JPH0245744A - X-ray spectroscopic analysis apparatus - Google Patents

Abstract

PURPOSE:To shorten the time required for element identification by providing means for rearranging the wavelength data of the respective elements stored in a wavelength chart memory to the appearance frequencies added by an appearance frequency adding means. CONSTITUTION:The identification of the sample component elements is executed by contrasting the respective peak wavelength and intensity ratios of the main peaks and sub-peaks of the characteristic X-rays of the respective elements stored in the wavelength chart memory 6 after the measurement data of the X-ray spectra with the sample from an X-ray detector is stored into a measurement data memory 4 via a control part body 2. One appearance frequency is added to the wavelength data of the identification element of the memory 6 in the appearance frequency adding means 8 at every identification when all the component elements of the sample are identified. The wavelength data of the respective wavelengths of the memory 6 are then rearranged in order of the larger appearance frequencies in a rearranging means 10. The wavelength data are, therefore, contrasted in order of the elements having the higher probability of the appearance frequency in case of making qualitative analysis of the same kind of the sample in the next time. The time required for the identification is thus shortened.

Description

DETAILED DESCRIPTION OF THE INVENTION (A) Field of Industrial Application The present invention relates to an X-ray spectrometer, and particularly relates to an improvement in the data processing section of the same device for element identification.

(b) Conventional technology In qualitative analysis to identify sample component elements, the sample is excited with an electron beam, the characteristic X-rays emitted from the sample are detected, the X-ray spectrum is measured, and the peaks of the X-ray spectrum are measured. The degree of agreement between the two is checked by comparing the intensity, peak wavelength, etc. with a database that collects data on spectral peaks of all elements.

If there is a spectral peak that best matches the measured X-ray spectrum, it is assumed that the sample contains an element corresponding to this spectral peak, and the element is identified.

(c) Problems to be Solved by the Invention The characteristic X-ray spectral peaks of each element contain a large amount of information from the main peak to the peaks of higher-order lines. Searching for each peak one by one takes a very long time.

Therefore, rapid qualitative analysis cannot be performed.

In order to speed up qualitative analysis, for example, you can set in advance only one peak wavelength of the most characteristic characteristic X-ray for each element, and use the presence or absence of this single peak wavelength to determine the By determining the corresponding elements, the peak search time can be shortened.

However, in general, the number of peaks in the measured X-ray spectrum is large, and the peaks of coexisting elements often appear close to each other. Therefore, if the corresponding element is determined based on only a single peak wavelength, there is a high possibility of incorrect identification, leading to a lack of reliability in the automatic analysis results.

The present invention has been made in view of these circumstances, and it is an object of the present invention to make it possible to shorten the time required for element identification as much as possible without impairing the reliability of analysis results.

(d) Means for Solving the Problems When performing qualitative analysis using an X-ray spectrometer, samples containing the same type of component elements are often analyzed. For this reason, there is a bias in the appearance frequency of elements contained in the sample. For example, in fields related to steel, there is a high probability that Fe, Ni, Cr, etc. will be identified. In addition, in the field of semiconductors, S
There is a high probability that i, Ga, As, Ge, etc. will be identified. Therefore, if data is searched preferentially starting from data with a high probability of being identified, the search time will be reduced accordingly.

The present invention has been made with attention to this point, and in order to achieve the above object, it adopts the following configuration.

That is, the X-ray spectrometer of the present invention has a wavelength table memory and a wavelength table memory in which the wavelengths of at least the strongest peak and the second strongest peak of characteristic X-rays of each element are classified by element and stored as a database. , appearance frequency adding means for adding one appearance frequency of an identified element each time an element contained in the sample is identified to the wavelength data for each element stored in the wavelength table memory; and rearranging means for rearranging the wavelength data of each element stored in the wavelength table memory in the order of appearance frequencies added by the appearance frequency adding means.

(E) Effect In the above configuration, identification of sample component elements is performed by checking at least both the wavelength of the first strongest peak and the wavelength of the second strongest peak of the characteristic X-rays of each element stored in the wavelength table memory. Therefore, the accuracy of element identification is increased.

When the component elements of the sample are identified in this way, the appearance frequency addition means adds the appearance frequency of the identified element by 1 to the wavelength data for each element stored in the wavelength table memory. Add one. Subsequently, the sorting means sorts the wavelength data of each element in the wavelength table memory in order of appearance frequency.

Therefore, the next time a sample containing similar components is qualitatively analyzed, the peak wavelengths of the elements are compared in descending order of probability of occurrence, reducing the time required for identification each time. become. In other words, a data search learning function has been added.

(F) Embodiment FIG. 1 is a block diagram showing the main part configuration of an X-ray spectrometer. In the same figure, the symbol 1 indicates the entire X-ray spectrometer (EPMA in this example), and 2 is the main body of the control section in the device 1. Measurement data of an X-ray spectrum obtained by detection with a detector is input. Reference numeral 4 denotes a measurement data memory 1- for storing measurement data of the X-ray spectrum of the sample. That is, the measurement data memory 4 stores the peak wavelength and peak intensity of all peaks included in the X-ray spectrum obtained from the sample in association with each other. 6 is a wavelength table memory, and this wavelength table memory 6 includes a second
As shown in Figure (a), the wavelength λ of the strongest peak (hereinafter referred to as main peak) of the characteristic X-rays of each element. and intensity ratio I. /■, the wavelength λ of the second strongest peak (hereinafter referred to as sub-peak), and the intensity ratio 11/I are stored separately for each element. Reference numeral 8 denotes an appearance frequency addition means, IO, which adds one appearance frequency of an identified element to the wavelength data for each element stored in the wavelength table memory 6 each time an element contained in a sample is identified. is a rearranging means for rearranging the wavelength data of each element stored in the wavelength table memory 6 in the order of appearance frequencies added by the appearance frequency adding means 8.

