JP3972307B2 - Ni emission spectroscopic analysis method and steel making method using this method - Google Patents

Description

  The present invention relates to an emission spectroscopic analysis method for Ni (nickel), and more particularly to a Ni emission spectroscopic analysis method suitable for analyzing Ni content in molten steel with high accuracy during melting.

  Conventionally, in an electric furnace facility, the molten steel is sampled in the middle of melting, and the components are subjected to emission spectroscopic analysis to adjust the additional input amount of each component. Among such component adjustments, especially Ni has a high unit price. Therefore, if the excess input can be reduced by increasing the analysis accuracy, there is an advantage that the raw material purchase cost can be greatly suppressed. Improvement is desired.

  Emission spectroscopic analysis is to measure and analyze the emission intensity of the emission line light specific to the element to be analyzed by repeating a spark discharge several thousand times between the discharge electrode and the sample and performing spectral analysis of the discharge light. . Here, even if the content of the analysis target contained in the sample is the same, the emission intensity of the discharge light generally varies, and therefore the analysis accuracy is reduced due to the variation. Various attempts have been made to avoid this problem.

  For example, as a general measure for improving the accuracy of analysis, an element contained in a sample at a high concentration (for example, Fe in molten steel) is used as a monitor of the discharge state, the emission intensity of the emission line light of the element to be analyzed, By evaluating the ratio of the emission light of the monitor element to the emission intensity of the monitor element, that is, by normalizing the emission intensity of the emission light of the element being analyzed, the emission intensity of the discharge light What reduces the influence of variation is known.

  Also, analysis is performed by designating multiple elements in the sample as monitor elements and integrating the emission line light intensity of the element to be measured in each discharge where the emission line intensity of the multiple elements per discharge was within a predetermined range. A method for improving the accuracy has been proposed. In other words, a method has been proposed in which variation in emission intensity of the discharge light is suppressed by deleting data on the emission line intensity outside the predetermined range as abnormal data, thereby improving analysis accuracy (for example, Patent Document 1).

  In the above-mentioned Patent Document 1, there is no clear guideline regarding the method for determining the predetermined range (effective range), and as a specific example, the range of the average value of the emission line light intensity ± 2σ is shown. This is only to disclose a general statistical method for noise removal on the assumption that the distribution of the emission line light intensity is a normal distribution.

Here, the inventors of the present invention emitted the emission intensity of Fe emission light and the emission intensity of the emission line light of Ni / the emission intensity of the emission line light of Fe using Fe in the molten steel as a monitor element. An example actually measured by spectroscopic analysis is shown in FIG. The horizontal axis of FIG. 7 plots the emission intensity of Ni emission line light / the emission intensity of Fe emission line light, and the vertical axis shows the emission intensity (dimensionless unit) of Fe emission line light. As shown in FIG. 7, the emission intensity of the Fe emission line light tends to be dipolarized into a distribution A and a distribution B, but as in Patent Document 1, the average value of the emission line light intensity of Fe ± 2σ. If the range is an effective range, the effective range is 1997 to 9577 (dimensionless unit), and almost all data is included in the effective range. Therefore, even if the emission intensity of the Ni emission line light is normalized by the emission intensity of the Fe emission line light, as shown in FIG. 7, there is a large variation in emission intensity after normalization, and the analysis accuracy is not improved. is there. The same applies to the case where a plurality of monitor elements are used as in Patent Document 1 described above.
Japanese Patent No. 252216

  The present invention has been made to solve such problems of the prior art, and is capable of reducing the influence of the variation in the emission intensity of the discharge light and analyzing the Ni content in the sample with high accuracy. It is an object to provide an emission spectroscopic analysis method.

