JPS647337B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for simultaneously measuring cracks and inclusions, which measures a wide two-dimensional distribution of cracks and inclusions in a slab sample using a large-diameter electron beam.

In the sulfur print test method, a photographic paper soaked in sulfuric acid solution is attached to the surface to be tested (for example, a cross section of a slab) in order to detect the segregation and distribution state of sulfur (S) in steel materials. Since the smoothness is required to be on the same level as No. 1, it has the drawback that polishing requires a large amount of time. Moreover, only sulfur and phosphorus can be directly detected with sulfur print, and other inclusions cannot be directly observed. Among inclusions other than sulfur, aluminum (Al), manganese (Mn), etc. are alumina (Al ₂ O ₃ ) or sulfur compounds (e.g. MnS) that behave in the same way as sulfur.
This was estimated empirically from the distribution pattern of sulfur prints, but as the content of sulfur in steel materials has decreased in recent years, these measurements tend to become increasingly difficult.

On the other hand, there is an X-ray microanalyzer (EPMA) as a device that directly measures various inclusions on a surface to be inspected. This method irradiates a sample with an electron beam with a very narrow beam diameter (1 to 100μφ), detects the X-rays emitted from inclusions, and measures the elements and amounts of inclusions two-dimensionally. However, since this is an X-ray focusing method that uses a curved crystal in the X-ray spectroscopy system, the Rowland circle (focusing circle) is located below the sample position. Therefore, the size of the sample is limited, and only about 20 x 20 mm ² can be analyzed. Therefore, the scanning range is narrow, from 1 to several millimeters, so it is not suitable for obtaining a wide distribution pattern like a sulfur print, and the surface to be examined requires flatness close to that of an electron microscope sample. It is similar to sulfur print in that it takes time to polish. In addition, when directly measuring inclusions on a cross section of a slab by irradiating an electron beam, if cracks (general term for defects including pinholes, alumina clusters, etc.; the same applies hereinafter) occur in the cross section of the slab, these can be detected as inclusions. becomes noise in elemental analysis.

The present invention has been made in view of the above circumstances, and even though the polishing of the cross section of the slab as the surface to be inspected is significantly simpler than that of the conventional sulfa print method, it still produces the same inclusion distribution pattern as sulfa print. Furthermore, it is possible to directly measure a wide range of distribution patterns of other inclusions regardless of the content of sulfur, and furthermore, cracks can be simultaneously detected by irradiation with an electron beam for elemental analysis, and this detection signal can be to ensure elemental analysis and
Since cracking is an important factor in the quality of steel materials, we have made it possible to use this as cracking information.The feature is that the electron beam, with a beam diameter selected in the range of 0.1 to 10 mmφ, is used to continuously cast slabs. A cross-sectional sample is irradiated in vacuum, the X-rays emitted from the sample upon irradiation with the electron beam are separated by a flat crystal, and the intensity of the characteristic X-rays unique to each element is detected by the flat crystal. The amount of iron X-rays emitted when the slab sample is irradiated with the electron beam, or the amount of reflected electrons or absorbed electrons of the electron beam is detected by a detector, and the amount of electrons relative to the slab sample surface is detected by a detector. The point of this method is to simultaneously measure the two-dimensional distribution of cracks and inclusions in the slab by changing the beam irradiation position two-dimensionally.

Hereinafter, the present invention will be explained in detail with reference to the drawings. Figure 1 shows the configuration of an apparatus for explaining the outline of the present invention.
It is kept in a vacuum container except for the electron optics. Reference numeral 2 designates a sample stage that is moved in the X and Y directions by pulse motors 3 and 4, and a sample 5 is placed on its upper surface. Sample 5 is, for example, a cross-section of a CC (continuously cast) slab with a width (Y direction) of about 30 cm and a length (X direction) of about 2 m (collected by gas cutting and the surface milled). electron gun 6
The beam diameter is selected within the range of 0.1mm to 10mm from
An electron beam EB with a deep focal depth is irradiated intermittently (or continuously). 7a to 7d are X-ray spectrometers that use flat plate crystals as spectroscopic crystals, which detect characteristic X-rays of Al, P, S, and Mn, respectively, and convert them into electrical signals (one pulse corresponds to one X-ray quantum). pulse train). 8 is a high voltage circuit that applies a high voltage for acceleration to the electron gun 6, and its output is confirmed through the high voltage control system 10. Electron beam
The beam diameter of the EB is controlled by a beam control circuit 11,
Further, the degree of vacuum within the apparatus 1 is confirmed through the vacuum state detection section 12. The X-ray spectrometers 13a to 13d are counters for counting output pulses, and the counting period thereof is defined (gate-controlled) by the gate control circuit 14. pulse motor 3,
4 are driven by the outputs of the X-axis pulse generation circuit 15 and the Y-axis pulse generation circuit 16, respectively, to change the electron beam EB irradiation position on the sample 5. Scanning in this example is mechanical in both the X and Y directions, but electron beam
Since it is possible to deflect the EB by about 30 cm, the Y-axis direction may be electrically scanned. 17 is an alarm circuit, and if any abnormality occurs, lamp 9 will be activated.
lights up.

