JPH02291993A - Method for eliminating count loss effect of X-ray counter and linear amplifier - Google Patents

Abstract

PURPOSE:To achieve elimination of counting loss effect to be generated between other X-ray counters due to a dead time by making a waveform shaping time constant variable continuously at a linear amplifier of a radiation counter. CONSTITUTION:At a linear amplifier, a stepped input signal is shaped in waveform into a Gauss type through a differentiation circuit and an integration circuit to be make time constants of the differentiation circuit and the integration circuit between 0.5 - 1.5mus. When a local structure of material is clarified by measuring a fine vibration (EXAFS) of an X-ray absorption coefficient near an X-ray absorption end of the material, a linear amplifier thus obtained is used for a 2-counter type EXAFS measuring device as I counter to make dead time of a counter system variable continuously. If so, a data glitch of an X-ray absorption coefficient curve is eliminated completely, thereby enabling obtaining of accurate information on a distance between atoms and a coordination number of atoms from a sample material.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention is directed to the micro-oscillation of the X-ray absorption coefficient near the X-ray absorption edge of a substance.
ion [jineStructure (hereinafter referred to as 8XA
It is concerned with X-ray structure analysis equipment that elucidates the local structure of materials by specifying FS (FS)], and is particularly concerned with the counting loss effect of X-ray counters suitable for structural analysis of amorphous, quasicrystal, etc. This invention relates to a solution method and a continuously variable linear amplifier with a waveform shaping time constant.

(Prior art) Figure 10 shows a conventional basic angle dispersion laboratory system BX.
FIG. 3 is a configuration diagram of an AFS measurement pedestal. In the same figure, the thickness t
The sample has energy E1 intensity■. When the X-ray (E) is incident and the intensity of the X-ray transmitted through the sample is I (E), the relational expression with the absorption coefficient μ(E) is defined as follows. X-ray tube (
The white X-rays emitted from the spectroscopic crystal ('MC) are made monochromatic and used as incident X-rays to the sample (SP).
Obtain the linear intensity Io(E) (blank measurement), then mount the sample and scan again in the same angular range to obtain the sample transmitted intensity I
(Obtaining E> (sample measurement) Figure 9 shows the amorphous 'gto2
n3. Incident X-ray intensity curve near the 2n K absorption edge of the alloy (
(lower part) and the X-ray absorption coefficient curve (upper part) obtained from equation (1). In the measured incident X-ray intensity curve, sharp characteristic X-rays WLβ1, WLβ2, and WLβ of tungsten, which is contained as an impurity in the molybdenum of the anti-shading metal of the X-ray tube, can be seen, and the absorption coefficient curve shows their energy positions. Noise, called data glitch due to unnatural fluctuations, (G.,
G2, GO) is appearing. The energy of certain characteristic X-rays due to various impurities contained in the anticathode of such X-ray tubes may be located within the range of EXAFS measurements, and local structures around the X-ray absorbing atoms, especially This has been a major obstacle to obtaining accurate information such as the distance between adjacent atoms and the coordination number of atoms. The causes of the data glitch mentioned above are: (1) Discrepancies in the angular position of the goniometer due to two scans (blank measurement and sample measurement) (reproducibility), (2)
This is due to counting loss due to dead time in the X-ray counter system caused by the high intensity of characteristic X-rays, and unstable fluctuations in the X-ray source intensity.

One of the inventors developed I in one angular scan without removing the sample, with the aim of obtaining a smooth absorption curve without data glitches. A two-counter type εXA with a structure that can measure the X-ray intensity of , and also correct the counting loss.
We developed an FS measuring device and proposed it in Japanese Patent Application No. 7780/1983.

FIG. 8 is a configuration diagram of the two-counter type IEXAFS measuring device. In the same figure, an anticathode type MO tube (XS)
The white X-rays generated at an accelerating voltage of 20 KV and a tube current of 40 mA pass through a Soler slit (33) for diffusion suppression and a diverging slit (DS), and are sent to a goniometer (GM).
A diffraction surface installed on the central axis (plane spacing 2.013 people (
A flat LiF spectroscopic crystal (M
C) is irradiated. The separated monochromatic X-rays are introduced into a detector system placed on a counter arm (CA) that rotates around the central axis of the diffraction surface. In the detector system, after the diffusion is suppressed again by the Soler slit (SS) and the light receiving slit (RS), the first transmission-type Io counter (C
1) The incident X-ray intensity to is monitored. ■．
The X-rays that have passed through the counter are in a vacuum chamber and are exposed to liquid N2.
Sample cooled by thermal conduction of Dewar (CS) (S P)
, and then the second sample transmitted X-ray intensity ■ becomes the second
is detected by the I counter (C2). This I counter has a partition wall that allows the detection area (effective length) of the counter to be continuously varied.
Linear absorption rate can be adjusted. X generated in I0, I counter
In the counting system and control system, the linear photon pulse is subjected to impedance conversion as a detection signal in a preamplifier (PA), and is subjected to Gaussian waveform shaping in a linear (main) amplifier (MA). The height of the output pulse of the main amplifier fluctuates in proportion to the X-ray photon energy, and this output pulse is
SA), where pulses due to harmonic X-rays from the spectroscopic crystal, natural cosmic rays, preamplifier noise, etc. are removed. The output pulses of the pulse height discriminator are counted by a 3-channel scaler (ST). The third channel of this scaler has a preset function and is based on the constant counting method (Fi
xed Count Method). Measured at each angular position θ! . counter count value,
The count value of the I counter and the time required for measurement are stored in the memory of the microcomputer (μCP) after the count is completed, and then a command pulse is sent from the microcomputer to the goniometer, and the goniometer is set at a predetermined angle Δθ. and is sent to the next angular position θ+Δθ.

