MLA data correction method and mineral analysis method

Abstract

To provide a correction method of MLA data which can more accurately analyze a composition ratio of minerals present in ore using MLA and a mineral analysis method.SOLUTION: A correction method of MLA data includes: a process S2 of obtaining a reflection electron image of mineral particles on a measurement surface obtained by measuring the measurement surface of an ore sample in which mineral particles are embedded in resin with a mineral particle analysis device and a mineral map containing information on a mineral region and mineral species of the constituent minerals constituting the mineral particles; and a process S3 which corrects the information on the mineral region and mineral species of a target mineral by using brightness correction information of the reflection electron image of the constituent minerals for the mineral particles containing the target mineral being the correction object as the constituent minerals.SELECTED DRAWING: Figure 1

Description

Technical Field

[0005] , ,

[0001] The present invention relates to a method for correcting MLA data and a method for mineral analysis.

Background Art

[0002] A method for analyzing minerals using a mineral particle analyzer (MLA) is known. For example, in Japanese Patent Application Laid-Open No. 2020-153738 (Patent Document 1), the magnitude relationship of the density of each mineral in a sample obtained based on the measurement results by MLA for a sample containing minerals is associated with the magnitude relationship of the luminance in the X-ray CT image of the sample, and based on the relationship obtained by the association, particle data of a predetermined mineral corresponding to a predetermined luminance in the X-ray CT image is obtained, and a predetermined mineral is evaluated based on the particle data.

[0003] Japanese Patent Application Laid-Open No. 2016-156826 (Patent Document 2) describes a method for identifying substances in a sample, including steps of obtaining a spectrum for a selected point on the sample, performing a first analysis of the spectrum for each point to identify substances in the pure phase, forming a sample-specific library of substances identified as being in the pure phase at any of the selected points on the sample, and performing a second analysis of the spectrum obtained at the selected point using the sample-specific library to determine the composition of the sample at the selected point.

[0004] Japanese Patent Application Laid-Open No. *************** (Patent Document 3) describes an example of a method for performing correction to remove a portion of a mineral region of a mineral particle where the BSE luminance is below a threshold value in a mineral particle analyzer based on a scanning electron microscope having an energy dispersive X-ray analyzer called MLA or QEMSCAN in order to accurately analyze the abundance ratio of mineral species in an ore.

Prior Art Documents

Patent Documents

[0005] Please note that the number in the patent document number in the translation of item is replaced with asterisks as the original number is incomplete. If you can provide the complete number, it can be accurately translated. [Patent Document 1] Japanese Patent Publication No. 2020-153738 [Patent Document 2] Japanese Patent Publication No. 2016-156826 [Patent Document 3] Japanese Patent Publication No. 2015-40724 [Overview of the project] [Problems that the invention aims to solve]

[0006] Various measurement and correction methods have been attempted, as described in Patent Documents 1 to 3. Analysis using MLA allows for obtaining various information such as the types and grain sizes of minerals contained in the ore using analysis software attached to the MLA. However, MLA uses a scanning electron microscope for analysis, and the depth of field of a scanning electron microscope is relatively deep. Therefore, misrecognition may occur due to overlapping minerals in the depth direction of the sample or the presence of boundaries between minerals, which can lead to low accuracy in analyzing the composition ratio of minerals present in the ore.

[0007] In view of the above issues, this disclosure provides a method for correcting MLA data and a mineral analysis method that enable more accurate analysis of the composition ratio of minerals present in ore using MLA. [Means for solving the problem]

[0008] To solve the above problems, according to one aspect of this disclosure, a method for correcting MLA data is provided, which includes the step of obtaining a backscattered electron image of the mineral particles on the measurement surface of an ore sample in which mineral particles are embedded in resin by measuring the measurement surface with a mineral particle analyzer, and a mineral map including information on the mineral regions and mineral species of the constituent minerals constituting the mineral particles, and correcting the information on the mineral regions and mineral species of the mineral of interest to be corrected for mineral particles that contain the mineral of interest as a constituent mineral using brightness correction information of the backscattered electron image of the constituent minerals.

[0009] According to another aspect of this disclosure, a mineral analysis method using a mineral particle analyzer is provided, comprising the steps of: measuring a measurement surface of an ore sample in which mineral particles are embedded in resin with the mineral particle analyzer; obtaining a backscattered electron image of the mineral particles on the measurement surface and a mineral map including information on the mineral regions and mineral species of the constituent minerals constituting the mineral particles; correcting the mineral regions and mineral species of the mineral of interest to be corrected for mineral particles containing the mineral of interest as a constituent mineral using brightness correction information of the backscattered electron image of the constituent minerals; and analyzing the composition ratio of constituent minerals present in the ore sample based on the area of ​​the mineral region of the corrected mineral particles and the types of constituent minerals. [Effects of the Invention]

[0010] According to this disclosure, a method for correcting MLA data and a method for analyzing minerals are provided that enable more accurate analysis of the mineral composition ratio present in ore using MLA. [Brief explanation of the drawing]

