JPH10267848A - Method for emission spectroscopic analysis - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve analysis accuracy by obtaining a relationship of a repeatability and a spectral intensity of a base element beforehand, setting as a threshold value the spectral intensity of the base element of a highest measurement accuracy, and obtaining a concentration when the spectral intensity is within a range of the threshold value. SOLUTION: In comparing changes in spectral intensity of Al/Fe, Al, O and Fe at points (A) and (B), while a spectral intensity ratio of Al/Fe is high, Al shows no intensity change, O shows high intensity and Fe has low intensity. When an Al oxide or the like is discharged, the intensity of Fe is considered to be abnormally low. As such, when a concentration of Al dissolved in a steel plate is to be measured, a relationship of the intensity ratio of Al/Fe and intensity of Fe is obtained, the intensity of Fe showing a large variance is detected, data are removed actually in the vicinity of the intensity, and accuracy is obtained. The intensity of Fe of a best accuracy is set as a threshold value. An optimum value is thus detected for every measurement.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an emission spectroscopy method for performing elemental analysis in a sample by performing a large number of spark discharges between a metal sample and a counter electrode in an inert gas atmosphere.

[0002]

2. Description of the Related Art In conventional emission spectroscopy, the intensity of matrix elements for each discharge is monitored. Sparks with reduced intensity are attributed to poor light emission. When the intensity is extremely high, abnormal emission is determined. Therefore, it was considered that the abnormalities occurred, and these abnormal values were removed to obtain an analysis value.
Excluded values (thresholds) were set to several percent to 20% at both low and high spark intensity locations. The measurement data is appropriately cut at such a threshold value to improve the accuracy. That is, the threshold value has been appropriately determined until now. If the threshold value is too large, there is a problem that the data amount decreases, the variation increases, and the accuracy becomes poor. If the threshold value is reduced, an abnormal value varies.

However, in such a conventional method, the analysis accuracy is not always improved, and it is unclear whether or not the above-mentioned appropriately determined threshold value is the best value.

[0004]

SUMMARY OF THE INVENTION In view of the above circumstances, the present invention has been completed as a result of a search for an appropriate method for determining the threshold value. The purpose is to:

[0005]

SUMMARY OF THE INVENTION The present invention is a method of actually calculating accuracy by using a large number of samples and determining the accuracy so that the accuracy is maximized. In an emission spectroscopy method in which a number of spark discharges are performed between a sample and a counter electrode to perform elemental analysis in the sample, the relationship between the measurement accuracy and the spectrum intensity of the base material element is determined in advance, and the measurement accuracy is the highest. The spectrum intensity of the base material element is determined, and the obtained value is used as a threshold value. The spectrum intensity of the base material element is measured for each discharge, and the concentration is determined from the data when the spectrum intensity is within the range of the threshold value. This is an emission spectroscopic analysis method characterized by being determined.

When the measured data is a spectrum intensity ratio between the measured element and the base material element, and a relationship between the spectrum intensity ratio and the spectrum intensity of the base material element is obtained, highly accurate measurement can be performed.

[0007]

BEST MODE FOR CARRYING OUT THE INVENTION In the Al concentration analysis of a silicon steel sheet having an Al concentration of 250 ppm, FIG. 2 was obtained by examining the spectral intensity ratio for each discharge obtained by the emission spectral analysis method. In FIG. 2, the horizontal axis represents the number of discharges, and the vertical axis represents the intensity ratio of the Al / Fe spectrum for each discharge. Al / F
The e-spectral intensity ratio is usually in the range of about 1 to 2, but may be abnormally high, and is called abnormal discharge, which causes a decrease in analysis accuracy. Therefore, the accuracy can be improved by rejecting the data having an abnormal discharge.

FIG. 3 is an enlarged view of the portion of the 1120 to 1200 discharges shown in FIG. FIG. 3 also shows the Al, O, and Fe spectral intensities along with the Al / Fe spectral intensity ratio. In FIG. 3, the Al / Fe spectral intensity ratio is high at points A and B. The fluctuations in the spectrum intensities of Al, O, and Fe at points A and B are as follows: (a) Al intensity is normal; (b) O intensity is high, which is considered to be due to discharge to the oxide ( C) Fe strength is low. It has become. It is conceivable that the discharge to the Al oxide or the like causes an abnormal decrease in the Fe intensity. If data in the discharge in which the Fe intensity is abnormally reduced is removed, the accuracy is improved. That is, if data with low Fe intensity is rejected, A
Points where the 1 intensity is normal but the spectral intensity ratio is abnormal are excluded. Next, at the point C, (a) the Al intensity is high, (b) the Fe intensity is high, and (c) the Al / Fe intensity ratio is normal. From the above, the relationship between the Al / Fe intensity ratio and the Fe intensity was extracted and examined. The result is shown in FIG. As shown in FIG. 4, when the Fe intensity decreases, the variation in the Al / Fe intensity increases. In particular, it has been found that when the strength of Fe is 1000 or less, it greatly varies.

Therefore, in order to remove defective data, Fe
When the repetition accuracy was obtained by changing the strength, the result of FIG. 1 was obtained. That is, it has been clarified that there is an optimum value at which the repetition accuracy is lowest. The analysis accuracy of the embodiment shown in FIG. 1 is such that the repetition accuracy (σ) for all data is 4.7.
ppm, but the repetition accuracy (σ) when removing data having an Fe intensity of less than 1100 was 2.6 ppm. In the examples according to the present invention, the analysis accuracy was significantly improved.

In the examples, the threshold value was determined from the relationship between the spectral intensity ratio of the measured element and the base material element and the spectral intensity of the base material element, using data obtained by measuring the calibration curve sample a plurality of times. At this time, (1) an optimum value is determined in advance by a number of measurement results,
Data processing is performed using this value. (2) There are two methods of determining an optimum value using actual data measured several times and performing data processing using this value. Either method may be used.

[0011]

Conventionally, in order to improve the measurement accuracy in the emission spectroscopy, the threshold value is specified by an appropriate percentage to exclude defective data. When measuring the concentration, use Al / F
Find the relationship between the e-intensity ratio and the Fe intensity, find the Fe intensity at which the variation increases, actually remove the data near this intensity, determine the accuracy, measure the Fe intensity at which the accuracy is the best as the threshold, and measure each time Since the optimum value is found every time, data with high analysis accuracy can be obtained quickly.

[Brief description of the drawings]

FIG. 1 is a graph showing a change in repetition accuracy when data is removed by changing an Fe spectrum intensity serving as a threshold.

FIG. 2 is a graph showing an Al / Fe spectrum intensity ratio for each discharge.

FIG. 3 is a graph showing Al, O, Fe, and Al / Fe spectral intensity ratios for each discharge.

FIG. 4 is a graph showing the relationship between the Fe spectrum intensity and the Al / Fe spectrum intensity.

Claims (2)

[Claims] 1. An emission spectroscopic analysis method for performing elemental analysis in a sample by performing a number of spark discharges between a metal sample and a counter electrode in an inert gas atmosphere. Determine the relationship with the intensity, determine the spectral intensity of the base metal element is the highest measurement accuracy,
An emission spectroscopy method comprising: measuring a spectrum intensity of a base material element for each discharge with the threshold value as a threshold; and obtaining a concentration from data when the spectrum intensity is within the range of the threshold value. . 2. The emission spectral analysis method according to claim 1, wherein said data is a spectral intensity ratio of a measured element and a base material element.