JP2589575B2 - Method for analyzing multi-element compounds in microscopic X-ray analysis - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION <Industrial Application Field> The present invention relates to an epoch-making method for analyzing multi-element compounds in microscopic X-ray analysis.

<Conventional technology> For example, an electron microscope and a so-called EPMA (Electron Probe X-ra)
y Analyzer: There are energy dispersive EDX and wavelength dispersive WDX), and more specifically, scanning and irradiating an electron beam within a predetermined small area of a sample, A minute portion X that measures each characteristic X-ray radiated from the matrix-like detection point area within the predetermined range and analyzes the sample based on the measurement result of the characteristic X-rays
In the field of line analysis, in recent years, several attempts have been made to apply image processing technology to analysis results by EPMA, but at present, the image processing function based on them has been The results of peak separation, line analysis, surface analysis, etc., which have been performed, are simply imaged.

Therefore, in the image processing method in the microscopic X-ray analysis according to the prior art, the type of element contained in the sample and the concentration thereof are analyzed while applying the image processing technique to the analysis result by EPMA. [Elemental identification] could not escape from the traditional EPMA analysis.

Therefore, the present inventors have applied a completely new image processing means to the analysis results by EPMA in earnest research, and thereby a new analysis field which was impossible in the past, that is, the type of element contained in the sample, Image processing method in micro-part X-ray analysis, which is a base that enables not only analysis of concentration (element identification) but also analysis (substance identification) of types and distributions of compounds constituting a sample (this is Which is a prior art for the present invention), which has already been proposed in Japanese Patent Application No. 1-61209.

That is, the image processing method in the microscopic X-ray analysis according to the prior art is to scan and irradiate an electron beam to a minute predetermined range in the sample, and to scan and irradiate the sample from a matrix-like detection point area within the predetermined range. In the minute part X-ray analysis in which the characteristic X-rays regarding the emitted contained elements are measured and the sample is analyzed based on the measurement results of the characteristic X-rays, corresponding to a matrix-like detection point area within a predetermined range of the sample. In a memory having a memory area and an address to be stored, the measurement result of the characteristic X-ray is stored for each address,
The characteristic X-ray measurement result stored in the memory is converted into a luminance value of image information, and then an image luminance value histogram is created. Each peak in the characteristic X-ray image luminance value histogram is converted into an adjacent peak. The number of pixels corresponding to a matrix-like detection point area within a predetermined range of the sample is determined by multi-leveling so that the foot portion of each sample does not interfere with each other, and only the multi-valued part of each peak in the image luminance value histogram of the characteristic X-ray. And an image processing means for displaying on the matrix screen having addresses and addresses in different colors or symbols, and a specific method thereof will be described below.

FIG. 7 and FIG. 8 show a specific example of an analysis result using the image processing method in the microscopic X-ray analysis using dolomite ore as a sample. The dolomite ore as the sample is composed of three kinds of compounds, namely, dolomite [CaMg (CO ₃ ) ₂ ], dolomite [CaCO ₃ ], and limestone [Ca
CO ₃ · MgCO ₃ ].

First, similarly to the conventional method, an electron microscope and an EPMA are used to scan and irradiate an electron beam on a minute predetermined area of the sample, and accordingly, each of the matrix-like detection point areas within the predetermined area of the sample is irradiated. The energy intensity of characteristic X-rays (X-rays unique to each element) emitted from each detection point is measured.

Next, the measurement results of the energy intensities of the characteristic X-rays for each element obtained by the EPMA are respectively converted into a matrix detection point region (for example, 256 × 256) within a predetermined range of the sample.
In a memory having a memory area (65536 memory) corresponding to a pixel) and an address, the data is stored for each address.

Here, if the energy intensity measurement result of the characteristic X-ray stored in each memory is analyzed according to a method similar to the conventional method, the type of element contained in the sample (in this example, C
a, Mg, C, O) and the concentration (content ratio) are each analyzed, and the element identification of the sample can be performed.