[0011] [Figure 1] This flowchart shows an example of a method for correcting MLA and analyzing ore according to an embodiment of the present invention. [Figure 2] This is an explanatory diagram showing an example of a mineral map of mineral particles obtained by MLA analysis of secondary enriched copper minerals. [Figure 3] Figure 3(a) is a bar graph showing an example of the results of analyzing the mineral composition ratios obtained from mineral maps obtained by MLA analysis for ore samples using secondary enriched copper minerals, and Figure 3(b) is a bar graph showing an example of the results of analyzing the mineral composition ratios using an optical microscope. [Figure 4] Figure 1 is a flowchart showing an example of a correction method in the MLA data correction process. [Figure 5] Figure 5(a) shows the mineral map of the mineral particles before correction, Figure 5(b) shows the backscattered electron image, and Figure 5(c) is an explanatory diagram showing the mineral map of the mineral particles after correction. [Figure 6]FIG. 6(a) is a bar graph showing the analysis results of the mineral composition ratios of Samples 1 to 6 after the correction process of MLA data, and FIG. 6(b) is a bar graph showing the analysis results of the measured values analyzed using an optical microscope. [Figure 7] FIG. 7(a) represents a schematic diagram of single mineral particles, and FIG. 7(b) is a graph explaining the correlation between the backscattering coefficient of the main constituent minerals and the measured values of the luminance of the reflected electron image. [Figure 8] It is an explanatory diagram showing an example of a method for calculating the first luminance correction information. [Figure 9] It is an example of the first luminance correction information, and is a histogram schematically showing the luminance ranges of each mineral after correction when the target mineral is Bn and the constituent minerals existing around the target mineral are Ccp and Clc. [Figure 10] FIG. 10(a) is a mineral map of mineral particles before correction showing a state where ores of Clc, Ccp, and Cv are adjacent to the periphery of Bn in the mineral map, FIG. 10(b) is a reflected electron image before correction, and FIG. 10(c) shows a mineral map after correction. [Figure 11] It is a graph showing the results of analyzing the composition ratio of minerals using the MLA data correction method according to the embodiment of the present invention with ores when the concentration of copper-insoluble substances in copper concentrate collected from a copper mine is high, low, and when the copper grade is lower than normal as ore samples. [Figure 12] It is a graph showing the difference in the composition analysis results due to the difference in the threshold value (0.1 to 0.9) of the area ratio with respect to the mineral particles of the target mineral. [Figure 13] It is a graph obtained by magnifying the analysis results of bornite ore in FIG. 11.

Embodiments for Carrying Out the Invention

[0012] Hereinafter, embodiments of the present invention will be described with reference to the drawings. The embodiments shown below illustrate devices and methods for embodying the technical idea of this invention, and the technical idea of this invention does not specify the structure, arrangement of the device, or the order of the method, etc. as the following.

[0013] The method for correcting MLA data according to an embodiment of the present invention includes, as shown in FIG. 1, a step of measuring a measurement surface of an ore sample in which mineral particles are embedded in a resin with a mineral particle analyzer (MLA measurement step S1), and a reflected electron image of the mineral particles present on the measurement surface, and a step of obtaining MLA data including a mineral map containing information on the mineral region and mineral species of the constituent minerals constituting the mineral particles (MLA data acquisition step S2), and a step of correcting, for the mineral particles containing the target mineral to be corrected as the constituent minerals, the information on the mineral region and mineral species of the target mineral based on the luminance correction information of the reflected electron image of the constituent minerals present around the constituent minerals or the target mineral (MLA data correction step S3). The mineral analysis method according to an embodiment of the present invention further includes a step of analyzing the composition ratio of the constituent minerals present in the ore sample based on the area of the mineral region of the corrected mineral particles and the types of the constituent minerals obtained in the MLA data correction step S3 (mineral analysis step S4).

[0014] In the MLA measurement step S1, an ore sample is prepared by embedding the mineral particles to be analyzed in a resin and polishing the surface of the measurement surface. The ore to be analyzed is not particularly limited. For example, ores containing chalcocite (Cu2S: hereinafter also referred to as "Chalcocite" or "Clc"), chalcopyrite (CuFeS2: hereinafter also referred to as "Chalcopyrite" or "Ccp"), pyrite (FeS2: hereinafter also referred to as "Pyrite" or "Py"), covellite (CuS: hereinafter also referred to as "Covellite" or "Cv"), bornite ore (Cu5FeS4: hereinafter also referred to as "Bornite" or "Bn"), molybdenite (hereinafter also referred to as "Molybdenite" or "Mo"), etc. are preferred.

[0015] In this embodiment, in particular, the composition ratio of the ore in the secondary enrichment zone (hereinafter also referred to as "secondary enrichment copper mineral") as the analysis target can be accurately analyzed using MLA.

