JP2021051053A - Fluorescent x-ray analyzing device - Google Patents

Abstract

To facilitate evaluation of whether or not a device sensitivity curve is properly prepared in a fluorescent X-ray analyzer according to a fundamental parameter method.SOLUTION: A fluorescent X-ray analyzer includes quantification means using a fundamental parameter method. About each standard sample, the quantification means calculates a quantitative value of the content ratio of a component corresponding to a measurement element, using the measurement intensity, a device sensitivity constant, and a proportional coefficient multiplied by a mass fraction of the measurement element for calculating the theoretical intensity in a theoretical intensity expression. Further, about each component, the quantification means outputs the accuracy about the entire quantitative value owing to a graph showing a correlation between a standard value and the quantitative value and/or the standard value, the quantitative value, a quantitative error for each standard sample, and a set of used standard samples.SELECTED DRAWING: Figure 4

Description

The present invention relates to a fluorescent X-ray analyzer based on the fundamental parameter method.

Conventionally, a fluorescent X-ray analyzer that performs quantitative analysis is roughly classified into a calibration curve method and a fundamental parameter method (also referred to as an FP method). In the quantitative analysis by the calibration curve method, for the analysis of an unknown sample, a set of standard samples in which the content of the component is known as a standard value is used, and the content of the component and the fluorescence X of the measurement element corresponding to the component are used. As a correlation with the measured intensity of the line, a calibration curve formula represented by the following formula (1) is obtained. The component is an element or a compound, and the content of the component is generally represented by a mass percentage (mass%) including a standard value and a quantitative value of the content. When the component is an element, the element itself is the measuring element corresponding to the component, and when the component is a compound, the element representing the compound is the measuring element corresponding to the component.

_Wi = (AI _i ² + BI _i + C) (1 + Σα _j W _j )… (1)
_Wi : Content of component i I _i : Measurement intensity of fluorescent X-rays of the measurement element corresponding to component i A, B, C: Calibration curve constant W _j : Content of correction component j α _j : Correction component Matrix correction coefficient of j

Then, for the evaluation of the created calibration curve equation (1), a graph of the calibration curve represented by the following equation (2) is output as shown in FIG. 2 for, for example, the component Cr. Specifically, a graph of the calibration curve is displayed by a display or a printer. Here, the estimated reference value X _i of the component i is assumed that the content W _j of the correction component j is zero, that is, there is no matrix effect of absorption / excitation by the coexisting components on the component i. In this case, it is the content rate of the component i in _{the measured intensity I i.} In FIG. 2, the calibration curve is a linear equation (A = 0), the horizontal axis coordinates of the rectangular points are the estimated reference values of the standard sample, and the horizontal axis coordinates of the white circle points are the standard values of the standard sample (chemistry). Analytical value) is shown.

X _i = AI _i ² + BI _i + C ... (2)
X _i : Estimated reference value of component i

Further, for the standard sample, the measurement intensity I _i corresponding to the component i and the standard value of the component j as the content rate W _j of the correction component j are substituted into the calibration curve equation (1) to obtain the component i. obtains a quantitative value W ^ _i of component i as a content W _i, further by the following equation (3), the standard value of the component i as a content W _i of quantitative values W ^ _i and the true component i of component i the standard deviation of the quantitative error (W ^ i _{-W i)} which is a difference between the obtained accuracy S _a for the entire quantitative value of the component i by a set of standard sample used.

_{S A = (Σ (W ^} i -W i) 2 / (n-m)) 1/2 ... (3)
n: Number of standard samples used m: Number of calibration curve constants used

These evaluation standard values _{W i} for each standard sample, quantitative values W _{^ i} and quantification error _(W ^ i _-W i), also accuracy _{S A} for the entire quantitative value, a calibration curve formula created (1) Is displayed for. Thus, in the quantitative analysis by calibration curve method, the estimated reference value X _i in the graph of the calibration _curve, quantitative values W ^ _i and quantification error of each standard sample (W ^ i _{-W i),} exactly for the entire quantitative values degrees S _a is, all displayed in the same units as the standard value W _i, calibration equation (1) can be easily properly whether created evaluated.

