JPH0821808A - Analysis position determining method - Google Patents

Abstract

PURPOSE:To determine a position having a uniform composition to a relatively deep point as an analysis position by generating a map of concentration for each specified element, generating a dispersion diagram based on the map, determining the concentration value of the element at the center of gravity position of a cluster appearing on the dispersion diagram, and determining the position having this concentration value on the map as the analysis position. CONSTITUTION:When the analysis position determination of a specific element is instructed by an input device 14, a scanning electronic microscope(SEM) 1 scans a sample 3 over the prescribed range, and a detector 4 detects radiated X-rays. A CPU 11 generates a map written with the concentration of the specified element for each picture element based on the analyzed result of an X-ray analyzer 5 and stores it in an X-ray image memory 13. The CPU 11 specifies the type of the dispersion diagram to be generated at this time and stores the generated dispersion diagram in the dispersion diagram memory 13. The dispersion diagram is displayed on a monitor 15, a closed area is drawn with a light pen, the concentration of the center of gravity position of each cluster is obtained, and the position having this concentration on the map is determined as the analysis position. This analysis position holds a high possibility to have a uniform composition to a deep point and is preferable.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION The present invention relates to an Electron Probe x-ray Micro Analyzer:
Hereinafter, it will be referred to as EPMA) or the like, and relates to a method for determining an analysis position in an apparatus for performing X-ray analysis.

[0002]

2. Description of the Related Art Conventionally, when performing elemental analysis or the like by using EPMA, the analysis position of a sample is usually determined by observing a backscattered electron image. That is, first, by raster-scanning an electron beam over the analysis range of the sample, detecting backscattered electrons generated from the sample at that time to obtain a backscattered electron image of the analysis range, and observing the backscattered electron image. The analysis position is determined.

After determining the analysis position, the determined analysis position is irradiated with an electron beam, and X-rays emitted from the sample at that time are detected and analyzed to perform elemental analysis. The current situation.

[0004]

By the way, when performing elemental analysis by EPMA, the points to be irradiated with electron beams are
As shown in FIG. 6A, it is desirable to irradiate a portion having a uniform composition to a relatively deep portion. This is because X-rays are also emitted from deep places of about 1 to 2 μm.

On the other hand, since the backscattered electrons are generated from a place shallower than several hundred Å, for example, as shown in FIG. 6B, the same substance as that shown in FIG. 6A exists only in a shallow place on the sample surface. As shown in FIG. 6A, the backscattered electron image of No. 1 becomes the same as the backscattered electron image of a uniform portion even at a relatively deep composition.

Therefore, it is impossible to determine the state of uniformity of the internal composition only by observing the backscattered electron image, and as a result, the composition is originally uniform to a relatively deep portion as shown in FIG. 6A. Although there is a desire to analyze a location, there is a problem that a location where the same composition exists only in a shallow location on the sample surface is determined as an analysis location as shown in FIG. 6B.

However, when the portion shown in FIG. 6B is analyzed, the composition in the depth direction is not uniform. Therefore, there is a difference in the analysis result from the case where the portion shown in FIG. 6A is analyzed. It is clear that this will happen. Of course, even in such a case, although it is possible to identify the element with a certain probability by subjecting the detection result to statistical processing, such statistical processing is troublesome.
Moreover, the analysis time is lengthened because it also analyzes a portion that is not desirable for analysis.

The present invention has been made in view of the above problems, and it is an object of the present invention to provide an analysis position determining method capable of determining a uniform position to a relatively deep composition as an analysis position. is there.

[0009]

In order to achieve the above object, the analytical position determining method of the present invention creates a concentration map for each designated element of a sample and creates a scatter diagram based on the map. The concentration value of the element at the barycentric position of the cluster appearing in the scatter diagram is obtained, and the position on the map where the obtained concentration value is obtained is set as the analysis position.

[0010]

In the present invention, first, a concentration map is prepared for each designated element of the sample. Then, a scatter plot is created based on the map. This scatter diagram may be a scatter diagram showing the relationship of the concentrations of a plurality of arbitrarily selected elements, or may be a scatter diagram by principal component analysis.