[0016] In preparing the ore sample, it is preferable to embed the mineral particles in resin so that they are uniformly dispersed within the measurement surface, thereby preventing the mineral particles near the measurement surface from overlapping in the depth direction. The prepared ore sample is then introduced into the MLA equipped with an EDS-equipped scanning electron microscope (SEM-EDS). Various adjustments and measurement conditions are then set for the SEM-EDS, and the ore sample is measured using SEM-EDS to obtain a backscattered electron image (BSE image) of the measurement surface.

[0017] In the MLA data acquisition process S2, based on the SEM-EDS measurement results of the measurement surface of the ore sample, MLA data is obtained that includes at least a backscattered electron image of the mineral particles present on the measurement surface and a mineral map containing information on the mineral regions and mineral species of the constituent minerals that make up the mineral particles.

[0018] This MLA data can be obtained, for example, using general-purpose or dedicated analysis software included with the MLA. The analysis software analyzes the shape and area of ​​thousands to tens of thousands of mineral particles present on the measurement surface of the ore sample from the brightness information of the backscattered electron image of the measurement surface obtained by SEM-EDS measurement. It also analyzes the type of one or more constituent minerals (mineral species) that make up each mineral particle, and the location and area of ​​the region where those constituent minerals exist (mineral region). The software then creates a "mineral map" that includes image information such as that shown in Figure 2, representing the distribution of mineral regions and mineral species with different classifications (different colors or shades of color) for each mineral species.

[0019] The brightness of the backscattered electron image is relatively low for the resin portion and high for the mineral particle portion. Therefore, the analysis software can determine the shape and area of ​​the mineral particles by utilizing the difference in brightness of the backscattered electron image. In the MLA data acquisition process S2, analysis is performed using predetermined analysis software to acquire information on the backscattered electron image of the region corresponding to the mineral map for each of the thousands to tens of thousands of mineral particles present on the measurement surface of the ore sample.

[0020] In the MLA data correction process S3, the mineral of interest to be corrected is extracted from the mineral map of the mineral particles obtained in the MLA data acquisition process S2, and correction is performed on the mineral of interest in the mineral map based on the backscattered electron images of the constituent minerals that make up the mineral particles. Figure 2 shows an example of the mineral map of 10 mineral particles obtained from MLA analysis of secondary enriched copper minerals. From its formation reaction, it can be seen that secondary enriched copper minerals contain a large amount of Clc and Ccp as constituent minerals. Bn appears frequently at the intermineral boundary between Clc (or Cv) / Ccp in the 10 single-edged particles in Figure 2.

[0021] Figure 3(a) shows the results of the analysis of the mineral composition ratio (MLA analysis results) obtained from the mineral map acquired in the MLA data acquisition process S2 for the same ore sample using secondary enriched copper minerals, and Figure 3(b) shows the results of image analysis using the colors of the color images captured by the optical microscope.

[0022] As shown in Figures 3(a) and 3(b), there is a significant discrepancy between the MLA analysis results and the results of image analysis obtained with an optical microscope, particularly in the composition ratio of Bn (Bornite), which is represented in black in the graph. This Bn is a constituent mineral that exists near the boundary between adjacent minerals (Clc (or Cv) / Ccp), as shown in Figure 2. MLA analysis tends to misjudge the content of mineral species that exist near the boundary between the mineral species with the relatively highest content and adjacent mineral species, and this is considered to be one of the causes of the reduced accuracy of MLA measurements.

[0023] Therefore, in the MLA data correction step S3, it is preferable to set the mineral of interest to be corrected as the mineral of interest that exists at the boundary between the constituent mineral with the relatively largest area among the constituent minerals that make up the mineral particle and the constituent mineral adjacent to this constituent mineral. Typically, it is preferable to set "Bn" that appears at the Clc / Ccp boundary as the mineral of interest. This allows for more accurate correction of the MLA data, and makes it possible to analyze the composition ratio of constituent minerals present in the ore with greater accuracy using MLA. In particular, mineral particles containing Clc and Ccp as constituent minerals are frequently found in the secondary enrichment zone of copper deposits, so correcting Bn has a significant effect on improving the accuracy of MLA analysis in copper deposit applications. Note that when secondary enriched copper minerals are used as a sample, setting "Ccp" that appears at the Clc / Py boundary as the mineral of interest also allows for more accurate analysis of the composition ratio of minerals present in the ore, similar to Bn.

[0024] In the MLA data correction process S3 shown in Figure 1, first, a mineral of interest to be corrected is selected from among multiple mineral particles, and information on mineral particles containing this mineral of interest as a constituent mineral is extracted from the MLA data. Then, for the mineral particles containing the mineral of interest as a constituent mineral, the information of the mineral region of the mineral of interest is corrected using the brightness correction information of the backscattered electron image of the constituent mineral of the constituent particle containing that mineral of interest.

[0025] More specifically, this correction process can be carried out using a computer or the like equipped with a processor (processing device) that converts, calculates, or processes data based on a set procedure, and predetermined storage means, as shown in Figure 4.