  In order to solve the above-mentioned problems, the present invention, as described in claim 1, performs a plurality of spark discharges between a discharge electrode and a sample containing Ni that is disposed opposite to the discharge electrode, and generates discharge light. An emission spectroscopic analysis method for calculating the Ni content in a sample by performing spectral analysis, and collecting emission intensity data of emission line light for each spark discharge for Ni and other elements contained in the sample. Calculating the emission intensity of the bright line light of Ni / the emission intensity of the bright line light of other elements for the plurality of spark discharges; and the emission intensity of the bright line light of the other elements for the multiple spark discharges. A step of creating a frequency distribution diagram of the data, and only for the spark discharge from which emission intensity data having a numerical value larger than the valley portion of the frequency distribution diagram was obtained, Extracting the emission intensity of the line light / the emission intensity of the bright line light of the other element, calculating the average value of the emission intensity of the extracted Ni emission line light / the emission intensity of the bright line light of the other element, And a step of calculating the Ni content based on the calculated average value.

  According to the first aspect of the present invention, the emission intensity of the emission line light of Ni / the emission intensity of the emission line light of another element is calculated for a plurality of spark discharges using other elements other than Ni contained in the sample as monitor elements. After that, a frequency distribution diagram of the emission intensity data of the emission light of the other element is created, and only for the spark discharge in which the emission intensity data having a numerical value larger than the valley portion of the frequency distribution diagram is obtained, the calculated Ni The emission intensity of the emission line light / the emission intensity of the emission line light of another element is extracted. In other words, when the inventors of the present invention create a frequency distribution diagram of emission intensity data of emission line light of other elements for a plurality of spark discharges, the frequency distribution diagram has a bimodal shape having two peaks. It is found that it becomes a frequency distribution of, and only the spark discharge from which the emission intensity data having a numerical value larger than the valley located between the peaks is obtained is effective, and the emission intensity of the Ni emission line light for the effective spark discharge / It has been found that the analysis accuracy can be improved by extracting the emission intensity of emission light from other elements. The reason why the frequency distribution diagram has a bimodal frequency distribution is that the discharge path from the discharge electrode to the sample follows the shortest path substantially perpendicular to the sample surface and the side opposite to the shortest path. This is considered to be caused by the fact that the discharge path is different for each spark discharge and the both are mixed. Here, since the light emission intensity of the discharge light is higher when the discharge path is the shortest path, only the spark discharge in which light emission intensity data having a numerical value larger than the valley part of the frequency distribution diagram is obtained is valid. According to the invention according to item 1, the spark discharge that has followed the shortest path that can obtain a relatively stable light emission intensity among the two kinds of discharge paths is made effective, thereby improving the analysis accuracy. Is possible.

  Furthermore, according to the first aspect of the present invention, the Ni content is calculated based on the average value of the emission intensity of the extracted Ni emission line light / the emission intensity of the emission light of other elements. Here, in order to calculate the Ni content based on the average value of the emission intensity of the bright line light of Ni / the emission intensity of the bright line light of other elements, the average value is calculated in advance for the known samples having different Ni contents, A relationship with the Ni content calculated by chemical analysis may be acquired as a calibration curve, and the calibration curve may be used.

  In the invention according to claim 1, for all spark discharges, after calculating the emission intensity of Ni emission line light / emission intensity of emission light of other elements, the emission intensity having a value larger than the valley of the frequency distribution diagram Only for the spark discharge for which data was obtained, the emission intensity of Ni emission line light / the emission intensity of emission light of other elements is extracted. In other words, for all spark discharges, unnecessary normalized emission intensity data is deleted after normalizing the emission intensity of the Ni emission line light with the emission intensity of the emission line light of the other element as the monitor element. However, it is also possible to change the order of these and delete unnecessary emission intensity data (emission intensity data of emission light of Ni and other elements) first, and then normalize the remaining emission intensity data. is there.