Next, the position of the center of gravity of the cluster appearing in the scatter diagram is obtained. The position of this center of gravity is the position having the maximum value in the cluster.

Then, the concentration value of each element at the position of the center of gravity of the cluster is obtained, and the position on the map where the concentration value is obtained is set as the analysis position.

It is highly possible that the analysis position obtained as described above is uniform to a relatively deep composition, and therefore the analysis result at this analysis position is good.
Therefore, elemental analysis can be performed efficiently in a short time.

[0014]

Embodiments will be described below with reference to the drawings.
FIG. 1 is an EPM to which an analysis position determining method according to the present invention is applied.
It is a figure which shows the structure of one Example of A, 1 is a scanning electron microscope (henceforth SEM), 2 is a deflection coil, and 3 is a figure.
Is a sample, 4 is an X-ray detector, 5 is an X-ray analyzer, 6 is a scanning circuit, 10 is a controller, 11 is a CPU, 12 is an X-ray image memory, 13 is a scatter diagram memory, 14 is an input device, 15 Indicates a monitor.

In FIG. 1, a sample 3 is placed at a predetermined position on the SEM 1. Further, the SEM 1 includes a deflection coil 2 for raster-scanning the electron beam over a predetermined range of the sample 3. A deflection current for raster-scanning the electron beam is supplied from the scanning circuit 6. In addition, SE
Various coils other than the deflection coil 2 are arranged in M1, but they are omitted here because they are not essential matters in the present invention.

The X-ray detector 4 detects X-rays emitted from the sample. There are various types of X-ray detectors, but any type may be used.

The X-ray analyzer 5 analyzes and outputs what kind of element the characteristic X-ray of the X-ray detected by the X-ray detector 4 is.

The control device 10 performs the processing for determining the analysis position according to the present invention based on the output from the X-ray analysis device 5, and includes a CPU 11, an X-ray image memory 12, and a scatter diagram memory 13. . The operation of the CPU 11, the X-ray image and the scatter diagram will be described later in detail.

The input device 14 is composed of a keyboard, a light pen, etc., and the monitor 15 is provided for displaying various data such as the display of the obtained analysis position.

Next, the operation of the operator and the operation of the CPU 11 at that time when the analysis position determination processing is performed will be described. When the operator gives an instruction from the input device 14 to execute the processing for determining the analysis position, the CPU 11 gives an instruction to the scanning circuit 6 to generate a deflection current for raster scanning.

As a result, raster scanning of the electron beam is performed over a predetermined range of the sample, and the X-ray emitted from the sample at this time is detected by the X-ray detector 4 and analyzed by the X-ray analyzer 5. .

Then, the control device takes in the analysis result of the X-ray analysis device 5, creates a map for the designated element,
The created map is stored in the X-ray image memory 12. In addition,
The designated element is previously input by the operator from the input device 14. That is, the operator designates which element a map is to be created for, prior to the instruction to execute the process, and the designated element is the designated element. Further, since the map is well known, detailed description thereof is omitted, but the concentration value of the element, that is, the count value of the characteristic X-ray of the element is written for each pixel.

Next, the CPU 11 creates a scatter diagram based on the map of the designated element. The scatter diagram has at least the following two types.

One scatter diagram is a scatter diagram showing the relationship between the concentrations of the elements. If, for example, it is instructed to create a map showing the relationship between the concentrations of the elements A and B, in this case, The CPU 11 reads the concentration values at the same coordinates on the element A map and the element B map, and the combination of the concentration values is used with the element A concentration value on one axis and the element B concentration value on the other axis. Create a scatter plot by plotting on the Cartesian coordinates as the axis.

More specifically, as shown in FIG. 2A, the concentration value of coordinates (X _P , Y _P ) is N _AP in the map of element A, and in the map of element B as shown in FIG. 2B. Assuming that the concentration value of the coordinates (X _P , Y _P ) is N _BP , for example, as shown in FIG. 2C, the concentration of the element A is plotted on the horizontal axis and the concentration of the element B is plotted on the vertical axis (N _AP , N _BP ) is plotted for all pixels in the map.