[0026] In step S31 of Figure 4, first, the following information is obtained from the MLA data, including the mineral map and backscattered electron image, obtained in the MLA data acquisition process S2 of Figure 1. (a) Mineral proportion (area ratio) of constituent minerals of each mineral particle in the mineral map (b) Types of constituent minerals surrounding the mineral of interest in the mineral map (adjacent mineral species) (c) Brightness distribution of the backscattered electron image of each mineral particle and the type of constituent mineral corresponding to each peak shown in the brightness distribution

[0027] In step S32, the operator sets the mineral of interest to be corrected. Here, we will explain using the example of setting "Bn" as the mineral of interest, but of course, the mineral of interest is not limited to this.

[0028] In step S33, it is determined whether the area ratio of the mineral of interest to the mineral particles is below a threshold. The threshold can be set as appropriate by the operator. Here, we will explain using an example where the threshold is set to 50%. The appropriate range for setting the threshold will be discussed later.

[0029] If the area ratio of the mineral of interest to the mineral particles in the mineral map is less than the threshold (50%), that is, if the abundance of the mineral of interest is small relative to the total mineral particles, proceed to step S34 in Figure 4. If the area ratio of the mineral of interest to the mineral particles is greater than or equal to the threshold (50%), that is, if the abundance of the mineral of interest is large relative to the total mineral particles, proceed to step S35.

[0030] In step S33, if the area ratio of the mineral of interest to the mineral particles is less than a threshold, the processing device corrects the mineral region of the mineral of interest using first brightness correction information prepared using brightness information of backscattered electron images of mineral species present around the mineral of interest. Note that mineral species present around the mineral of interest refer to mineral species in contact with the mineral of interest.

[0031] Specifically, the processing device first extracts mineral particles that contain the mineral of interest to be corrected as a constituent mineral. Mineral particles that do not contain the mineral of interest as a constituent mineral do not need to be corrected. For the extracted mineral particles, the processing device extracts the region containing the mineral of interest shown on the mineral map, for example as shown in Figure 5(a), for each unit crystal grain (hereinafter referred to as "grain") g1 to g3, and performs the correction process for each grain g1 to g3.

[0032] Figure 5(a) shows an example of a mineral map of mineral particles before correction, Figure 5(b) shows an example of a backscattered electron image, and Figure 5(c) shows an example of a mineral map of mineral particles after correction. For example, in grains g1 to g3 shown in the mineral map of Figure 5(a), the constituent minerals surrounding the mineral of interest, Bn, are Clc and Ccp. Therefore, in the correction process of the mineral map of Figure 5(a), the mineral map is corrected using first luminance correction information composed of luminance information from the backscattered electron images of the two constituent minerals "Clc" and "Ccp" that surround Bn. The method for creating the first luminance correction information will be described later.

[0033] On the other hand, in step S33 of Figure 4, if the area ratio of the mineral of interest to the mineral particles is above a threshold, that is, if the proportion of the mineral of interest is relatively high in one mineral particle, the process proceeds to step S35, where the mineral region of the mineral of interest is corrected using second brightness correction information, which is created using brightness information of the backscattered electron images of the mineral species constituting the mineral particles.

[0034] Specifically, first, the processing device extracts mineral particles containing the mineral of interest to be corrected as a constituent mineral, similar to step S34. Although a diagrammatic explanation is omitted here, the processing device extracts information on the mineral particles containing the mineral of interest from the extracted mineral particles and performs correction for each mineral particle. If the area ratio of the mineral of interest to the mineral particles is greater than or equal to a threshold, the possibility of the mineral of interest being present as a main constituent mineral in the mineral map cannot be ruled out. Therefore, in step S35, the information on the mineral region and the backscattered electron image of the mineral of interest on the mineral map are corrected based on second luminance correction information created using the luminance information of the backscattered electron images of the mineral species constituting the mineral particles. An example of the method for creating the second luminance correction information will be described later.

[0035] Figure 6(a) shows the analysis results of the mineral composition ratios of samples 1 to 6 after correcting the MLA data according to the procedure described above, and Figure 6(b) shows an example of the results of image analysis using the colors of the color images captured by an optical microscope (measured values). In the example shown in Figure 6(b), it can be seen that the composition ratio of Bn, which is thought to be incorrectly determined in terms of mineral content, has been significantly reduced in all samples compared to the mineral composition ratio before correction in Figure 3(a). Thus, it can be seen that by performing the correction process according to this embodiment on the MLA data, composition analysis results that are more consistent with measured values ​​can be obtained.