  That is, in order to solve the above-mentioned problems, the present invention performs a plurality of spark discharges between a discharge electrode and a sample containing Ni disposed opposite to the discharge electrode, as described in claim 2, An emission spectroscopic analysis method for calculating the Ni content in a sample by spectrally analyzing light, wherein emission intensity data of emission line light is obtained for each spark discharge for each of Ni and other elements contained in the sample. A step of collecting, a step of creating a frequency distribution diagram of emission intensity data of emission line light of the other elements for the plurality of spark discharges, and emission intensity data having a numerical value larger than a valley portion of the frequency distribution diagram is obtained. Extracting emission intensity data of emission line light of Ni and other elements only for the spark discharge thus obtained, and emission intensity of emission line light of the extracted Ni and other elements, respectively For each spark discharge, the step of calculating the emission intensity of the Ni emission line light / the emission intensity of the emission light of the other element, and the calculated emission intensity of the Ni emission line light / the emission line light of the other element. A Ni emission spectroscopic analysis method characterized by including a step of calculating an average value of emission intensity and a step of calculating a Ni content based on the calculated average value is also possible.

  When the sample is an analytical sample sampled from molten steel, as described in claim 3, the other element is preferably Fe having a high content.

  In the present invention, as described in claim 4, the Ni content in the sample is calculated using the method according to any of claims 1 to 3, and the amount of Ni input is calculated based on the calculated Ni content. It is also provided as a steelmaking method characterized by adjusting the amount of steel, and this provides the advantage that the excessive amount of Ni input during steelmaking can be reduced and the raw material purchase cost can be greatly suppressed.

  As described above, according to the Ni emission spectroscopic analysis method according to the present invention, emission of bright line light of the other elements is performed for a plurality of spark discharges using an element other than Ni contained in the sample as a monitor element. Creates a frequency distribution map of intensity data, and enables only spark discharges with emission intensity data that is larger than the troughs of the frequency distribution chart, enabling data variation to be reduced. As a result, it is possible to improve the measurement accuracy of the Ni content.

  Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings.

  FIG. 1 is a diagram showing a schematic configuration of an emission spectroscopic analysis apparatus for carrying out a Ni emission spectroscopic analysis method according to an embodiment of the present invention. As shown in FIG. 1, the emission spectroscopic analysis apparatus according to this embodiment includes a discharge circuit 1, a discharge chamber 2, a spectrometer 3, an integrator 4, an A / D converter 5, and an interface (I / I). F) A substrate 6 and a microcomputer 7 are provided. More specifically, the discharge chamber 2 is provided with a discharge electrode 21 connected to the discharge circuit 1, and a sample containing Ni so as to face the discharge electrode 21 (sampled from molten steel in this embodiment). Analysis sample) S can be arranged. The spectrometer 3 is in a vacuum state and includes an entrance slit 31, a diffraction grating 32, an exit slit 33, and a plurality of photomultiplier tubes.

  With the emission spectroscopic analysis apparatus having the above configuration, spark discharge is performed a plurality of times (1200 times in the present embodiment) between the discharge electrode 21 and the sample S disposed opposite to the discharge electrode 21, and the spectrometer 3 By analyzing the discharge light spectrum, the Ni content in the sample S is calculated. More specifically, first, a high voltage pulse is repeatedly applied from the discharge circuit 1 to the discharge electrode 21 to generate a plurality of spark discharges between the discharge electrode 21 and the sample S. Discharge light generated by the spark discharge in the discharge chamber 2 enters the spectroscope 3 and is converted into a parallel light flux through the entrance slit 31. Next, the discharge light converted into a parallel light beam is split by the diffraction grating 32 and is incident on each photomultiplier tube 34 for each emission line light of each element contained in the sample S via the exit slit 33. Become. The emission intensity signal of the bright line light detected by each high electron multiplier 34 is integrated by the integrator 4 for each spark discharge and then converted to digital data by the A / D converter 5. The converted emission intensity data of the emission line light for each spark discharge is transmitted to the microcomputer 7 via the I / F substrate 6 and stored therein, and then a predetermined calculation program installed in the microcomputer 7 Thus, a calculation as described later is performed, and the Ni content in the sample S is calculated. The microcomputer 7 is also connected to the discharge circuit 1 via the I / F board 6 and operates the discharge circuit 1 at a predetermined timing by a predetermined control program installed in the microcomputer 7. In synchronism with this, the emission intensity data of the bright line light is captured.

