JP2655562B2 - Qualitative analysis method by X-ray fluorescence analysis - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for automatically performing optimal spectrum identification by a computer in qualitative analysis of a sample by a fluorescent X-ray analyzer.

[0002]

2. Description of the Related Art A conventional automatic spectrum identification analysis procedure is as follows.
First, a peak of an intrinsic X-ray of a target substance of an X-ray tube is searched for and removed from the obtained measured spectrum data (& 1). The strongest peak is searched from the measured spectrum data from which the influence of the X-ray tube has been removed, and the angle value of the strongest peak is read. By converting the wavelength of the search peak from the angle value, comparing the converted wavelength with the spectrum table data that lists the name and wavelength of the specific X-ray of each element, and searching for an element that matches the above search peak Identify peak elements (&
2). In the measured spectrum data, all the peaks that match the intrinsic X-ray of the identified peak element and the associated X-ray peak (eg, second-order diffracted X-ray, etc.) are marked (& 3). Return to operation (& 2) for unmarked peaks, and for all peaks,
Operate until mark is added, that is, until identification is completed (& 2)
Repeat ~ (& 4). The spectrum table used in the operation (& 2) is a table in which the spectral wavelengths of all the elements from ₅ B to ₉₂ U that can be analyzed by X-ray fluorescence are used as data.

However, in the above-described analysis method, since a search spectrum table is prepared in the order of atomic numbers and for all elements, the search table until the wavelength of the detected peak matches the wavelength in the spectrum table is obtained. Since the time becomes longer in proportion to the number of detected peaks and the number of spectrum table data, there is a problem that it takes much time. In addition, the correct answer rate of the identified spectrum is also worse because the information amount of the spectrum table data is larger, since the number of adjacent peaks is increased.
There was a problem that the accuracy rate was low. In order to solve these problems, in the conventional equipment, before analysis, the analyst specifies the elements that are considered not to be present in the sample, so that the spectral data of the elements not necessary for analysis can be obtained. Can be removed from the search spectrum table, and at the time of analysis, the measured spectrum data is identified using the search spectrum table from which the spectrum data of elements not required for analysis has been removed. However, even in this state, since the search is not performed in the order of the main components of the analysis sample, there is a problem that the search time and the accuracy rate are not sufficiently improved.

[0004]

SUMMARY OF THE INVENTION An object of the present invention is to improve search time and correct answer rate in automatic identification work in X-ray fluorescence analysis.

[0005]

In the qualitative analysis of X-ray fluorescence analysis, an element to be searched is specified from an element spectrum list, and a search order is set for each specified element to create a search order table. Means for storing, and means for selecting detected peaks in the order of signal intensity from a detected peak table obtained by X-ray fluorescence analysis, and identifying the strongest detected peak selected by the above means using the search order table; In addition to deleting the selected element peak data from the search rank table, all peak data resulting from the identified element is deleted from the detected peak table, and identification is performed using the search rank table in which the detected peaks sequentially selected are deleted. did.

[0006]

According to the present invention, the search time is long and the accuracy rate is low because peak data of many elements are randomly arranged in a search spectrum table. In the search spectrum table for each element,
In addition to specifying unnecessary, an element to be written in the search spectrum table is assigned a search order from the element that is predicted to be a strong peak, and a search order table is created in order from the peak data of the element to be searched preferentially, Create a detection peak table in which the detected peaks are arranged in order of strongest from the measured spectrum data, and identify the detected peaks in order from the upper detection peak in the detection peak table, the element data on the priority order of the ranking table created above Since the identification is performed by comparing in order from, the strongest detected peak is the detected peak of the main element, so that the search time is greatly reduced. In addition, the incidental peak derived from the main element is often stronger than the main detection peak of a small amount of the element, and the erroneous determination can be reduced by deleting the incidental detection peak derived from the main element. In the present invention, the higher the content of the element, the more certain the element name is, and the less erroneous judgment is made. Since the detection peak is stronger as the content of the element is higher, it is necessary to identify the detection peak in order from the strongest.
This reduces erroneous determinations. In addition, when the detected peak is identified, the associated peak derived from the identified element is deleted from the detected peak table. Therefore, all the associated detected peaks stronger than the main detected peak to be identified are deleted when the main detected peak is identified. Thus, erroneous determinations are reduced, and the accuracy of identification can be increased. Further, by sequentially deleting the peak data of the identified elements from the ranking table, it becomes possible to identify the detected peak at a higher accuracy rate in a shorter time.

[0007]

FIG. 1 shows a flow chart of an embodiment of the present invention. First, before analysis, analysis conditions are set (# 1).
The measurement conditions are set (# 2). A search element ranking table is created by deleting search unnecessary elements and sequentially assigning search priority to elements expected to have a high content (# 3). The analysis is started, the spectrum is measured (# 4), and the measured spectrum data is stored (# 5). Next, data processing is started. First, a peak is searched from the stored measured spectrum data, and a detected peak table is created (# 6). Remove the peak due to the tube target from the detection peak table (#
7). The strongest detected peak is searched from the detected peak table, and the wavelength of the strongest detected peak is set to λ ₀ (# 8). N is set to 1 (# 9). The N-th element wavelength in the ranking table is set to λ _N (# 10). It is determined whether λ ₀ = λ _N (# 1
1). If not λ ₀ = λ _N , add 1 to N and N ←
N + 1 is set (# 12), and the process returns to step # 10. λ ₀ =
In the case of λ _N , the element of the detected peak is determined to be the N-th element (# 13), the data of the same element is deleted from the ranking table (# 14), and the peak of the incidental spectrum of the same element (specific X-ray and The peak due to diffraction X-rays) is deleted from the detected peak table (# 15), and it is determined whether there is a residual peak in the deleted detected peak table (# 16). Then, the head of the remaining elements in the ranking table is set to N = 1, the process returns to step # 8, and the identification analysis of the residual peak is continued. If there is no residual peak, the analysis result is output (# 17), and the analysis ends.

[0008]

According to the present invention, prior to analysis, a ranking table is created by rearranging a spectrum table for determination in an order in which strong peaks are expected for elements contained in an analysis sample, and the ranking table is prepared. By identifying the detected peaks by using, the peaks are identified in order from the strongest peak, and other peaks derived from the element that generates the strong peak will also be deleted from the ranking table. At the time of identification, only the peaks after all the accompanying peaks derived from the main elements have been deleted, so
The number of errors in the determination is reduced, and the analysis speed and the accuracy rate are greatly improved. In addition, since analysis priorities can be set for sample elements for each sample form, automatic data processing by a computer can be made closer to the same data processing method as the data processing intended by the analyst. It became so.

[Brief description of the drawings]

FIG. 1 is a flowchart of an embodiment of the present invention.

 ──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-60-135549 (JP, A) Nikkan Kogyo Shimbun "Radiation Measurement Handbook", pages 609-610, published in 1982

Claims (1)

(57) [Claims] 1. A means for designating an element to be searched from an element spectrum list, setting a search order for each element, creating and storing a search order table, and a means for detecting a search peak table obtained by X-ray fluorescence analysis. Means for selecting detection peaks in order of signal intensity, identifying the detection peaks selected by the means in accordance with the search order table in order of intensity, deleting the identified element peak data from the search order table, A qualitative analysis method based on X-ray fluorescence analysis, wherein all peak data generated from an identified element are deleted from the above, and the above-described identification operation is performed on a detected peak of the next intensity.