On the other hand, in the quantitative analysis by the fundamental parameter method, for the analysis of an unknown sample, a set of standard samples whose component contents are known as standard values is used, and each fluorescent X-ray of the measurement element corresponding to the component is used. , Correlation between the theoretical strength and the measured strength calculated by the theoretical strength formula using the mass fraction (1/100 of the mass fraction) of the measurement element corresponding to the standard value and the mass fraction of the sample constituent element obtained from the standard value. The device sensitivity curve represented by the following equation (4) is obtained, and the device sensitivity constant is determined (see, for example, paragraph 0003 of Patent Document 1 and steps S1 to S3 in FIG. 4).

I _Ti = aI _Mi ² + bI _Mi + c ... (4)
I _Ti : The theoretical intensity of the fluorescent X-ray of _{the measurement element corresponding to the component i I Mi} : The measurement intensity of the fluorescent X-ray of the measurement element corresponding to the component i a, b, c: Device sensitivity constant

Then, for the evaluation of the created device sensitivity curve, the graph of the device sensitivity curve represented by the above equation (4) is displayed as shown in FIG. 3 for, for example, the component Cr. In FIG. 3, the horizontal axis coordinates of the black circle points indicate the theoretical intensity for the standard sample.

Further, for the standard sample, _{by substituting the measurement intensity I Mi} corresponding to the component i into the equation (4) of the device sensitivity curve, the conversion measurement intensity I ^ _Ti , which is the measurement intensity converted to the theoretical intensity scale, is obtained, and further. , Conversion of component i by a set of standard samples used as the standard deviation of the error (I ^ _Ti −I _{Ti ), which is the difference between the converted measured strength I ^ Ti} and the theoretical strength I _Ti , according to the following equation (5). determining the accuracy of S _B for the entire measured intensities.

_{S B = (Σ (I ^} Ti -I Ti) 2 / (n-m)) 1/2 ... (5)
n: Number of standard samples used m: Number of device sensitivity constants used

These theoretical intensity _{I Ti} per standard sample, in terms of the measured intensity _{I ^ Ti} and error _(I ^ _{Ti -I} Ti), also accuracy _{S B} for the entire conversion measured intensities for evaluation of the apparatus sensitivity curve prepared Is displayed in.

International Publication No. 2018/168939

In quantitative analysis by the fundamental parameter method, the theoretical intensity _{I Ti} in the graph of device sensitivity curve, the theoretical intensity of each standard sample _{I Ti,} in terms of the measured intensity _{I ^ Ti} and error _(I ^ _{Ti -I} Ti), for the entire conversion measured intensity accuracy S _B of all because the standard value W _i is expressed in units i.e. the theoretical intensity scale of different dimensions, device sensitivity curve is not easy to properly whether created evaluated.

The present invention has been made in view of the above-mentioned conventional problems, and an object of the present invention is to facilitate evaluation of whether or not an apparatus sensitivity curve is appropriately created in a fluorescence X-ray analyzer by a fundamental parameter method.

In order to achieve the above object, the present invention first irradiates a sample with primary X-rays, and based on the measured intensity of the generated fluorescent X-rays, contains components in the sample by a quantitative means using a fundamental parameter method. This is a fluorescent X-ray analyzer for obtaining a quantitative value of the rate. The quantification means is used for a set of standard samples in which the content of the component is known as a standard value, and the mass fraction and the standard value of the measurement element corresponding to the standard value for each fluorescent X-ray of the measurement element corresponding to the component. The device sensitivity constant is determined by obtaining the device sensitivity curve which is the correlation between the theoretical strength calculated by the theoretical strength formula and the measured strength using the mass fraction of the sample constituent elements obtained from.