[0036] (Method for creating brightness correction information) The method for creating luminance correction information includes the steps of (1) extracting information to be used for correction, (2) determining the initial values ​​of the peak luminance value and mineral boundary value of the backscattered electron image of each constituent mineral, and (3) creating first and second luminance correction information. Furthermore, the step of creating first and second luminance correction information includes determining the combination of constituent minerals that constitute the luminance correction information based on the area ratio of the mineral of interest to the mineral particles, extracting adjacent constituent minerals when the peak luminance values ​​of the backscattered electron images of the determined constituent minerals are arranged in descending order, and creating luminance correction information in which the intermediate value between the initial lower limit of luminance of the backscattered electron image of the constituent mineral with the higher peak luminance value and the initial upper limit of luminance of the backscattered electron image of the constituent mineral with the lower peak luminance value is set as the corrected mineral boundary value between adjacent constituent minerals. Specifically, for example, it can be created according to the following procedure.

[0037] (1) Extraction of information to be used for correction First, from the set of mineral map images and backscattered electron images for each mineral particle output from the MLA, the following information is extracted as needed. (a) Mineral proportion (area ratio) of constituent minerals contained in each mineral particle of the mineral map (b) Types of constituent minerals (adjacent mineral species) surrounding the mineral of interest on the mineral map. (c) Brightness distribution information of backscattered electron images of the constituent minerals of each mineral particle

[0038] (2) Determination of the initial values ​​of the peak brightness and mineral boundary values ​​of the backscattered electron images of each constituent mineral. From MLA data containing information on numerous mineral particles, one or more mineral particles composed of only one of the main constituent minerals (Clc, Bn, Cv, Ccp, Py), as shown in Figure 7(a), are extracted, and information on their backscattered electron images is extracted. Specifically, the brightness distribution information of the backscattered electron images of the single mineral particles (for example, a histogram with brightness on the horizontal axis and number of pixels on the vertical axis) is used to determine the brightness value at the peak position, which is then used as the initial value for the peak brightness of the backscattered electron image of each constituent mineral.

[0039] As shown in Figure 7(b), there is a correlation between the backscattering intensity (backscattering coefficient) of each constituent mineral and the brightness of the backscattered electron image. Therefore, if brightness distribution information for the backscattered electron image of a single mineral particle of a desired constituent mineral cannot be obtained from the MLA data, the initial value of the peak brightness of the backscattered electron image of the single mineral particle of the desired constituent mineral can be determined by interpolation based on the relationship in Figure 7(b) using the initial value of the peak brightness of the backscattered electron image of other adjacent constituent minerals.

[0040] For example, suppose the peak brightness values ​​of the backscattering electron images for individual mineral particles of Clc, Bn, Ccp, and Py have been determined, but the peak brightness value of the backscattering electron image for individual mineral particles of Cv has not been obtained. In such a case, one method of interpolating the peak brightness value of the backscattering electron image of Cv is to determine the initial value of the peak brightness value of Cv based on the relationship between the peak brightness value of the backscattering electron image of individual mineral particles and the measured ratio of the backscattering electron brightness value, as shown in Table 1. In Table 1, "coefficient" refers to the ratio of backscattering coefficients η calculated based on an empirical formula that depends on the atomic number of the atoms constituting the constituent mineral, and "measured ratio" refers to the ratio obtained from the peak brightness values ​​of the backscattering electron images of individual mineral particles. The coefficient η was calculated based on the empirical formula described in the following document: Hong Zhao and David Darwin, "QUANTITATIVE BACKSCATTERED ELECTRON ANALYSIS TECHNIQUES FOR CEMENT-BASED MATERIALS," The University of Kansas, Lawrence, Kansas, June 1990, pp. 5-6). By interpolating the relationship between the initial and measured peak brightness values ​​of these constituent minerals, the estimated initial value of the peak brightness value of Cv is 123.

[0041] [Table 1]

[0042] As shown in Figure 7(b), the luminance values ​​of the backscattered electron images of each constituent mineral in a typical copper ore are, in descending order, Clc, Bn, Cv, Ccp, and Py. When the peak luminance values ​​of the backscattered electron images are arranged in descending order in this order, the midpoint between the backscattered electron images of adjacent constituent minerals is defined as the "mineral boundary value," and the mineral boundary value set based on the initial peak luminance values ​​of the backscattered electron images of each mineral particle in Table 1 is defined as the "initial mineral boundary value." For example, if the initial peak luminance values ​​of Clc, Bn, Ccp, and Py determined using the luminance distribution information of individual mineral particles in Table 1 are 130, 126, 114, and 98, respectively, and the initial peak luminance value of Cv interpolated using the procedure described above is 123, then the initial mineral boundary value of Clc / Bn will be 128, which is the midpoint between the mineral boundary values ​​of Clc and Bn. Similarly, the initial mineral boundary values ​​for Bn / Cv are 124.5, for Cv / Ccp they are 118.5, and for Ccp / Py they are 106.