In other words, the above-mentioned device has the following features: Eliminate angular position deviation for each angular scan of the goniometer, measure Io and ■ at once to avoid being affected by fluctuations in X-ray source intensity, do not move the sample, It was intended to eliminate data glitches by accurately correcting the count loss of the line detector.

In particular, by moving the partition wall of the I counter that can continuously vary the detection area (effective length) for sample-transmitted It was intended to eliminate data glitches by eliminating the effects of loss.

(Problem to be solved by the invention) However, in the two-counter EXAFS measurement described above, when it is necessary to use a semiconductor detector with excellent energy resolution as the I counter (mainly the absorption edge is on the low energy side, In addition, although it is originally desirable for the detection efficiency of the counter to be 100%, it is necessary to satisfy the matching condition of counting loss characteristics. However, there was a drawback that the detection efficiency of the I counter was forced to drop below 100%. In order to overcome these drawbacks, in order to obtain the maximum detection efficiency of the aforementioned effective length variable proportional counter, if you try to use it at its longest length,
The problem of count loss effects due to dead time of the line detector system will arise again.

The dead time of the X-ray detector system will be explained below. 7th
The figure is a flowchart of electrical signals in a detector system using the proportional counter described above. A small signal from the preamplifier (PA) is amplified by the main amplifier (MA) and shaped into a Gaussian waveform. At this time, when the pulse interval is sufficiently long (case (a)), each pulse is counted as an independent pulse.

However, when the pulse interval is short (case (C)), the pulses overlap and only one pulse is counted.

This is counting loss. The dead time of the detector system is the minimum pulse interval τ (in case (b)) at which two pulses are counted independently. this is,
It is approximately the same as the pulse width.

The time constant of a typical commercially available main amplifier is 1μSN
Since it can only be changed discretely, such as 2 μs...
When used in a two-counter type BXAFS measurement device, two counters are used when an e-counter such as a semiconductor detector is used, or when the effective length variable I counter is used with the counter length fixed at the maximum length. There was a problem in that it was not possible to satisfy the matching condition of the counting loss characteristics.

The present invention improves the pulse width of electrical signals by continuously varying the waveform shaping time constant of the linear amplifier of the X-ray counter.
In other words, characteristic X-ray The purpose is to obtain a smooth X-ray absorption coefficient curve that is not affected by

(Means for solving the problem) The above problem is solved in the linear amplifier of the radiation counter.
This can be achieved by configuring the waveform shaping time constant to be continuously variable.

(Function) In the two-counter type IEXAFs measuring device shown in Fig. 8,
■. ■ The actual counted value of the counter. , its true value (
(value for ideal case with no counting loss). 'The actual value of the I counter is I1 Its true value is 11R=IO/I, Rt=
l. If t/It, then R=R' + (r-Rt
ro) Io...(2) holds true.

Here, ro and τ are each ■. The dead time of the counter and the I counter has the following relationship: Iot = Io/ (I Ioτ.) I' = I/ (1-Iτ). The second term on the right side of equation (2) above represents the counting loss effect.

The already developed X-ray counter can be adjusted by adjusting the effective length of the counter, that is, Rt=τ/r 6
, and set the coefficient of Io in the second term on the right side of equation (2) to (rRtro)=0 to prevent data glitches from occurring. The present invention
The coefficient of I0 in the second term on the right side of equation (2) above is (τ−Rt
A method for setting Z''o)=O is to continuously change τ or τ instead of Rt to prevent data glitches due to dead time from occurring.