Then, for each standard sample, the quantification means obtains the measurement intensity, the device sensitivity constant, and a proportional coefficient multiplied by the mass fraction of the measurement element in order to calculate the theoretical intensity in the theoretical intensity formula. It is used to calculate the quantitative value of the content of the component corresponding to the measurement element. Further, the quantification means is a graph showing the correlation between the standard value and the quantification value for each component, and / or the standard value, the quantification value, the quantification error for each standard sample and the entire quantification value by the set of standard samples. Outputs the accuracy of.

In the fluorescent X-ray analyzer of the present invention, a graph showing the correlation between the standard value and the quantitative value and / or the standard value, the quantitative value, and the quantitative error for each standard sample were used for each component by the quantitative means. The accuracy of the entire quantification value from a set of standard samples is output. Here, since the quantitative value in the graph, the quantitative value and the quantitative error for each standard sample, and the accuracy of the entire quantitative value are all displayed in the same unit as the standard value, the calibration curve formula is used in the quantitative analysis by the calibration curve method. Just as it is easy to evaluate whether or not the device sensitivity curve is properly created, it is easy to evaluate whether or not the device sensitivity curve is properly created.

It is the schematic which shows the fluorescence X-ray analyzer of one Embodiment of this invention. This is an example of a graph of a calibration curve output by a fluorescent X-ray analyzer using a conventional calibration curve method. This is an example of a graph of the device sensitivity curve output in the fluorescent X-ray analyzer by the conventional fundamental parameter method. It is a flowchart which shows the operation of the quantification means provided in the fluorescence X-ray analyzer of one Embodiment of this invention. This is an example of a graph showing the correlation between the standard value and the quantitative value output by the quantitative means.

Hereinafter, the fluorescent X-ray analyzer according to the embodiment of the present invention will be described. As shown in FIG. 1, this apparatus measures the intensity of secondary X-rays 5 generated by irradiating samples 1 and 14 (including both unknown sample 1 and standard sample 14) with primary X-rays 3. A scanning type fluorescent X-ray analyzer, the sample table 2 on which the samples 1 and 14 are placed, the X-ray source 4 such as an X-ray tube that irradiates the samples 1 and 14 with the primary X-ray 3, and the X-ray source 4. A spectroscopic element 6 that disperses secondary X-rays 5 such as fluorescent X-rays generated from samples 1 and 14, and a detector 8 that detects the intensity of the incident secondary X-rays 7 dissociated by the spectroscopic element 6. And have. The output of the detector 8 is input to a control means 11 such as a computer that controls the entire apparatus via an amplifier (not shown), a pulse height analyzer, a counting means, and the like.

This device is a wavelength dispersive and scanning type fluorescent X-ray analyzer, and interlocks the spectroscopic element 6 and the detector 8 so that the wavelength of the secondary X-ray 7 incident on the detector 8 changes. The means 10, that is, a so-called goniometer is provided. When the secondary X-ray 5 is incident on the spectroscopic element 6 at a certain incident angle θ, the extension line 9 of the secondary X-ray 5 and the secondary X-ray 7 spectroscopically (diffused) by the spectroscopic element 6 are 2 at the incident angle θ. Although the spectral angle 2θ is doubled, the interlocking means 10 changes the spectral angle 2θ to change the wavelength of the second-order X-ray 7 dispersed, and the dispersed second-order X-ray 7 is sent to the detector 8. The spectroscopic element 6 is rotated about an axis O perpendicular to the paper surface passing through the center of the surface so as to be incident, and the detector 8 is rotated along the circle 12 about the axis O by twice the rotation angle. And rotate. The value of the spectral angle 2θ (2θ angle) is input from the interlocking means 10 to the control means 11.

This device includes a quantification means 13 as a program mounted on the control means 11, and the quantification means 13 using the fundamental parameter method based on the measurement intensity of the fluorescent X-ray 5 is used to display the components in the samples 1 and 14. Obtain a quantitative value of the content rate. In the present invention, the fluorescent X-ray analyzer may be a wavelength dispersive and multi-element simultaneous analysis type fluorescent X-ray analyzer, or an energy dispersive fluorescent X-ray analyzer.