[0043] Furthermore, the initial upper limit of the brightness of the backscattered electron image of Clc, the constituent mineral with the highest peak brightness value, is set to the peak value of Clc + a, and the initial lower limit of the brightness of the backscattered electron image of Py, the constituent mineral with the lowest peak brightness value, is set to the peak value of Py - b. The values ​​of a and b are set based on the position where the brightness of the convex peak curve converges to approximately 0 (background), referring to the brightness distribution information of Clc and Py (histogram with brightness on the horizontal axis and number of pixels on the vertical axis). For example, if the set values ​​are a=7 and b=7, the initial upper limit of the brightness of the backscattered electron image of Clc will be 137, and the initial lower limit of the brightness of the backscattered electron image of Py will be 91.

[0044] (Creation of the first brightness correction information) An example of a first method for creating brightness correction information, which is applied when the area ratio of the mineral of interest within the mineral particles is below a threshold, i.e., when the proportion of the mineral of interest is small relative to the total mineral particles, is described below.

[0045] First, the combination of constituent minerals that make up the first brightness correction information is determined. A grain containing the mineral of interest is extracted from arbitrary mineral particles, and assuming that the grain is composed only of constituent minerals present around the mineral of interest, the mineral species appearing in this region are determined to be the constituent minerals present around the mineral of interest, thereby determining the combination of constituent minerals that make up the first brightness correction information. However, if the mineral species present around the mineral of interest consist only of mineral species with lower brightness than the mineral of interest, in order not to rule out the possibility of a mineral of interest with higher brightness, the grain is determined to be composed of the mineral of interest and constituent minerals in contact with the mineral of interest, thereby determining the combination of constituent minerals.

[0046] For the determined combination of constituent minerals, adjacent constituent minerals are extracted when the peak brightness values ​​of the backscattered electron images of the constituent minerals are arranged in descending order. The intermediate value between the initial lower limit of brightness of the backscattered electron image of the constituent mineral with the higher peak brightness value and the initial upper limit of brightness of the backscattered electron image of the constituent mineral with the lower peak brightness value is determined as the mineral boundary value between the constituent mineral with the higher peak brightness value and the constituent mineral with the lower peak brightness value. This creates a first brightness correction information that defines the brightness range of the backscattered electron images of multiple constituent minerals.

[0047] For example, let's consider the case where the mineral of interest is Bn, and the mineral species in contact with the mineral of interest are Clc and Ccp, as shown in the mineral particles in Figure 5(a). As shown in Figure 8, the mineral species in contact with the mineral of interest are Clc and Ccp, so a first brightness correction information is created based on the mineral species Clc and Ccp. In the calculation of the peak brightness value, mineral boundary initial value, brightness upper initial value, and brightness lower initial value described above, the peak brightness value of Clc is 130, the brightness upper initial value of Clc is 137, and the brightness lower initial value is 128, which is the mineral boundary initial value for Clc / Bn. Also, the peak brightness of Ccp is 114, the brightness upper initial value of Ccp is 118.5, which is the mineral boundary initial value for Cv / Ccp, and the brightness lower initial value is 106, which is the mineral boundary initial value for Ccp / Py.

[0048] Therefore, when creating the first luminance correction information based on the mineral species Clc and Ccp, the mineral boundary value after correction for Clc / Ccp is set to the midpoint between 128, which is the initial lower limit of luminance for Clc (which has a high peak luminance value), and 118.5, which is the initial upper limit of luminance for Ccp (which has a low peak luminance value), i.e., (128 + 118.5) / 2 = 123.25.

[0049] The upper limit of luminance for Clc is set to 137, using the initial value of the upper limit of luminance. The corrected lower limit of luminance for Ccp is set to the midpoint between 106, which is the initial value of the lower limit of luminance for Ccp with a high peak luminance value, and 91, which is the initial value of the mineral boundary of the detection limit (Py / detection limit), i.e., (106 + 91) / 2 = 98.5. As a result, as shown in Figure 9, a first luminance correction information (see Figure 9) is obtained in which the range of luminance in the backscattered electron image from 123.25 to 137 is judged as Clc, and the range of luminance in the backscattered electron image from 98.5 to less than 123.25 is judged as Ccp.

[0050] Using a similar procedure, all combinations of mineral species (Clc, Bn, Cv, Ccp, Py) of the multiple constituent minerals contained in the mineral particles are determined, and a first luminance correction information is created for each mineral particle having this combination. Note that when creating the first luminance correction information, if the mineral particle with the highest luminance and the mineral particle with the lowest luminance are included as mineral species in contact with the mineral of interest, the corrected upper limit of luminance for the mineral particle with the highest luminance uses the initial luminance upper limit value. The corrected lower limit of luminance for the mineral particle with the lowest luminance uses the initial luminance lower limit value.

[0051] When correcting the mineral region of a mineral of interest using first luminance correction information, which includes luminance correction information for the backscattered electron images of constituent minerals surrounding the mineral of interest in such a mineral map, the luminance of the backscattered electron image within the mineral region of the mineral of interest is read pixel by pixel, and the luminance corresponding to that luminance is compared with the first luminance correction information to correct it to the constituent mineral corresponding to the corresponding luminance. Figure 10(a) shows the mineral map of ore particles before correction, Figure 10(b) shows the backscattered electron image, and Figure 10(c) shows the mineral map after correction when ores of Clc, Ccp, and Cv are adjacent to Bn in the mineral map.