20 is a data processing system for control analysis;
Microcomputer 21, minicomputer system 22 and graphic display 23
Consisting of The minicomputer system 22 includes a minicomputer main body 24, an analysis calculation memory 25,
A data bus 2 is connected between the main body 24 and the microcomputer 21 and between the memory 25 and the display 23.
Connected at 7,28. 29 is a line for control signals.

Figures 2, 5, and 7 are different examples showing the main parts of the measuring device 1.
The cross section is taken from the state shown in FIG. 1 where the light is irradiated. In FIG. 2, the X-ray spectrometer 7c (7a, 7
b, 7d) consists of a plane crystal 30 that spectrally spectra the X-rays emitted from the sample 5, an X-ray detector 31 that converts the spectroscopic X-rays into electrical signals, and an amplifier 32 for the output thereof. Some X-ray spectrometers use a condensing spectroscopy method that uses a curved crystal, which is used in X-ray microanalyzers, etc.
In this method, the light focusing conditions are strict and a slight positional shift has a large effect on the measurement results, so it is difficult to apply it to the large samples that the present invention targets, although it is difficult to apply it to the large samples targeted by the present invention. If a flat crystal is used, even if the distance to the sample surface changes by about 10 mm, the spectral X-rays will only shift by about 10 mm and will fall within the range of the X-ray counter opening, so no measurement errors will occur. The electron beam EB irradiated on sample 5 is the iron (Fe) of sample 5, as well as sulfur (S) and phosphorus (P).
The characteristic X-rays Fe-Kα, S
−Kα, etc. are generated. S-Kα is X-ray spectrometer 7c
, and Fe-Kα is detected by an X-ray spectrometer 7e comprising a plane crystal 34, an X-ray detector 35, and an amplifier 36. The X-ray spectrometer 7c has inclusions 33
The X-ray spectrometer 7e is used to detect a crack 37 of approximately 1 mm in size that has occurred in the sample 5.
When the sample 5 is sequentially scanned from the left end with the electron beam EB, the output C of the X-ray spectrometer 7c changes as shown in FIG. 3, showing a peak C ₁ at the position of the inclusion 33.
At the position of the crack 37, a peak C ₂ of opposite polarity appears. ba indicates the background level. C ₂ is primarily a noise in the analysis, so the X-ray spectrometer 7e
The crack information A ₁ contained in the output A of is removed (the level is significantly reduced). There is also a level change in the output A that corresponds to the inclusion 33, but this is slight, so by waveform shaping the output A, a binary level crack signal B corresponding to the presence or absence of the crack 37 can be obtained.

The reason why the plane crystal 30 is used as the spectroscopic crystal of the X-ray spectrometers 7a to 7d is as follows. In other words, in order to increase the sensitivity of an X-ray microanalyzer, it is customary to use a spherical focusing crystal for the spectroscopic crystal, but in this type of spectroscopic crystal, the focal point and the electron beam irradiation position are not always placed in a constant relationship. Since this disturbs the spectral conditions, elemental analysis can only be performed by scanning the sample by moving it in the X and Y directions with extremely strict accuracy and thus by a very small distance. Since the sample of the present invention is large, such exact precision is not economically achievable, and in order to simplify the movement mechanism, scanning in either the X or Y direction may be performed by deflecting the electron beam. It is inappropriate to use a spherical condensing spectrometer. Furthermore, although the plane crystal type has lower sensitivity than the spherical crystal type, in the present invention, since a large diameter beam is used, the amount of X-rays emitted is large, so even if a flat crystal is used, lack of sensitivity will not be a problem. Note that since the flat crystal is placed facing the sample surface at the angle shown in FIG. 2, etc., there is no problem even if the irradiation position moves on the sample as long as the scanning direction of the electron beam is selected appropriately.