Next, the operation of the quantification means 13 will be described with reference to the flowchart of FIG. First, as in the quantitative analysis by the conventional fundamental parameter method described above, for analysis of an unknown sample 1, in step S1, components i, the set content of the j W _i, W _j are known as the standard value using a standard sample 14, for each X-ray fluorescence 5 measurement element corresponding to component i, _(1/100 of mass percentage) mass fraction w _i of the measuring element corresponding to the standard values W _i and standard values W _i, the theoretical intensity I _Ti calculated by theoretical strength equation using the mass fraction w _k of the sample constituent element k obtained from W _j, as a function of the measured intensity I _Mi, tables as equation (4) The device sensitivity curves to be obtained are obtained to determine the device sensitivity constants a, b, and c. The sample constituent element k is all the elements constituting the samples 1 and 14, and includes the measurement element corresponding to the component i. Further, when the components i and j are compounds, the sample constituent element k has a one-to-one correspondence with the components i and j, so different symbols k are used.

Here, as the theoretical intensity equation for calculating the theoretical intensity, the following equation (6) is used when it is assumed that only the primary excitation is generated by the excitation by the primary X-ray of a single wavelength.

_ITpi = K _i w _i / Σμ _k w _k … (6)
I _Tpi : Theoretical intensity of fluorescent X-rays of the measurement element corresponding to component i (primary excitation)
K _i : Constant w _i : Mass fraction of the measurement element corresponding to the component i μ _k : Total absorption count of the sample constituent element k with respect to the fluorescent X-ray corresponding to the component i w _k : Mass fraction of the sample constituent element k

As can be understood from formula _{(6), K i / Σμ} k w k = I Tpi ' , in theoretical strength formula (6), calculates the theoretical intensity _{I Tpi} of the fluorescent X-ray measurement element corresponding to component i to have a proportional coefficient to be multiplied to the mass fraction w _i of the measuring element corresponding to the component i. This proportional coefficient I _{Tpi'is a theoretical intensity formula (6) for calculating the theoretical intensity I Tpi} of the fluorescent X-ray of the measurement element corresponding to the component i, and the mass fraction _{wi of the measurement element corresponding to the component i.} in it can be said that the numerical value which is calculated by dividing the formula _{K i} / _{Σμ k} w _k. Theoretical strength I _Tsi corresponding to component i of the secondary excitation is similarly proportional to the mass fraction w _i of the measuring element corresponding to component i, when the proportionality coefficient I _{Tsi ',} the primary excitation and a secondary even theoretical strength I _Ti corresponding to component i of the combined following excitation is proportional to the mass fraction w _i of the measuring element corresponding to the component i, the proportionality factor, I _Ti by the following equation (7) ' It becomes.

_{I Ti = I Tpi + I Tsi} = (I Tpi '+ I Tsi') w i = I Ti 'w i ... (7)

When the samples 1 and 14 are single-layer thin film samples, a term depending on the thickness of the samples 1 and 14 is added to the theoretical strength equation, but it corresponds to the component i as in the equation (7). theoretical strength I _Ti is proportional to the mass fraction w _i of the measuring element corresponding to the component i. _{Further, even when the samples 1 and 14 are multi-layer thin film samples including a substrate, the theoretical strength I Ti} corresponding to the component i is the component i unless the same measurement element is contained in the plurality of layers including the substrate. It is proportional to the mass fraction w _i of the measurement element corresponding to.

In theoretical strength formula described above, the mass fraction of the measurement element is calculated from the standard value W _i of component i as a mass fraction w _i of the measuring element corresponding to the component i, the mass fraction of the sample constituent element k using mass fraction _{w k} of each sample constituent element k to the rate _{w k} is calculated from the standard values _W i, _{W j,} calculating the theoretical intensity _{I Ti.} Then, the device sensitivity curve represented by the above equation (4) as the correlation between the _{theoretical intensity I Ti and the measurement intensity I Mi} of the fluorescent X-ray 5 of the measurement element corresponding to the component i of the standard sample 14. To determine the device sensitivity constants a, b, and c.

