) To compare the evaluation results of samples 1-4, the metal content of the metal-containing samples was measured by chemical analysis, and it was confirmed that the metal content was 15.4% by weight. From this, it was confirmed that samples 1 and 2 could evaluate metal samples with an accuracy close to that of chemical analysis, while also reducing the evaluation time, compared to samples 3 and 4.

[0071] In Sample 3, which was evaluated based solely on CT images at FOV7, a higher metal content was confirmed compared to the chemical analysis. This is presumed to be because, due to the short scanning distance and small sample volume in the scanning field, coarse metal particles were present, and the resulting particle size distribution did not accurately reflect the original particle size distribution of the metal sample.

[0072] Furthermore, in Sample 4, which was evaluated based solely on FOV30 CT images, it was confirmed that the metal content was lower than that determined by chemical analysis. This is presumed to be because FOV30 CT images have low resolution and cannot accurately detect the number of minute metal particles.

[0073] On the other hand, in samples 1 and 2, which were extracted and integrated from multiple CT images for evaluation, the detectable particle size was reduced, and the variability in analysis due to coarse metal particles was suppressed. Therefore, it is presumed that the metal samples were evaluated more accurately compared to samples 3 and 4. According to the present invention's method for analyzing metal-containing samples, the particle size distribution of metal particles scattered in non-metallic substances can be obtained accurately and efficiently. Furthermore, it was found that the weight ratio of metal particles can also be rapidly evaluated, albeit as a guideline. Incidentally, in general chemical analysis, all metal elements are quantified regardless of whether they are in compound form or elemental form. Therefore, quantifying only those in the form of elemental metals requires special analytical methods, which are time-consuming and laborious.

[0074] As explained above, by acquiring multiple CT images of a metal-containing sample at different imaging distances, obtaining the particle size distribution from the multiple CT images, normalizing it, and appropriately extracting a predetermined particle size range based on the normalized particle size distribution, it is possible to evaluate the metal particles contained in the metal-containing sample with high accuracy while shortening the measurement time. [Explanation of Symbols]

[0075] 1 X-ray CT device 10 X-ray sources 20 detectors

Claims

1. Preparation steps for preparing a metal-containing sample, which is composed of metal particles present in a nonmetallic substance, Image acquisition step: Using an X-ray CT apparatus, image is taken of the metal-containing sample with different sample amounts by changing the imaging distance from the X-ray source to the metal-containing sample, and multiple CT images with different imaging fields according to the imaging distance are obtained. An analysis step of obtaining multiple particle size distributions from each of the multiple CT images, which show the relationship between the particle size and number of metal particles contained in the metal-containing sample, A normalization step to obtain multiple normalized particle size distributions by normalizing the sample amount which changes according to the imaging distance when acquiring the CT image, An extraction and integration step is performed for each of the multiple normalized particle size distributions, which involves comparing one normalized particle size distribution obtained at a predetermined shooting distance with other normalized particle size distributions obtained at a longer shooting distance, setting a threshold for particle size that correlates with the number of metal particles, extracting normalized particle size distributions below the threshold from the one normalized particle size distribution, and extracting normalized particle size distributions above the threshold from the other normalized particle size distributions. The system includes an evaluation step of evaluating the metal particles contained in the metal-containing sample based on the extracted normalized particle size distribution. Methods for analyzing metal-containing samples.

2. In the extraction and integration process, the threshold is set from a range of three times or more the resolution corresponding to the shooting distance when acquiring the other normalized particle size distribution. The method for analyzing a metal-containing sample according to claim 1.

3. In the extraction and integration step, the threshold is set from a particle size range in which the ratio of the number of particles in one normalized particle size distribution to the other normalized particle size distribution is 0.5 times or more and 1.5 times or less. The method for analyzing a metal-containing sample according to claim 1.

4. In the image acquisition step, the ratio of the sample quantities of the two samples taken at close shooting distances is set such that the sample quantity taken at a relatively long shooting distance is 3 to 100 times the sample quantity taken at a relatively short shooting distance. A method for analyzing a metal-containing sample according to any one of claims 1 to 3.

5. In the standardization step, when the sample quantity in the CT image acquired at the maximum imaging distance is A, and the sample quantity in the CT image acquired at other imaging distances is B, the particle size distribution obtained at the other imaging distances is standardized by multiplying the number by A / B. A method for analyzing a metal-containing sample according to any one of claims 1 to 3.

Images