Further, the result of measuring the energy intensity of the characteristic X-ray stored in each memory is converted into a luminance value (image density value) of image information (normalized to, for example, 0 to 255), and then each element is analyzed. The image brightness value histogram (the relationship between the image brightness value and the count number of pixels having the same image brightness value) is created. FIG. 7 shows an example of the image luminance value histogram, which is a characteristic X-ray of Ca selected as an element contained in all three compounds constituting the sample (dolomite ore). Of the image brightness value histogram H, and it can be seen that there are three peaks A, B, and C corresponding to these three types of compounds.

Here, data (65536 pieces of image luminance value data) in all areas of each peak A, B, and C in the image luminance value histogram H of the characteristic X-ray of Ca obtained as described above is
Assuming that the brightness is simply displayed as it is on a single matrix screen having the number of images (256 × 256 pixels) and addresses corresponding to the matrix detection point area within a predetermined range of the sample, all pixels on the image are displayed. Since the image brightness changes only in a continuous plane (analog), the boundaries between the peaks A, B, and C cannot be grasped at all.

Therefore, as shown in FIG. 7, the peaks A, B, and C in the image characteristic value histogram H of the characteristic X-ray of Ca are compared with the peaks (A and B, B and C) of adjacent peaks. Multi-valued so that the foot part does not interfere (divided into a, b, c parts)
And the multivalued parts (a, b, c) of each of the peaks A, B, C
8, only one matrix screen P (for example, CRT) having the number of pixels (256 × 256 pixels) and the address corresponding to the matrix-like detection point area within a predetermined range of the sample as shown in FIG. ), Different colors (for example, red, blue, yellow) or the above (for example, ○,
×, △) to display (map) the peaks A, B, C
Respectively, and the boundaries, that is, the phases of the three types of compounds constituting the sample (dolomite ore) can be clearly distinguished and grasped.

As described above, according to the image processing method in the microscopic X-ray analysis, in particular, an image luminance histogram of characteristic X-rays of the elements constituting the sample is created, and each peak in the image luminance value histogram is adjacent to each other. Multi-valued so that the foot portions of the peaks do not interfere with each other, and only the multi-valued portion of each peak in the image luminance value histogram is represented by the number of pixels and the address corresponding to the matrix detection area within a predetermined range of the sample. On the matrix screen having, by using a unique image processing means of displaying with different colors or symbols,
The area and each boundary occupied by each of the peaks, that is,
The phases (distribution state) of the compounds constituting the sample can be clearly distinguished and grasped, and thus, a new analysis field which has not been possible in the past, that is, the types of elements constituting the sample Thus, the possibility of analyzing (identifying substances) the types and distributions of the compounds constituting the sample is not limited to analysis of concentrations (element identification).

<Problems to be Solved by the Invention> However, the following problems still remain in the prior art (image processing method in microscopic X-ray analysis) in which very excellent advantages can be expected as described above. .

That is, in the case of using the image processing method in the microscopic X-ray analysis according to the above-mentioned prior art, (a) the number and distribution of the compounds constituting the sample can be analyzed, but once for the compound, The only information obtained by the analysis is that each of these compounds contains one selected element, and that each compound contains any other kind of element in any ratio. Not only has the basic limitation that it cannot be specified by itself, but (a)
As in the case of the above analysis example, peaks A, B, and C in image luminance value histogram H of characteristic X-rays of Ca selected only as one element contained in the sample (dolomite ore)
There is no problem because the number (3) of the compound (3) coincides with the number (3) of the compounds (kukai stone, dolomite, limestone) constituting the sample. The number of peaks in the characteristic X-ray image luminance value histogram does not always match the number of compounds constituting the sample (for example, in the characteristic X-ray image luminance value histogram of one selected element, Even if a peak looks like one, it may actually be made up of multiple peaks.) In such a case, multiple compounds may be mistaken for one compound. There is also a problem that correct analysis cannot be performed.

The present invention has been made as a result of further research in view of such circumstances, and its object is to provide a large number (n kinds: n ≧ 3).
(Sample of a multi-component system) composed of a compound containing the following elements (first element to n-th element), the sample is formed by first focusing on two types of elements contained in the sample. A micrometer that allows the number of compounds to be analyzed easily and accurately, as well as easily obtains more detailed information on each compound (what other elements are included in what ratio). Another object of the present invention is to develop a method for analyzing a multi-element compound in a part X-ray analysis.