Gaussian waveform shaping of a linear amplifier is performed by a differentiating circuit and an integrating circuit, and the waveform shaping time constant is the differentiating/integrating circuit time constant. Usually, this time constant is determined by a circuit configuration that combines a fixed resistor and a fixed capacitor. In the present invention, these are configured by, for example, a variable resistor and a variable capacitor, and the time constant is made to be continuously variable, for example, between 0.5 μs and 1.5 μs. As a result, the waveform shaping time constant of the linear amplifier for the X-ray counter, that is, the dead time of the counter system can be continuously varied, so the two-counter type EX
When applied to an AFS measuring device, the matching condition of counting loss characteristics (Rt=τ/τ) can be satisfied even if any X-ray counter is used as the second count.

(Effects of the Invention) According to the present invention, the waveform shaping time constant of the linear amplifier for an X-ray counter and the dead time of the counter system can be continuously varied. Data glitches in the linear absorption coefficient curve can be completely eliminated.

Therefore, it has become possible to obtain accurate information such as the distance between atoms and the coordination number of atoms from the sample substance. Furthermore, in order to completely remove high-order harmonics, it is also effective when it is necessary to use a semiconductor detector with excellent energy resolution.

The present invention is suitable for analyzing the atomic structure of non-quality materials, quasicrystals, liquid solutions, etc., and therefore can be expected to be used for the development of new materials such as various ceramics and amorphous alloys.

(Example) The present invention will be explained in detail below.

The present invention uses the waveform shaping time constant continuously variable linear amplifier of the present invention as the linear amplifier (MA') for the I counter in the two-counter EXAFS measuring device shown in FIG.
A normal time constant discrete linear amplifier (MA) is used for the Io counter.

The first block diagram of the general linear amplifier circuit of the present invention is shown below.
As shown in the figure. The step-like human input signal is shaped into a Gaussian waveform through a differentiating circuit and an integrating circuit. The time constant of waveform shaping is the time constant of the differentiating circuit and the integrating circuit. This is usually determined by the resistance and capacitance used in the differential/integrator circuit. In conventional commercially available linear amplifiers, the resistors and capacitors are fixed, so the time constant can only be changed discretely (discontinuously). In the present invention, this is a variable type, and the time constant is 0. It was designed to be able to be changed continuously between 5 and 1.5 μs.

Next, a method to clarify the relationship between the waveform shaping time constant of the linear amplifier and the dead time of the X-ray counter system will be described.

Normally, the dead time of an X-ray counter system is approximately determined by the waveform shaping time constant of a linear amplifier, so it is expected that this time constant and dead time are proportional. First, the effective length of the counter is set to its maximum value (30ca+) and fixed at that length.

The variable time constant type of the present invention is used as an I-counter type linear amplifier. This time constant is set to an arbitrary value between 0.5 and 1.5 μs, X and gravel are actually made incident on the I counter, and the dead time is determined by the single thin film method.

Here, a method for determining the dead time using the single thin film method will be explained. A conceptual diagram showing the actual situation is shown in Figure 6A and Figure 6.
Shown in Figure B. Let the intensity of X-rays incident on the thin film be 10, and the intensity of X-rays transmitted through the thin film be i.

In reality, the thin film is removed from the X-ray beam bath (
The X-ray intensity is measured by an X-ray counter (a counter for determining the dead time, in this case an I counter) (Fig. 6A), and this counting rate is defined as i.

Next, insert the thin film into the X-ray beam bath (Figure 6B) and
Measure the line intensity and let it be i. The dead time of the counter system is rd, the true value of the actual measurement value 10 (the ideal case when τ' = 0, that is, the value when there is no counting loss) is 1OL, and the true value of the actual measurement value l. 11, then 1ot-io/(l loτd) i' = i/(1-ir') (7) There is a relationship. Furthermore, if r = lo/1 s rt = i0t/it, r=r'+ (1-r') r'i., holds true.This means: ■Measure io and i while changing the incident X-ray intensity,
A straight line can be obtained by plotting the actual measurement @r=io/i against the actual measurement value il1, ■ r1 can be found from the extrapolated value of the straight line to l o - 0, and ■ the slope of the straight line is (1 −
r') τ6, indicating that 7d can be found from this slope and the value of rt. In this example, the thin film is ^β(
(thickness: 100 μm), and the incident X-ray energy was set to 9.9
The experiment was conducted at 6 keV (energy of WLβ2). Figure 5 shows the four waveform shaping time constants of the linear amplifier, sh
( , sh= 0.5, 0.8, 1.2, 1.5
μs>. An example of plotting rVSjo is shown.

However, in FIG. 5, the vertical axis is r/rt normalized by rt (=2.62). Each τ6 can be found from the slope of the straight line in FIG. In this way, τ6 is found for each τ゛7, and τ'vs. A τ゛h plot is obtained. FIG. 4 shows the τ' vs, τ1 plot obtained in this example. From this figure, τ4 is proportional to τ1, and r
' (μs) = 5.5 rsh(, us). As a result of the above, the waveform shaping time constant τ1 is set to 0.5~
If it is changed to 1.5 μs, the dead time τ6 will be 2.7 to 8.
It was found that the time period changes to 3 μs.