FIG. 4 shows an example of a circuit for processing outputs A and B. Signal A passes through a pulse height discriminator (and rate meter) 40 and is converted into a cracking signal B by a comparator 41.
On the other hand, the signal C passes through the pulse height discriminator 42 and is measured by the scaler timer 43, but at this time the cracking signal B
The cracked part C ₂ is removed.

Figure 5 shows the incidence of incident electrons when irradiated with electron beam EB.
The backscattered electrons RE, which appear at about 30%, are detected by backscattered electron detectors 44 and 45 (although only one may be used), and the output thereof is amplified by amplifiers 46 and 47 and then combined.
A cracking signal B is obtained from cracking information (deep level changes) included in the sum of the outputs D and E of these amplifiers. In this example, cracks are detected by utilizing the fact that the amount of reflected electrons RE is significantly reduced at the crack 37 portion. FIG. 6 shows signal waveforms at each part.

Figure 7 shows sample 5 when irradiated with electron beam EB.
This is an example in which absorbed electrons absorbed by the resistor 48 are detected by an absorbing current resistor 48, and the voltage across the resistor 48 is amplified by an amplifier 49. In this case, the output G of the amplifier 49 shows a peak _G1 when the number of absorbed electrons increases at the position of the crack 37, as shown in FIG. 8, so that a crack signal B can be obtained using this as crack information.

FIG. 9 is a further development of FIG. 4, in which a crack detector 60 represents each of the crack detection means shown in FIGS. 2, 5, and 7, and a crack discrimination circuit 61 that discriminates the output thereof. The output becomes the cracking signal B. timer 62
is used to start counting and reset the counter 63 with the repeated output T to determine its steady counting period.However, since the output T is combined with the crack signal B in the NAND gate 64, the counter 63 is Counting does not start during the period when signal B exists, and is cleared by crack signal B. FIG. 10 shows the waveforms of each part, and J is the output of the counter 63. In this output J, the shaded area indicates the count value according to the intensity of the X-ray signal C, but the circle area is the area where the count is cleared and counting is stopped due to the presence of the crack signal B. This is effective for noise removal. The output J of the counter 63 is taken into the computer 65 together with the crack signal B, and at this time a position signal P indicating the position where these signals J and B were obtained is added as an address. IN in FIG. 10 shows its storage format. In this way, the position, crack information, and X-ray elemental analysis results are stored side by side in the memory, so the operator can read them out and make various decisions. The position signal P is X, Y when moving the sample stage 2 in Fig. 1.
This is coordinate information, and in the case of scanning the electron beam EB, it is sawtooth deflection voltage information. This position signal P is also used as a drive signal for the timer 62 for synchronization. In addition, 66 in Fig. 9 is an amplifier (buffer).
The computer 65 is a generalized version of the microcomputer 21 and minicomputer 24 shown in FIG.

2 to 10 are explanatory diagrams of the portion of FIG. 1 excluding the data processing system 20, that is, the processing steps up to obtaining the computer input IN. computer input
IN is obtained two-dimensionally for the entire region of the sample 5 or a partial region thereof at a required scanning density for each element, and is processed by the data processing system 20. FIG. 11 is a block diagram showing the processing system 20 in detail. Computer input IN from the measuring device 1 is stored in image memories 70a and 70d for each element. The standard value image memories 71a to 71d paired with the image memories 70a to 70d include a disk image file 26 or a drum type scanning image input terminal 7.
Standard values from 2 are selectively stored. Data from the input terminal 72 is obtained by converting a known sulfur print pattern or the like into a digital signal by an A/D converter 73, and is guided through a switching section 74 to image memories 71a to 71d. The control unit 75 switches, synthesizes, and superimposes the contents of the actual measurement data memories 70a to 70d and the standard data memories 71a to 71d, and provides the output to the color image processing unit 76 for display on the display unit. Displayed at 23.
While viewing this display, the operator can identify, for example, which of the standard samples the sample corresponds to, and automatic identification is also possible using arithmetic logic.

FIG. 12 shows image memories 70a to 70n (11th
This is a block diagram mainly showing a configuration for associating an address (n=d) with the irradiation position of the electron beam EB on the sample 5. Pulses from a reference clock source 80 are guided through a direction control circuit 81 to an X drive circuit 82 and a Y drive circuit 83. Drive circuit 8
2 and 83 are the X-axis and Y-axis pulse generation circuit 1 in Figure 1.
5 and 16, and drive the pulse motors 3 and 4, respectively, but at the same time, the X address counter 84 and the Y
The address counter 85 counts the number of drive pulses. This number of drive pulses corresponds to the distance from the origin of the X and Y axes, so the counters 84 and 85 at each time
The contents are the X and Y coordinates indicating the irradiation position of the electron beam EB. The X address conversion circuit 87 and the Y address conversion circuit 86 convert the contents of the counters 84 and 85 into codes suitable for indexing the image memory. This is the first
This becomes the position signal P shown in Figure 0. Image memory 70a-7
Elemental analysis values (S)...(Al) are sequentially derived from 0n. This corresponds to the X-ray signal J in FIG. At the same time, the crack signal B is also written using the position signal P as an address. Arithmetic unit 88 in FIG.
and the color control unit 89 is the computer 6 in FIG.
5. A block that consolidates the functions of the control section 75 and color image processing section 76, and various processing contents can be considered.