<Means for Solving the Problems> In order to achieve the above object, the present invention provides a method for preparing a sample comprising a compound containing a plurality of (n types: n ≧ 3) elements (first element to n-th element). Scans and irradiates an electron beam within a predetermined range, and measures characteristic X-rays of the respective element elements respectively emitted from the matrix-like detection point areas within the predetermined range, thereby measuring the characteristic X-rays within the predetermined range of the sample. The measurement results of the characteristic X-rays relating to the n kinds of contained elements are stored for each address in n memories having a memory area and an address corresponding to the matrix detection point area in the above, and stored in each of the memories. The measurement results of the characteristic X-rays of each of the contained elements are respectively converted into luminance values of image information, and image luminance value histograms of the n contained elements are respectively created. (I) First, the first Former From the image luminance value histogram of the first element and the image luminance value histogram of the second element, (II) Next, in the first image luminance value correlation chart for the two elements, A group of data groups is extracted, and is converted into a first image luminance value histogram for the first element or the second element based on the extracted data result of the group of data groups, and the first converted image luminance value histogram is obtained. And the image brightness histogram of the third element, and the second image brightness value correlation diagram for the three elements is sequentially and repeatedly performed up to the n-th element. (N-1) From the image brightness value correlation diagram, the content ratio of n kinds of elements (first element to n-th element) constituting the compound is analyzed. An object of the present invention is to provide a method for analyzing a multi-element compound in a microscopic X-ray analysis characterized by the above.

<Operation> The operation exhibited by adopting such a characteristic means is as follows.

That is, in the method for analyzing a multi-element compound in the microscopic X-ray analysis according to the present invention, one kind of compound constituting the sample is contained as in the image processing method in the microscopic X-ray analysis according to the prior art. Instead of focusing only on the elements, as will become clearer from the description of Examples described later, first, two kinds of contained elements (the first element and the second
Element), and a first image luminance value correlation diagram for both elements is created based on the two image luminance value histograms obtained from the measurement results of the characteristic X-rays of the two elements. Based on the number and positions of the data groups appearing in the figure, the number of compounds containing both elements and the content ratio of both elements in each compound can be easily and accurately determined. Further, a group of data groups in the first image luminance value correlation diagram for the first element and the second element is extracted, and based on the data result of the extracted group of data groups, the first element or the second element is extracted. The first image luminance value histogram is converted into a first image luminance value histogram, and the second image luminance values of the three elements are obtained from the first converted image luminance value histogram and the image luminance histograms of other elements (third element,..., N-th element). The process of creating a correlation diagram is sequentially repeated, and finally the (n-1) th
By obtaining the image brightness value correlation diagram, not only the sample composed of the binary and ternary compounds but also the case of the multi-component sample in which the contained elements in each compound are very various, It has become possible to easily determine the content ratios of all elements in a compound.

In short, the gist of the present invention is as follows. For example, the first element, the second element, and the third element are respectively
In the case of Mg, Si, and Ca, it can be seen from the first image luminance value correlation diagram (see FIG. 4) for the first element and the second element.

From the number of data groups G1, G2, and G3 that appeared, the number of compounds containing (Mg, Si) is three.

Among the three compounds, the content ratio (Mg: Si) of the compound (Mg, Si) in the compound corresponding to the data group G1 can be read from the position of the data group G1. Similarly, the content ratio of (Mg, Si) in the compounds corresponding to the data groups G2 and G3 can be read from each position.

Subsequently, from the data groups G1, G2, and G3 that have appeared, only one data group G1 (a group of data groups) is selected, and only the values of the data group G1 are used as shown in FIG. For example, after creating a first converted image luminance value histogram for Mg (similarly for Si), the first converted image luminance value histogram and the third element Ca shown in FIG.
Then, a second image luminance value correlation diagram of three elements as shown in FIG. 6 is created from the image luminance histogram H3 of FIG. It can be seen from this second image luminance value correlation diagram.

(Mg, Si) in the compound corresponding to the data group G1
And the content ratio of Ca and [(Mg, Si): Ca] can be read from the position of the data group GI appearing in the second image luminance value correlation diagram.