Next, the linear amplifier of the present invention is made of amorphous Mgt. 2n. . An example of application to 8XAFS measurement of alloys is shown. The experimental conditions are as follows. In the two-counter EXAFS measuring device described above, the length of the variable effective length I counter is fixed at its maximum value (30 mm). The variable time constant amplifier of the present invention is used as the main amplifier for the I counter. The main amplifier of the I0 counter uses a normal time constant discrete type amplifier, and its time constant is 0. Fixed at 5 μs or 1 μs. In the two-counter EXAFS measuring device, as described above, R=RL-(R'-κ)τ. 6th grade. , holds true. Here, κ== r d / r od. τ6ro
d is rsh1 τ, respectively. Since it is proportional to 1, κ
= τl h / r0% h. In this example, τ.

. is fixed and τ1 is continuously variable, so κ is continuously variable. Non-quality in Figure 3! Jgto2n3. In the alloy sample, the incident X-ray energy was set to 9.96 keV (W
Lβ2), three values of κ (κ=0.6, 1.
2, 3.0), whereas RVS. An example of plotting Io is shown.

However, in FIG. 3, the vertical axis is normalized by Rt (=1.2). As can be seen from this figure, when κ=0.6, r becomes smaller as in becomes larger, and κ=3.
When it is 0, it is the opposite. As R becomes larger, R becomes larger. These reflect the fact that the count loss characteristics of the two counters do not match, and Rt>κ and RL<κ, respectively. When κ = 1.2, the value of R remains constant (R' = 1.2) even if Io increases, which indicates that the count loss characteristics match. . Therefore, if EXAFS measurement is performed under the conditions of κ=1, 2, data glitches due to counting loss can be prevented from occurring. In fact, amorphous'g 7
EXAFS on ZnK absorption edge of 0 l n 3 3 alloy
Measured examples are shown in FIGS. 2A and 2B. When κ = 0.6, a valley-shaped data glitch occurs, when κ = 3.0, a mountain-shaped data glitch occurs, but when κ = 1.2, no data glitch occurs. .

As described above, it has been shown that the waveform shaping time constant continuously variable linear amplifier of the present invention is extremely effective in completely eliminating data glitches in a two-counter BXAFS measuring device.

[Brief explanation of drawings]

FIG. 1 is a circuit block diagram of a linear (main) amplifier according to the present invention, and FIG. 2A is an amorphous Mg7.2n, .
The X-ray absorption coefficient curve near the 2nK absorption edge of the alloy, FIG. 2B is the incident X-ray intensity curve according to the example, and FIG. 3 is the amorphous Mgt. Zn3. x == at the position of WLβ2 of the alloy
7 Sh/r. RVS for S h, Io
Figure 4 shows the plot of rdVS. determined by the single thin film method. A plot diagram of rSh, Figure 5 is a plot diagram of rVS,io performed for the four waveform shaping time constants of the main amplifier (r' = 2.6
2), Figures 6A and 6B are conceptual diagrams of the single thin film method, Figure 7 is a flowchart of electrical signals in the detector system when using a conventional two-counter EXAFS measuring device, and Figure 8 9 is a block diagram of a two-counter EXAFS measuring device. FIG. 9 shows non-quality Mgt obtained using the angle dispersive εXAFS measuring device of FIG. 2130 alloy ZnK
Figure 10, a diagram showing an incident X-ray intensity curve (lower side) and an X-ray absorption coefficient curve (upper side) near the absorption edge, is a configuration diagram of a conventional basic angle dispersion type EX8FS measuring device. (Explanation of Sasa No.) XS...XL tube, SS...Solar slit,
DS: Divergent slit, MC: Spectroscopic crystal,
G1...coniometer, RS...light emitting slit, C1...Io counter, C2...I counter, CS...N2 dewar SP...sample, CA
... Counter arm, P A ... Preamplifier, MA, MA' ... Linear (main) amplifier, ST ...
Channel Kohler μCP...Microcomputer. Fig. 3 χ1τsh/1:05h 10 (cps) 10r Faction Fig. 4 τSh (μS) Faction Fig. 7 Time I Fig. 9 × 105 9.8 10.0 10.2 E (keY) Fig. 8 Figure 10

Claims (2)

[Claims] (1) An X-ray counter characterized in that the waveform shaping time constant of the linear amplifier of the X-ray counter is continuously changed to eliminate the count loss effect caused by dead time between the X-ray counter and other X-ray counters. How to eliminate counting loss effect. (2) In a linear amplifier incorporating a differentiating circuit and an integrating circuit for shaping the waveform of a signal from an X-ray counter, continuously variable elements are used as circuit elements constituting the differentiating circuit and integrating circuit. A linear amplifier characterized in that the waveform shaping time constant can be changed continuously.