Hereinafter, the details of the calculation performed by the calculation program will be described.
FIG. 2 is a flowchart schematically showing calculation contents in the emission spectroscopic device according to the present embodiment. As shown in FIG. 2, first, based on the emission intensity data of the emission line light for each spark discharge collected as described above, the emission intensity of Ni emission line light / the emission line light of Fe for a plurality of spark discharges. The light emission intensity (hereinafter referred to as Ni / Fe as appropriate) is calculated (S1). That is, the emission intensity of the Ni emission line light is normalized by the emission intensity of the Fe emission line light.

  Next, for the plurality of spark discharges, the data (Fe emission line light emission intensity data and Ni / Fe data) are rearranged in the order of the emission intensity of the Fe emission line light and the frequency of the emission intensity data of the Fe emission line light. A distribution map is created (S2).

  FIG. 3 shows an example of a frequency distribution diagram of the emission intensity data of the created Fe emission line light. In FIG. 3, the emission intensity (dimensionless unit) of the Fe bright line light is plotted for each of about 240 section units on the horizontal axis, and the frequency belonging to each section is plotted on the vertical axis. In addition, what is necessary is just to determine so that it may divide | segment between the maximum value and minimum value of the emission intensity data of bright line light into about 35-40 as a section unit. As shown in FIG. 3, the frequency distribution diagram of the emission intensity data of the emission line light of Fe is a bimodal frequency distribution having two peaks. As shown in FIG. 4, the discharge path from the discharge electrode 21 to the sample S follows the shortest path P1 that is substantially perpendicular to the sample surface, and the path P2 that warps to the side of the shortest path P1 (FIG. 4). However, for the sake of convenience, the route P2 is illustrated as one route, but there are actually two types of cases where the route follows a plurality of different routes), and the discharge route differs for each spark discharge. It is thought to be caused by mixing.

  After creating the frequency distribution chart as described above, the calculated Ni / Fe is calculated only for the spark discharge in which the emission intensity data having a numerical value larger than the valley V (see FIG. 3) of the frequency distribution chart is obtained. Extract. In other words, only the spark discharge from which emission intensity data having a numerical value larger than that of the valley V is obtained is made effective, and Ni / Fe for the effective spark discharge is extracted. More specifically, in this embodiment, the knowledge that the cumulative frequency of the emission intensity data having a numerical value larger than that of the valley portion V and the cumulative frequency of the emission intensity data having a numerical value lower than that of the valley portion V are substantially equal. As shown in FIG. 2, Ni / Fe having a frequency corresponding to 50% of all frequencies (corresponding to the total number of spark discharges) is deleted in order from the lowest emission intensity of the Fe emission line light, and the remaining Ni / Fe is extracted (S3, S4).

  However, the present invention is not limited to this. For example, the frequency distribution diagram is configured to be displayed on a monitor included in the microcomputer 7, and the valley V is visually confirmed. It is also possible to configure such that Ni / Fe is extracted only for the spark discharge in which the emission intensity data having a value larger than the valley V is obtained by inputting the emission intensity data with a keyboard or the like. In addition, it is shown in FIG. 7 described above that only the spark discharge in which the emission intensity data having a numerical value larger than that of the valley V is obtained is valid and Ni / Fe is extracted for the effective spark discharge. This corresponds to deleting Ni / Fe in the distribution B and extracting only Ni / Fe in the distribution A.

  Thus, if only the spark discharge from which emission intensity data having a numerical value larger than that of the valley V of the frequency distribution diagram is obtained is effective, the relatively stable large emission of the two types of discharge paths shown in FIG. The spark discharge that follows the shortest path P1 from which the intensity can be obtained is made effective, thereby improving the analysis accuracy.

  Next, as shown in FIG. 2, the average value of Ni / Fe extracted as described above is calculated (S5), and the Ni content is calculated based on the average value. More specifically, for known samples with different Ni contents, the relationship between the average value and the Ni content calculated by chemical analysis in advance is a calibration curve (see FIG. 5; plotted in FIG. 5). The obtained point data is the data used to create the calibration curve) (stored in the microcomputer 7), and the Ni content is calculated using the calibration curve (S6).