Since the value of Mg: Si is constant as apparent from FIG. 4, the horizontal axis in FIG. 5 may be either Mg or Si.

In other words, for the three component elements (Mg, Si, Ca) in the corresponding compound from one selected data group G1,
The following can be seen.

That is, from the position of the data group G1, for example, Mg: S
If i = A: B, then Mg: Si = Mg / Si = A / B (= C = constant value). Therefore, from the position of the data group GI, [M
g, Si): Ca] = C: D, the content ratio of the third element to the first and second elements in the compound corresponding to the data group G1 can be determined.

<Example> Hereinafter, as a specific example of an image processing method to which the analysis method of a multi-element compound is applied in the microscopic X-ray analysis according to the present invention, at least three kinds of elements (first element Mg, second element Mg, second element The analysis results using a meteorite composed of a compound containing the element Si and the third element Ca) will be described.

First, as in the conventional method, the electron microscope and
Using an EPMA, an electron beam is scanned and irradiated on a minute predetermined range of the sample, and the respective contents radiated from each detection point of a matrix-like detection point area within the predetermined range of the sample accordingly. The characteristic X-rays relating to the elements (Mg, Si, Ca,...) (X-ray intensity specific to each element is measured.

Next, the elements (Mg, Si, Ca,...) Obtained by the EPMA
The measurement result of another characteristic X-ray is stored in a memory area (65536 memory) corresponding to a matrix detection point area (for example, 256 × 256 pixels) and a memory having an address within a predetermined range of the sample, respectively. Store them separately.

Here, if the measurement result of the characteristic X-ray stored in each memory is analyzed according to a method similar to the conventional method, the types of elements contained in the sample (in this example, Mg, Si, Ca,...) Then, the concentration (content ratio) is analyzed, respectively, and element identification of the sample can be performed.

Further, the measurement results of the characteristic X-rays stored in each memory are converted into luminance values (image density values) of image information (normalized to, for example, 0 to 255), and then, the Image luminance value histogram (Relationship between image luminance value and count of pixels having the same image luminance value)
Is created as illustrated in FIGS. 1 to 3.

Among them, the image luminance value histogram H1 of the characteristic X-ray of the first element Mg shown in FIG. 1 has three peaks A1 and B in this example.
1, C1, and the image brightness value histogram H2 of the characteristic X-ray of the second element Si shown in FIG. 3 also has three peaks in this example.
A2, B2, and C2, and the third element Ca shown in FIG.
It can be understood that the characteristic X-ray image luminance value histogram H3 has two peaks A3 and B3 in this example.

Then, as seen from the image luminance value histograms H1 and H2 of both the first element Mg and the second element Si, except for the peaks A1 and A2 corresponding to the background (see FIG. 4 described later) which is a noise component. Considering this, it is presumed that this sample (meteorite) is composed of two types of compounds.

However, first, the first element Mg and the second element Si
Of the first element M using the image luminance value histograms H1 and H2 of
As shown in FIG. 4, a correlation diagram showing the correlation between the image luminance value of the characteristic X-ray of g and the image luminance value of the characteristic X-ray of the second element Si is shown in FIG. Pixel), it turns out that the above estimation is incorrect.

That is, in the first image luminance value correlation diagram for the first element Mg and the second element Si shown in FIG. 4, the horizontal axis (x-axis) indicates the characteristic X-ray image luminance value (0 to 255) of the first element Mg. ), The vertical axis (y-axis) is the image luminance value (0 to 25) of the characteristic X-ray of the second element Si.
5), and each point shown on this xy screen is
The number of data related to (Mg image luminance value, Si image luminance value) is represented by luminance of image information (image density: for example, 255 gradations from 0 to 255). For example, x = 50 and y = 10
The number of data that satisfies 0 is counted with reference to the original data, and the number is displayed at the pixel located at (x coordinate, y coordinate) = (50, 100) at a luminance corresponding to the number. In this example, if the above operation is performed for each of x and y in 256 units, one point at a time, the number of operations is 256 ² = 65536.