When the component i is an element, the element itself is a measurement element corresponding to the component i. Therefore, the content rate Wi of the component _i (including the standard value and the quantitative value of the content rate, etc., is generally a mass fraction. conversion from represented) by (mass%), the mass fraction w _i of the measuring element corresponding to component i is merely multiplied by 1/100. If component i is a compound, the content of W _i of component i, is converted to the mass fraction w _i of the measuring element corresponding to component i, representing the measurement element (the compound of molecular weight and the corresponding compounds It is done by a well-known technique based on the atomic weight of the element). In addition, when a sample such as an oxide is prepared as a glass bead and the glass bead is used as a sample for fluorescent X-ray analysis, or when a powder sample is mixed with a binder and the mixture is used as a sample for fluorescent X-ray analysis. in the case of processing the conversion from content W _i of component i in the sample, the mass fraction w _i of the measuring element corresponding to the component in the sample i is dilution of the sample is vaporized by a glass bead prepared It is done by a well-known technique based on the content of the igros component and the like.

To give a specific example of step S1, samples 1 and 14 are stainless steel, and five standard samples 14 having standard values as shown in Table 1 are used, and fluorescent X-rays of the measurement element Cr corresponding to the component Cr are used. For Cr-Kα rays, as the correlation between the theoretical intensity and the measured intensity as shown in Table 2, the device sensitivity curve (device sensitivity constant a = 0) of the linear equation was obtained by the minimum square method, and the device in the previous equation (4). The sensitivity constant is determined to be b = 1.10238 and c = 6.91666.

Next, in step S2, the quantification means 13 calculates the measured intensity I _Mi , the device sensitivity constants a, b, and c, and the theoretical intensity I _{Ti in the theoretical intensity formula for each standard sample 14.} It said measuring element by using the mass fraction w _i to be multiplied proportionality factor I _{Ti 'of} the calculated as follows quantitative value W ^ _i of the content of components i corresponding to the measurement element.

First, Equation (4) and the formula derived from (7) (8) to calculate the quantitative value w ^ _i mass fraction w _i of the measuring element corresponding to the component i. Incidentally, the proportional factor _{I Ti} 'is the theoretical intensity formula used in step S1, for example, the formula Equation (6), divided by the mass fraction _{w i} of the measuring element corresponding to the component i, for example, _{K i} / _{Σμ k} w Calculate with _k.

w ^ _i = (aI _Mi ² + bI _Mi + c) / I _Ti '… (8)

Then, the inverse conversion to the conversion from the content W _i of component i mentioned in step S1 to the mass fraction w _i of the measuring element corresponding to the component i, the mass fraction w of the measuring element corresponding to component i by performing relative quantitative value w ^ _i of _i, to calculate the quantitative value W ^ _i of the content of components i corresponding to the measurement element.

In the specific example described in step S1, as shown in Table 3, for each standard sample 14, (aI _Mi ² + bI _Mi + c) (a = 0 in this specific example, expressed as converted measurement intensity in Table 3) is proportional. Quantitative value w ^ _{i (denoted as Cr mass fraction in Table 3) of the mass fraction w i} of the measurement element Cr corresponding to the coefficient I _Ti ', the component Cr, and the quantitative value of the content Cr of the component Cr corresponding to the measurement element Cr. The value W ^ _i (denoted as Cr quantitative value in Table 3) is calculated.

Next, in step S3, the quantification means 13 is a graph showing the correlation between the _{standard value Wi} and the quantification value W ^ _{i for each component i, and / or the standard value Wi} and the quantification value for each standard sample 14. W _{^ i,} and outputs the accuracy _{S C} for the entire quantitative values by quantitative error _(W ^ _i -W i) a set of standard sample 14 used. Specifically, the graph and / or each numerical value is displayed on a display or a printer (not shown). It is also possible to display the accuracy S _C superimposed on a graph showing the correlation between the standard value W _i and quantitative values W ^ _i. Here, accuracy _{S C} is the following equation (9), determined as the standard deviation of quantitative error _(W ^ _i -W i).