[0052] (Creation of the second brightness correction information) An example of a second method for creating brightness correction information, which is applied when the area ratio of the mineral of interest to the mineral particles is above a threshold, i.e., when the proportion of the mineral of interest is large relative to the total mineral particles, is described below.

[0053] First, the combination of constituent minerals that make up the second brightness correction information is determined. From among the multiple mineral particles contained in the MLA data, mineral particles that contain the mineral of interest as a constituent mineral and in which the area ratio of the mineral of interest to the mineral particle is equal to or greater than the threshold are extracted. Next, all constituent minerals contained in the extracted mineral particles are extracted.

[0054] In creating the second luminance correction information, it is assumed that the mineral particles are composed of the constituent minerals determined from the backscattered electron image plus the mineral of interest, so as not to exclude the possibility that the mineral of interest is the main constituent mineral of the mineral map. When creating the second luminance correction information, as with the creation of the first luminance correction information, adjacent constituent minerals are extracted when the peak luminance values ​​of the backscattered electron images of the constituent minerals are arranged in descending order of magnitude. The intermediate value between the initial lower limit of luminance of the backscattered electron image of the constituent mineral with the higher peak luminance value and the initial upper limit of luminance of the backscattered electron image of the constituent mineral with the lower peak luminance value is determined as the mineral boundary value between the constituent mineral with the higher peak luminance value and the constituent mineral with the lower peak luminance value, thereby creating the second luminance correction information which defines the range of luminance of the backscattered electron images of multiple constituent minerals.

[0055] When correcting the mineral region of a mineral of interest using a second set of brightness correction information, which includes brightness correction information for the backscattered electron images of constituent minerals surrounding the mineral of interest in such a mineral map, the brightness of the backscattered electron image within the mineral region of the mineral of interest is read pixel by pixel, and the brightness corresponding to that brightness is compared with the second set of brightness correction information to correct the corresponding brightness to the corresponding constituent mineral.

[0056] Figure 11 shows the results of analyzing the mineral composition ratio using the MLA data correction method according to an embodiment of the present invention, using ore samples taken from an operating copper mine, with high and low concentrations of copper insoluble matter and with a copper grade lower than normal. Sample 7 shows the results when the concentration of copper insoluble matter is low and the copper grade is high, sample 8 shows the results when the concentration of copper insoluble matter is high and the copper grade is high, and sample 9 shows the results when the concentration of copper insoluble matter is low and the copper grade is low. In all of samples 7 to 9, the results were obtained that accurately reflected the composition of the copper concentrate with good reproducibility.

[0057] (Examination of the threshold for the area ratio of the mineral of interest to the mineral particles) Figure 4 shows the threshold for the area ratio of the mineral of interest to the mineral particles in step S33. The threshold was set in the range of 10% to 90%, and based on that threshold, the MLA data of sample 1 in Figures 3(a) and 3(b) was corrected according to the flow in Figure 4. Figure 12 shows the relationship between the threshold and the compositional analysis results when compositional analysis is performed using the corrected MLA data. Figure 13 shows an enlarged graph of the analysis results for Bn in Figure 12. As shown in Figure 12, for Ccp, Py, Cv, and Clc, there was no significant difference in the compositional analysis results depending on the threshold setting compared to Bn. As shown in Figure 13, for Bn, it can be seen that when the Bn threshold is low, the mineral proportion of Bn is calculated to be high. After examining this in comparison with the mineral proportion of Bn in sample 1 in Figures 6(a) and 6(b), it is preferable to set the threshold to 50% or higher, and even more preferably to 60% or higher, and even more preferably to 70% or higher.

[0058] Because MLA data is analyzed using a scanning electron microscope, misidentification can occur due to overlapping minerals in the depth direction of the sample and the presence of boundaries between minerals, which can lead to inaccurate analysis of the mineral composition present in the ore. To address this problem, for example, a known method involves focusing on BSE luminance and correcting the data by removing mineral portions with BSE luminance below a set threshold for mineral species with BSE luminance above a predetermined value. However, there is still room for further investigation into methods for correcting MLA data to improve the accuracy of compositional analysis.

[0059] According to the MLA data correction method of the embodiment of the present invention, instead of uniformly removing low-luminance areas of the backscattered electron image through correction, a mineral species prone to discrepancies between MLA data and measured values ​​is selected as the mineral of interest. Then, MLA data correction is performed on mineral particles containing this mineral of interest. When correcting MLA data, first or second luminance correction information, created using luminance information of constituent minerals constituting the mineral particles or constituent minerals surrounding the mineral of interest, is used to correct the mineral region of the mineral of interest, which is prone to misrecognition in MLA analysis, to a more appropriate mineral species, thereby enabling more accurate analysis of the mineral composition ratio present in the ore.