Now, the first element Mg and the second element Si described above
As is clear from the image luminance value correlation diagram, in this example, data groups (in this example, three data groups G1, G2, and G3 in addition to the data group G0 corresponding to the background) are respectively formed. You can see that. Therefore, this means:
Data groups G1, G2, G3 appearing in the first image brightness value correlation diagram
It can be seen from the number (3) that the compound containing the first element Mg and the second element Si exists not in the two types estimated from FIGS. 1 and 2, but actually in three types. . That is, in the image luminance value histogram H1 of the characteristic X-ray of the first element Mg shown in FIG.
2 as one peak B1, and in the image luminance value histogram H2 of the characteristic X-ray of the second element Si shown in FIG. 2, two data groups G2 and G3 appeared as one peak B2. Can understand.

Then, the content ratio of the first element Mg and the second element Si in each compound is read from the positions (x coordinate, y coordinate) of each data group G1, G2, G3 appearing in the first image luminance value correlation diagram. Can be.

Next, with respect to the three data groups G1, G2, G3 appearing in the first image luminance value correlation diagram of FIG. 4, the data groups G1, G2, G3
Only one data group is selected from the respective detection point groups, and a first converted image luminance value histogram for the first element Mg or the second element Si is created using only the value of the selected data group. . In this example, the data group G1 is selected from the three data groups G1, G2, and G3 appearing in the first image luminance value correlation diagram. In this case, for example, as shown by a square window mark, a representative portion is extracted (that is, multi-valued) from the data group G1 in FIG. 4, and based on the data corresponding to the multi-valued portion a ', A first converted image luminance value histogram relating to the first element Mg or the second element Si as shown in FIG. 5 is converted. In the first converted image luminance value histogram shown in FIG. 5, the horizontal axis (x-axis) represents the characteristic X-ray image luminance value (0 to 255) of the first element Mg or the second element Si, and the vertical axis (y).
The axis indicates the count number of pixels having the same image luminance value of the first element Mg or the second element Si.

Subsequently, the first element Mg and the second element shown in FIG.
From the first converted image luminance value histogram for Si and the third element Ca image luminance histogram shown in FIG. 3, a second (Mg / Si-Ca) image luminance value for these three elements (Mg, Si, Ca) If the correlation diagram is created as exemplified in FIG. 6, if the position (x coordinate, y coordinate) of the data group GI appearing in the second (Mg / Si-Ca) image brightness value correlation diagram is read, In the compound (the first element Mg, the second element Si: the third element C
The content ratio of a) can be determined.

The process according to the above procedure is sequentially repeated for the data groups G1, G2, and G3 appearing in FIG. 4 up to the n-th element, thereby obtaining the (n-1) -th image luminance value correlation diagram. With respect to all the compounds constituting the sample, the content ratio of n kinds of elements contained in each compound can be easily analyzed.

Then, as exemplified in FIG. 6, in the finally obtained (n-1) -th image luminance value correlation diagram, for example, as shown by a square window mark d, n kinds of elements A representative portion of the data group GI having a constant content ratio is extracted (that is, multivalued), and data corresponding to the multivalued portion of the data group GI,... After taking out the characteristic X-ray measurement results stored in each memory and converting them to image luminance values (image density values) (for example, normalizing them to 0 to 255), Although omitted, different colors (for example, for each data group) are displayed on one matrix screen (for example, CRT) having the number of pixels (256 × 256 pixels) and the address corresponding to the matrix detection point area within the predetermined range of the sample. , Red, blue, yellow) or symbols (for example, ○, ×, 等) and the like (mapping), the sample (meteorite) can be obtained in the same manner as shown in FIG. The region and each boundary occupied by all the constituent compounds, that is, the phases of the compounds can be clearly distinguished and grasped, and therefore, the content ratio of the compounds can be read.

In this example, a data group G1 is selected from the three data groups G1, G2, and G3, and a first converted image luminance value histogram for the first element Mg or the second element Si is determined based on the data result of the data group G1. I have shown what I created,
The data group G2 may be selected from the three data groups G1, G2, G3 (in this case, as shown in FIG. 5 based on the data corresponding to the multi-valued portion b 'in the data group G2 in FIG. 4). A first converted image brightness value histogram relating to the first element Mg or the second element Si is converted.) Alternatively, a data group G3 may be selected from the three data groups G1, G2, G3 (in this case, FIG. 4). Is converted into a first converted image luminance value histogram for the first element Mg or the second element Si as shown in FIG. 5 based on the data corresponding to the multi-valued portion c 'in the data group G2). By selecting only one data group, the corresponding compounds can be separated and analyzed.