It is known that the scatter diagram has the following properties. The points on the scatter plot correspond one-to-one to the underlying map. Portions of the same composition form a cluster. The spread of clusters represents variations in composition and statistical fluctuations in measurement. A cluster containing many points represents a composition that occupies a large area on the map. Clusters are connected in the region where the composition continuously changes from one composition to another. The exact same composition occupies the same coordinates on the scatter plot.

The creation of the scatter diagram showing the relationship between the concentrations of the two kinds of elements has been described above, but it is often possible to create a scatter diagram showing the relationship between the concentrations of the three kinds of elements in the same manner. It is a known matter.

Further, when creating the scatter diagram, the operator may set in advance which element the scatter diagram is to be created from the input device 14, or the CPU 11 may make the CPU 11 when the map is completed. Monitor 1
A menu for requesting the designation of elements for creating a scatter diagram may be displayed in 5, and a scatter diagram may be created for the element set by the input device 14 at that time.

Another scatter plot is a scatter plot by principal component analysis. The method of principal component analysis is widely known as one method of statistical processing, but it will be outlined below.

Now, if there are four clusters as shown in FIG. 3A in the scatter diagram created as described above, the direction Z ₁ in which the clusters are dispersed as shown in the diagram.
It is possible to take the (first principal component) and the direction Z ₂ (second principal component) which is orthogonal to it (thus, as shown in FIG. 3B), consider orthogonal coordinates having these Z ₁ and Z ₂ as two axes. You can This is the scatter plot by principal component analysis. The scatter diagram by the principal component analysis shown in FIG. 3B is obtained by performing parallel translation and rotation on the scatter diagram showing the relationship between the concentrations of the elements shown in FIG. 3A. It is clear that the coordinates have a one-to-one correspondence with the coordinates of the scatter diagram showing the relationship between the concentrations of the elements.

As described above, the concentration of elements may be used as the two axes of the scatter diagram, or the principal component functions of the first principal component and the second principal component may be calculated from the scatter diagram showing the concentrations of the elements. The obtained first and second principal components may be biaxial. The operator may set in advance which scatter diagram is to be created from the input device 14, or the CP may be set at the time when the creation of the map is completed.
The U11 may display a menu requesting the designation of the type of scatter chart to be created on the monitor 15, and at that time create a scatter chart of the type set by the input device 14.

After creating the scatter diagram, the CPU 11 stores the scatter diagram in the scatter diagram memory 13.

Now, when the scatter plot is created, the CPU
11 performs clustering. This is a process for making a region in which the points plotted in the scatter diagram are collected into one group, and this clustering process is performed by, for example, the CPU.
11 may calculate the density of points on the scatter plot and automatically recognize the regions having a certain density or more as one group, that is, a cluster, or monitor the scatter plot created. It is also possible to display on 15 and let the operator draw a closed area on the screen with a light pen, and recognize the drawn closed area as one cluster.

If a scatter diagram as shown in FIG. 4A is created by the above clustering process, for example, two clusters A and B will be recognized as shown in FIG. 4B.

By the way, as is clear from the above-mentioned properties of the scatter diagram, even if the position on the map is different, if the composition is the same, that is, the combination of the concentrations is the same, it is plotted at the same position on the scatter diagram. Because it will be
The position on the map corresponding to the area where the points are dense on the scatter diagram has a high possibility that the composition is uniform to a relatively deep place, which is desirable as the analysis position.

Then, the CPU 11 next finds the position of the maximum point where the most points are plotted for each cluster on the scatter diagram. This is the center of gravity of the cluster. Although it is clear that the position of the maximum point in which the points are plotted more than the surroundings may be obtained, the position of the center of gravity is obtained here.

Then, the CPU 11 obtains the obtained concentration of the center of gravity and stores the obtained concentration in a predetermined area of the internal memory. For example, the center of gravity of cluster A shown in FIG. 4B is Q in FIG.
If the position is indicated by, the CPU 11 obtains the concentration N _AQ of the element A and the concentration N _{BQ of the} element B at the position Q and stores these concentration values.