_{S C = (Σ (W ^} i -W i) 2 / (n-m)) 1/2 ... (9)
n: Number of standard samples used m: Number of device sensitivity constants used

In the specific example described in steps S1 and S2, the graph of FIG. 5 and / or Table 4 and the fact that the accuracy is 0.11 mass% are output and displayed on the display.

_{After that, the step of obtaining the quantitative value W ^ i} of the content of the component i in the unknown sample 1 using the prepared device sensitivity curve is performed in the same manner as the quantitative analysis by the conventional fundamental parameter method.

As described above, in the fluorescent X-ray analyzer of the present embodiment, the quantification means 13 is used _{to show the correlation between the standard value Wi} and the quantification value W ^ _i for each component i, and / or the standard sample. standard values _{W i} for each 14, quantitative values W _{^ i,} accuracy _{S C} for the entire quantitative values by quantitative error _(W ^ _i -W i) a set of standard sample 14 used is output. Here, the quantitative value W _{^ i} in the graph, quantitative values W _{^ i} and quantification error of each standard sample _{14 (W ^ i -W i)} , the accuracy of _{S C} for the entire quantitative value, all the standard values _{W i} Since it is displayed in the same unit, it is easy to evaluate whether or not the calibration curve formula is properly prepared in the quantitative analysis by the calibration curve method, and just as it is easy to evaluate whether or not the device sensitivity curve is properly prepared. Is easy.

Note that, in step S2 of the present embodiment, the theoretical intensity type using in step S1 by the formula divided by the mass fraction w _i of the measuring element corresponding to component i, and computes the proportionality factor I _{Ti '} .. In contrast, calculated in step S1, the theoretical intensity I _Ti of the fluorescent X-ray measurement element corresponding to component i, also calculated in step S1, the mass fraction w _i of the measuring element corresponding to component i It is also conceivable to simply divide by and obtain the proportionality coefficient I _Ti'. However, in this method, when the mass fraction wi of the measurement element corresponding to the component _i is 0, the divisor becomes 0 and it cannot be dealt with. Therefore, assuming as an improvement proposal, in step S1, the composition mass fraction w _i of the measuring element corresponding to component i is 0, the by replacing the mass fraction w _i in trace amounts, for example, 10 ^-8 Then, it is conceivable to calculate the theoretical intensity of each component. That way, the mass fraction w _i of the measuring element corresponding to component i is no longer zero. Thus, the theoretical intensity I _Ti of the fluorescent X-ray measurement element corresponding to components i, then simply divided by the mass fraction w _i of the measuring element corresponding to component i, can be determined proportionality factor I _{Ti '} ..

1 Unknown sample 3 Primary X-ray 5 Fluorescent X-ray 13 Quantitative means 14 Standard sample

Claims (1)

A fluorescent X-ray analyzer that irradiates a sample with primary X-rays and obtains a quantitative value of the content of the components in the sample by a quantitative means using a fundamental parameter method based on the measured intensity of the generated fluorescent X-rays. ,
The quantification means
For a set of standard samples whose component content is known as a standard value, the mass fraction of the measurement element corresponding to the standard value and the sample composition obtained from the standard value for each fluorescent X-ray of the measurement element corresponding to the component. The device sensitivity constant is determined by obtaining the device sensitivity curve, which is the correlation between the theoretical strength calculated by the theoretical strength formula using the mass fraction of the element and the measured strength.
For each standard sample, the measured element is supported by using the measured intensity, the device sensitivity constant, and a proportional coefficient multiplied by the mass fraction of the measured element in order to calculate the theoretical intensity in the theoretical strength formula. Calculate the quantitative value of the content of the component to be
For each component, a graph showing the correlation between the standard value and the quantitative value, and / or the standard value, the quantitative value, the quantitative error for each standard sample, and the accuracy of the entire quantitative value of the set of standard samples are output. , Fluorescent X-ray analyzer.

Images