FIG. 13a shows a typical pattern of a cross section of a CC slab, and each crack α to ε can be classified into five discrimination zones shown in FIG. 13b. The zone is a portion where so-called sulfur bands are formed, and alumina clusters β or lateral cracks γ occur here. The zone is a vertically symmetrical part centered on the zone, and a pinhole α occurs here. The zone is a portion that is symmetrical left and right around the zone, and a diagonal crack δ or a lateral crack ε occurs here. These flaws and cracks are detected at the same time as the elemental analysis of the inclusions, and the data processing system performs various processing according to the operation flow shown in FIG. 14, for example. The treatments shown in the figure are roughly divided into those related to cracks and those related to inclusions. Regarding flaws and cracks, after determining to which zone they belong, a determination specific to each zone is made. That is, pinhole discrimination (α) is performed for zones, types β and γ are discriminated for zones, and δ and ε are classified for zones by angle discrimination. After the classification of various types of cracks is completed, the degree of cracking is classified and recorded in a computer file, and an evaluation is added to the sample to be measured. On the other hand, a two-dimensional distribution for each element is created according to the distribution information input for each element, and after calculation is performed in combination with crack information, color display control is performed to display the color on the display unit 23, and if necessary, the hardware Take a copy.

The display form of the display unit 23 in FIG. 12 depends on the processing contents of the calculation unit 88 and color control unit 89. The calculation unit 88 aligns the image memories 70a to 70n and the image memories 71a to 71n. For example, as shown in FIG.
Two-dimensional image pattern based on the content of 70n
A difference in scale between GP ₁ and the two-dimensional image pattern GP ₂ based on the contents of the image memories 71a to 71n;
If there is a difference in rotational position, etc., (a) make the geometric scales of both the same, (b) correct the vertical, horizontal, and rotational positions, and (c) minimize the correlation between image patterns. Match it to suit your needs. GP′ ₁ changes the scale of GP ₁ ,
GP′ ₂ is a modified image pattern obtained by changing the rotational position of GP ₁ , and there is no positional shift between the two patterns even if they are displayed on the same screen of the display unit 23 in a superimposed manner. When superimposing and displaying the images GP' ₁ and GP' ₂ , the color control unit 89 displays a common part a between the actual measurement value and the standard value, and a part b where the difference between the two is positive with respect to one (for example, the standard value). and negative portion c, and these are displayed in different colors. Figure 16 is an example of this, with a, b,
c are colored differently. The display in the figure is switched for each element of inclusions, so in addition to sulfur (S), phosphorus (P) and aluminum (Al)
Inclusions such as manganese (Mn) can be directly observed, but as far as Sulfur is concerned, the display pattern in Figure 16 shows the difference from the known standard pattern of Sulfur prints, so it is possible to judge the quality of Sulfur 5 based on this. Can be distinguished. At this time, the control unit 75 (FIG. 11)
If only the actual measured values are displayed by switching, it becomes a distribution pattern equivalent to the sulfur print. This also applies to other elements.

There are various other possible display formats for the display section 23. FIG. 17 shows this. Figure a is the same as Figure 16, so the explanation will be omitted. Figure b shows the maximum value (Speak) and minimum value (Smin) of the Sulfur band in the width direction, and Figure c shows the length of the Sulfur band.さ (Se) is displayed. In addition, figure d shows the concentration distribution of Sulfur S along the center line of the Sulfur band, and figure e
shows the concentration distribution of sulfur S in the cross-sectional direction along the line crossing the sulfur band. In addition to this, the total area of the sulfur band portion can also be displayed. These displays are intended only for the operator to judge the quality of Sample 5, and can be selected from the operator console. It is also possible to perform automatic quality classification.