<Effects of the Invention> As is clear from the above description, according to the method for analyzing a multi-element compound in the microscopic X-ray analysis according to the present invention, particularly, the sample is contained in the sample as in the prior art. Instead of focusing on only one kind of element, first, two kinds of contained elements (first element and second element) are selected, and both image luminance value histograms obtained from the measurement results of characteristic X-rays of both elements By creating a first image luminance value correlation diagram for both elements based on the number of data groups appearing in the first image luminance correlation diagram and each position,
The number of compounds containing both elements and the content ratio of both elements in each compound can be easily and accurately determined, and furthermore, a group of data groups in the first image luminance value correlation diagram for the first element and the second element. Is extracted, and is converted into a first image luminance value histogram for the first element or the second element based on the data result of the extracted group of data, and the first converted image luminance value histogram and another type Element (third element, ..., n-th element
From the image luminance value histogram of the three elements) to sequentially obtain the (n-1) th image luminance value correlation diagram by successively repeating the process of creating a second image luminance value correlation diagram for the three elements. Not only for samples consisting of binary and ternary compounds, but also for multi-component samples containing a very large number of elements in each compound, the content ratio of all elements in each compound can be easily determined. Can solve the problems (a) and (b) in the prior art as described above.
A remarkably excellent effect is exhibited.

[Brief description of the drawings]

FIG. 1 to FIG. 6 are for describing a specific embodiment of the method for analyzing a multi-element compound in the microscopic X-ray analysis according to the present invention, and FIG. 1 shows the structure of a sample (meteorite). Histogram of the characteristic X-ray of the first element (Mg) contained in the compound to be converted, FIG. 2 is the image luminance histogram of the characteristic X-ray of the second element (Si), and FIG. Image luminance histogram of characteristic X-ray of element (Ca), FIG. 4 shows first element Mg and second element Si expressed by luminance display method
FIG. 5 is a first converted image luminance value histogram, and FIG. 6 is a second (Mg / Si-C
a) Image brightness value correlation diagrams are shown. 7 and 8 are for explaining the technical background of the present invention and the problems in the prior art, and FIG. 7 is a diagram showing the image processing in the microscopic X-ray analysis according to the prior art. In the method, a luminance histogram of characteristic X-rays of one kind of element (Ca) contained in a compound constituting a sample (dolomite ore) and an explanatory diagram of a multi-value conversion method thereof are shown, and FIG. FIG. 3 is a schematic explanatory view of a matrix screen displaying a distribution state of a compound containing (Ca). G1 …… A set of data.

Claims (1)

(57) [Claims] 1. The method according to claim 1, wherein a plurality of (n kinds: n ≧ 3) elements (first element to
The electron beam is scanned and radiated onto a minute predetermined range in a sample made of a compound containing the (n-th element), and accordingly, each of the organic elements emitted from the matrix-like detection point region in the predetermined range is irradiated. Measure the characteristic X-ray,
In n memory having a memory area and an address corresponding to a matrix-like detection point area within a predetermined range of the sample, measurement results of the characteristic X-rays regarding the n kinds of contained elements are respectively stored for each address, After converting the characteristic X-ray measurement results of each contained element stored in each of the memories into luminance values of image information, and creating image luminance value histograms of the n contained elements,
(I) First, the image luminance value histogram of the first element and the second
A first image luminance value correlation diagram for both elements is created from the image luminance value histogram of the elements, and (II)
In the first image brightness value correlation diagram relating to the two elements, a data group that becomes a block is extracted, and a first data group is extracted based on the data result of the extracted block data group.
Converting into a first image luminance value histogram for the element or the second element, and creating a second image luminance value correlation diagram for the three elements from the first converted image luminance value histogram and the image luminance histogram for the third element; From the finally obtained (n-1) th image luminance value correlation diagram, the n kinds of elements (first element to A method for analyzing a multi-element compound in microscopic X-ray analysis, wherein the content ratio of the (n-th element) is analyzed.