The above is the description of the case of using the scatter diagram showing the relationship of the concentrations of the elements, but the concentration of each element in the center of gravity is similarly obtained when the scatter diagram by the principal component analysis is used. It is clear that

That is, what is important here is not the position of the center of gravity on the scatter diagram, but the concentration value of each element at the center of gravity.

Next, it is necessary to find at which position on the map the density at the center of gravity was obtained. Therefore,
The CPU 11 reads the density values of the pixels at the same coordinates from the maps of all the elements used in the scatter diagram, compares the read density values with the previously stored density values of the center of gravity, and reads all the elements of the elements read from the map. When the density value matches the density value of the center of gravity, the position of the coordinate is determined as the analysis point, and this coordinate value is stored.

By performing this processing for all pixels of the map, all analysis positions can be obtained.

Taking the case of FIG. 2 as an example, it is as follows.
Now, it is assumed that the position indicated by “•” in FIG. 2C is the center of gravity.
At this time, the CPU 11 first determines the concentration at the origin (0,0) of the element A map shown in FIG. 2A and the element B shown in FIG. 2B.
Read the density at the origin (0,0) of the map. _Assuming that the concentration of the element A is N _A00 and the concentration of the element B is N _B00 at this coordinate, the CPU 11 causes the concentration N _A00 of the element A in the map and the concentration N of the element A in the center of gravity.
_AP is compared, and the concentration of element B in the map is also N _B00
And the concentration N _BP of the element B at the center of gravity are compared. Then, when N _A00 = N _AP and N _B00 = N _BP , the coordinate (0,0) on this map is determined as the analysis position.

When the processing for the coordinate (0,0) is completed, the CPU 11 performs the same processing for the next coordinate (1,0). It is obvious that all the analysis positions can be obtained by performing the above-mentioned processing for all the coordinates of the map, that is, all the pixels in this way.

When the analysis position is obtained as described above, C
The PU 11 displays the coordinates on the map determined as the analysis position on the monitor 15. Although not shown in FIG. 1, a printer may be provided to output the coordinates of the determined analysis position to the printer. It is clear that a plurality of analysis positions can be obtained.

The above is the processing of the analysis position determination method according to the present invention. After the analysis position is obtained in this way,
Which position is actually analyzed is arbitrary. It goes without saying that the analysis can be performed at all the obtained analysis positions, but it is also possible to select and analyze some analysis positions.

Although one embodiment of the present invention has been described above, it will be apparent to those skilled in the art that the present invention is not limited to the above embodiment and various modifications can be made. [Detailed Description of the Invention]

[0047]

INDUSTRIAL APPLICABILITY According to the present invention, it is very likely that the position determined as the analysis position is uniform to a comparatively deep position. Therefore, elemental analysis is performed by performing analysis at this analysis position. Can be performed accurately in a short time, and thus the efficiency of analysis can be improved.

If the analysis is automatically performed based on the analysis position determined by the present invention, the elemental analysis can be automated.

[Brief description of drawings]

FIG. 1E to which an analysis position determining method according to the present invention is applied
It is a figure which shows the structure of one Example of PMA.

FIG. 2 is a diagram for explaining how to create a scatter diagram showing the relationship between the concentrations of elements.

FIG. 3 is a diagram for explaining creation of a scatter diagram by principal component analysis.

FIG. 4 is a diagram for explaining clustering.

FIG. 5 is a diagram for explaining a process of obtaining the concentration of each element at the center of gravity of the scatter diagram.

FIG. 6 is a diagram for explaining a conventional problem.

[Explanation of symbols]

1 ... Scanning electron microscope, 2 ... Deflection coil, 3 ... Sample, 4 ...
X-ray detector, 5 ... X-ray analyzer, 6 ... Scanning circuit, 10 ...
Control device, 11 ... CPU, 12 ... X-ray image memory, 13 ...
Scatter plot memory, 14 ... Input device, 15 ... Monitor.

Claims (1)

[Claims] 1. A concentration map is created for each designated element of a sample, a scatter diagram is created based on the map, and the concentration value of the element at the barycentric position of the cluster appearing in the scatter diagram is determined, and the determined concentration is calculated. A method for determining an analytical position, characterized in that the analytical position is a position on the map where a value is obtained.