As mentioned above, a large-diameter electron beam with a beam diameter selected in the range of 0.1 to 10 mmφ is irradiated onto the cross section of the slab, the cross section is scanned, and the characteristic X-rays from the inclusions are detected after being separated by a plane crystal. At the same time, the iron X that is irradiated when the electron beam is irradiated on the slab cross section is
With the simultaneous measurement method of the present invention, which detects cracks occurring in the slab cross section from the amount of reflected or absorbed electrons of the electron beam (Fe-Kα) or the electron beam, (1) the electron beam flux is thick, so The smoothness only needs to be relatively rough, so the time it takes to obtain the smoothness of the test surface required for sulfa print is that it takes a whole day just for polishing, and the measurements are not taken until the next day. On the other hand, according to the X-ray analyzer of the present invention, the time spent on polishing can be significantly shortened, so the entire measurement time can be extremely shortened. In addition, (2) even if the sulfur concentration in steel decreases, other elements can be directly measured, so trends in other elements that have been estimated through sulfur in sulfur print can be measured regardless of the sulfur concentration. Furthermore, (3)
Since scanning with a large-diameter electron beam requires only a short time to scan the entire surface, a wide measurement area of the sulfur print, which is virtually impossible to scan with a micro-diameter beam such as an X-ray microanalyzer, can be measured in a short time. (4) Since crack detection can be performed at the same time as elemental analysis of inclusions, it is possible to obtain crack information without positional deviation with a single scan, which not only eliminates noise in elemental analysis but also detects cracks and inclusions in the cross section of the slab. It is extremely useful for comprehensive quality judgment.

[Brief explanation of drawings]

Figure 1 is a schematic configuration diagram showing the entire device, Figure 2 is a configuration diagram of the main parts of the multi-element simultaneous analysis device shown in Figure 1, and Figure 3
The figure is a signal waveform diagram of each part of Fig. 2, Fig. 4 is a block diagram showing an example of the signal processing system of Fig. 2, and Fig. 5 is a main part showing another example of the multi-element simultaneous analysis device of Fig. 1. 6 is a diagram of the signal waveforms of each part in Figure 5, Figure 7 is a diagram of the main part configuration showing a different example of the multi-element simultaneous analysis device of Figure 1, and Figure 8 is a signal waveform of each part of Figure 7. figure,
Fig. 9 is a block diagram showing another example of the signal processing system from the multi-element simultaneous analyzer shown in Fig. 1, Fig. 10 is a signal waveform diagram of each part of Fig. 9, and Fig. 11 is the data processing shown in Fig. 1. Block diagram showing the system in detail, No. 12
The figure is a block diagram mainly showing the assignment of addresses to input information to the data processing system in Figure 1;
Figures 13a and b are cross-sectional views showing typical crack patterns in a CC slab cross section, and a schematic diagram of the same cross-section divided into three types of determination zones. Figure 14 is a cross-sectional view of crack identification and inclusion elements in a CC slab cross-section. A flowchart comprehensively showing the analysis, FIG. 15 is an explanatory diagram showing the alignment of the measured image pattern obtained by electron beam scanning and the standard image pattern stored in advance, etc.
Fig. 6 is an explanatory diagram showing an example of a color image display pattern in which the differences between the measured image pattern and the standard image pattern are displayed in different colors, and Figs. 17 a to e are explanatory diagrams showing various display forms regarding sulfur in the cross section of the slab. It is. 1... Metal material multi-element simultaneous distribution measuring device, 5...
...Sample (slab cross section), 6...Electron gun, 7a
~7d...X-ray spectrometer, 20...Data processing system, 23...Display (display), 30,3
4... Planar crystal, 31, 35... X-ray detector, 3
3...Inclusion, 37...Crack, 44, 45...Backscattered electron detector, 48...Absorbed current detection resistor, EB
...Electron beam, S-Kα, Fe-Kα...X-ray, RE...
...Reflected electrons.

Claims (1)

[Claims] 1. An electron beam with a beam diameter selected in the range of 0.1 to 10 mmφ is irradiated in a vacuum to a cross-sectional sample of a continuously cast slab, and the X-rays emitted from the sample by the electron beam irradiation are analyzed using a flat crystal. Then, the intensity of the characteristic X-rays unique to each element spectrally analyzed by the planar crystal is measured with an X-ray detector, and the intensity of the iron X-rays or the electron beams emitted when the slab sample is irradiated with the electron beams is measured. A detector detects the amount of reflected electrons or absorbed electrons, and the two-dimensional distribution of cracks and inclusions in the slab is simultaneously measured by two-dimensionally changing the electron beam irradiation position on the slab sample surface. A method for simultaneously measuring cracks and inclusions in slab samples using an electron beam.