Next, the operation of the above configuration will be explained.

Initially, the appearance frequencies added to the wavelength data in the wavelength table memory 6 are all set to "0" for each element.

Identification of sample component elements is performed after the measurement data of the X-ray spectrum of the sample is once stored in the measurement data memory 4.
This is done by comparing the peak wavelengths and intensity ratios of the main peaks and sub-peaks of the characteristic X-rays of each element stored in the wavelength table memory 6. That is, the control unit main body 2
First, each wavelength and intensity data indicating the largest peak intensity is read out from the measurement data memory 4, and it is checked whether the wavelength of the data exists in the main peak column of the wavelength table memory 6. In this case, the search is performed in order from the smallest address in the wavelength table memory 6. That is, in this example, data is searched from top to bottom in FIG. 2(a). If an element with the corresponding peak wavelength exists, next, data on the peak wavelength and intensity ratio of the sub-peak of the same element is read out from the wavelength table memory 6, and a peak corresponding to this sub-peak exists in the measurement data memory 4. Find out whether or not. If the element exists, the element corresponding to the searched main peak and subpeak is determined as the sample component element. In this way, elements are sequentially identified based on the peak wavelength and intensity ratio of two characteristic X-rays for each element. Therefore, the accuracy of element identification becomes high.

When all the component elements of the sample are identified, the appearance frequency adding means 8 adds one appearance frequency to the wavelength data of the identified element in the wavelength table memory 6 each time the element is identified. for example,
When Fe, Cr, and Ni are identified in one sample, +1 is added for each element.

Subsequently, the rearranging means 10 rearranges the wavelength data of each element in the wavelength table memory 6 in descending order of appearance frequency.

This sorting is performed by the following steps. for example,
As shown in FIG. 3, if wavelength data of 100 elements is stored in the wavelength table memory 6, the addresses 1 to 100 are assigned in order from the top, so Compare the frequency of appearance above and below. In other words, first compare the appearance frequencies of the wavelength data at the 99th and 100th addresses, and if the frequency of appearance of the wavelength data at the 100th at1nos increases, it will be compared to the wavelength data at the 99th address. Replace. After the replacement, the frequencies of appearance of the wavelength data at the 98th and 99th addresses are compared, and if the frequency of appearance of the wavelength data at the 99th address is large, the wavelength data at the 98th address is replaced. From then on, perform the same procedure. For example, if we focus on the wavelength data at the i-th and i+1-th addresses, we can replace the data by copying the wavelength data at the i+-th address to another memory area, and then
Copy the wavelength data at the i-th address to the memory area at the i+1-th address. Next, the i+1th wavelength data that has already been copied is copied to the i-th address. When the data is sequentially replaced in this way, the wavelength data with the highest frequency of appearance will finally be located at the first address.

Therefore, next, the data is replaced in the same manner as above, targeting the wavelength data at the 2nd to 100th addresses. Then, the wavelength data having the second highest frequency of appearance is located at the second address. Furthermore, next, the data is replaced in the same manner as above for the wavelength data located at the 3rd to 100th addresses. By repeating this procedure, the wavelength table memory 6 has the following information as shown in FIG. 2(b):
The wavelength data is arranged in an order in which the frequency of appearance gradually decreases from the 1st address to the 100th address.

Therefore, when qualitatively analyzing the same type of sample next time, the wavelength data is collated in descending order of the probability of occurrence of the element, so the time required for identification is shortened each time.

(G) Effects of the Invention According to the present invention, since element identification is performed based on the peak wavelengths of two characteristic X-rays for one element, the reliability of the analysis results is not impaired. Moreover, each time a component element of a sample is identified, the wavelength data stored in the wavelength table memory is sorted in order of appearance frequency, so in subsequent qualitative analysis of the sample, the peak wavelength of the element with the highest probability of appearance is sorted. Verification with measurement data is performed. As a result, when analyzing samples of the same type, the time required for element identification is shortened.

[Brief explanation of the drawing]

The drawings show an embodiment of the present invention. Fig. 1 is a block diagram showing the main part configuration of an X-ray spectrometer, Fig. 2 is a memory map of the wavelength table memory, and Fig. 3 shows the elements in order of appearance frequency. FIG. 3 is an explanatory diagram of a procedure for rearranging wavelength table data. 1... X-ray spectrometer, 4... Measurement data memory, 6... Wavelength table memory, 8... Appearance frequency addition means,
10... Sorting means.

Claims (1)

[Claims] (1) A wavelength table memory in which the wavelengths of at least the first and second strongest peaks of the characteristic X-rays of each element are classified by element and stored as a database; appearance frequency adding means for adding one appearance frequency of the identified element each time an element contained in the sample is identified to the wavelength data for each element contained in the wavelength table memory; An X-ray spectroscopic analysis apparatus comprising: a sorting means for sorting wavelength data of elements in the order of appearance frequencies added by the appearance frequency adding means.