[0060] Although the present invention has been described using the embodiments described above, it is not limited to each embodiment, and the components can be modified and implemented without departing from the spirit of the invention. Furthermore, various inventions can be formed by appropriately combining the multiple components disclosed in each embodiment. For example, some components may be deleted from all the components shown in the embodiment. Moreover, components from different embodiments may be appropriately combined.

[0061] In addition to the correction method described above, SEM-EDS may also be measured at a lower acceleration voltage (e.g., 15 keV) than the normal setting (e.g., 25 keV) as a method for correcting MLA data. By measuring at a lower acceleration voltage, the spatial resolution of the MLA can be improved, thereby improving the accuracy of the MLA analysis. It is also preferable to set the measurement mode of the MLA analysis software to measure one point at a time for each grain within the mineral particles, and to exclude or not measure measurement data at the mineral boundaries where misidentification is likely to occur.

[0062] In the MLA data correction process S3, it is preferable to automate the correction of MLA data using machine learning with Python or similar software. This allows for the correction of tens of thousands or more mineral particles in a short time, enabling faster acquisition of analysis results. Furthermore, by creating correction information for a large amount of MLA data and further utilizing this in machine learning, the accuracy of MLA analysis can also be improved.

[0063] In the above-described embodiment, an example focusing on Bn as the mineral of interest was explained. However, it is of course possible to set other minerals, such as Clc, Ccp, Py, and Cv, as the mineral of interest. In particular, Clc, Cv, and Py are end-components of Cu-Fe-S copper sulfide minerals, while Bn and Ccp are intermediate compositions of end-components in the Cu-Fe-S system, so the effect of correcting them using this method is especially significant.

[0064] The flowchart in Figure 4 illustrates a method for correcting MLA data using either first or second luminance correction information based on a threshold value for the area ratio of the mineral of interest to the mineral particles. However, the present invention is not limited to the above-described embodiments. For example, this embodiment may also include a method in which, without determining the threshold value for the area ratio of the mineral of interest in step S33, the correction process is performed using only first luminance correction information created using luminance information of backscattered electron images of mineral species surrounding the mineral of interest, or only second luminance correction information created using luminance information of backscattered electron images of mineral species constituting the mineral particles, for the mineral region and mineral species information of the mineral of interest.

Claims

1. A process of obtaining a backscattered electron image of the mineral particles on a measurement surface obtained by measuring the measurement surface of an ore sample in which mineral particles are embedded in resin using a mineral particle analyzer, and a mineral map including information on the mineral regions and mineral species of the constituent minerals that make up the mineral particles. A process of correcting information on the mineral region and mineral species of mineral particles containing the mineral of interest to be corrected as a constituent mineral, using brightness correction information of the backscattered electron image of the constituent mineral. A method for correcting MLA data, including [specific data].

2. In mineral analysis methods using mineral particle analyzers, A process of measuring the measurement surface of an ore sample in which mineral particles are embedded in resin using a mineral particle analyzer, A step of obtaining a backscattered electron image of the mineral particles on the measuring surface and a mineral map including information on the mineral regions and mineral species of the constituent minerals that make up the mineral particles, A step of correcting information on the mineral region and mineral species of a mineral of interest, which includes the mineral of interest to be corrected as a constituent mineral, using brightness correction information of the backscattered electron image of the constituent mineral, A step of analyzing the composition ratio of the constituent minerals present in the ore sample based on the area of ​​the mineral region of the corrected mineral particles and the types of constituent minerals. Mineral analysis methods including

3. The aforementioned correction step is When the area ratio of the mineral of interest to the mineral particles is less than a threshold, the mineral region of the mineral of interest is corrected using first brightness correction information prepared using brightness information of backscattered electron images of mineral species present around the mineral of interest. When the area ratio is greater than or equal to a threshold, the mineral region of the mineral of interest is corrected using second brightness correction information prepared using brightness information of the backscattered electron images of the mineral species constituting the mineral particles. The mineral analysis method according to claim 2, which includes the following:

4. The process includes the step of creating the brightness correction information, and the step includes, Based on the area ratio of the mineral of interest to the mineral particles, the combination of constituent minerals that constitute the brightness correction information is determined, The process involves extracting adjacent constituent minerals when the peak brightness values ​​of the backscattered electron images of the determined constituent minerals are arranged in descending order, and creating brightness correction information by setting the midpoint between the initial lower brightness limit of the backscattered electron image of the constituent mineral with the higher peak brightness value and the initial upper brightness limit of the backscattered electron image of the constituent mineral with the lower peak brightness value as the corrected mineral boundary value between the adjacent constituent minerals. A mineral analysis method according to claim 2 or 3, including the following:

5. The mineral analysis method according to claim 3, wherein the threshold is 50% or more.

6. The mineral analysis method according to claim 2 or 3, comprising analyzing ore from a secondary enrichment zone as the ore sample.

7. The mineral analysis method according to claim 2 or 3, wherein the mineral of interest is bornite.

Images