  According to the emission spectroscopic analysis method according to the present embodiment described above, the standard deviation of the measurement error (difference between the measurement value and the chemical analysis value) is 0.102%, and all the collected emission intensity data are targeted. It was possible to sufficiently improve the measurement accuracy with respect to 0.162% according to the emission spectroscopic analysis method. Therefore, if the Ni content in the sample S is calculated using the method according to the present embodiment, and the Ni input is adjusted based on the calculated Ni content, the excessive Ni input during steel making is adjusted. Can be reduced and the cost of purchasing raw materials can be greatly reduced.

  In the emission spectroscopic analysis method according to the present embodiment, unnecessary Ni / Fe is deleted (S4 in FIG. 2) after calculating Ni / Fe (S1 in FIG. 2) for all spark discharges. However, the present invention is not limited to this, and unnecessary emission intensity data (emission intensity data of Ni and Fe emission line light) is deleted first, and then Ni / Fe is calculated for the remaining emission intensity data. It is also possible to do.

  In the present embodiment, the aspect of using Fe as a monitor element for normalizing the emission intensity of the Ni emission line light has been described. However, the present invention is not limited to this, as long as it is included in the sample S. It is also possible to use other elements. FIG. 6 shows an example of a frequency distribution diagram of emission intensity data of Mg emission line light contained in the sample S. As shown in FIG. 6, the frequency distribution diagram of the emission intensity data of the Mg emission line light is also a bimodal frequency distribution having two peaks, and thus the emission intensity having a numerical value larger than that of the valley portion of the frequency distribution chart. The measurement accuracy can be improved by making it effective only for the spark discharge from which data was obtained.

FIG. 1 is a diagram showing a schematic configuration of an emission spectroscopic analysis apparatus for carrying out a Ni emission spectroscopic analysis method according to an embodiment of the present invention. FIG. 2 is a flowchart schematically showing calculation contents in the emission spectroscopic apparatus shown in FIG. FIG. 3 shows an example of a frequency distribution diagram of emission intensity data of Fe emission line light created by the emission spectroscopic apparatus shown in FIG. FIG. 4 is an explanatory diagram for explaining a discharge path from the discharge electrode to the sample in the emission spectroscopic apparatus shown in FIG. FIG. 5 shows an example of a calibration curve for calculating the Ni content in the emission spectrometer shown in FIG. FIG. 6 shows an example of a frequency distribution diagram of the emission intensity data of Mg emission line light created by the emission spectroscopic apparatus shown in FIG. FIG. 7 shows an example of the result of emission spectroscopic analysis of the emission intensity of Fe emission line light and the emission intensity of Ni emission line light / emission intensity of Fe emission light in molten steel using Fe in the molten steel as a monitor element. Indicates.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Discharge circuit 3 ... Spectrometer 7 ... Microcomputer 21 ... Discharge electrode
S ... Sample

Claims (4)

An emission spectroscopic analysis method for calculating a Ni content in a sample by performing a plurality of spark discharges between a discharge electrode and a sample containing Ni disposed opposite to the discharge electrode and performing spectral analysis of the discharge light. Because
For each of Ni and other elements contained in the sample, the step of collecting emission intensity data of emission line light for each spark discharge;
Calculating the emission intensity of the bright line light of Ni / the emission intensity of the bright line light of other elements for the plurality of spark discharges;
For the plurality of spark discharges, creating a frequency distribution diagram of emission intensity data of the emission light of the other elements;
Extracting the calculated emission intensity of the emission line light of Ni / the emission line light of other elements only for the spark discharge from which emission intensity data having a value larger than the valley of the frequency distribution diagram was obtained;
Calculating an average value of emission intensity of the extracted Ni emission line light / emission intensity of emission light of other elements;
And a step of calculating a Ni content based on the calculated average value. An emission spectroscopic analysis method for calculating a Ni content in a sample by performing a plurality of spark discharges between a discharge electrode and a sample containing Ni disposed opposite to the discharge electrode and performing spectral analysis of the discharge light. Because
For each of Ni and other elements contained in the sample, the step of collecting emission intensity data of emission line light for each spark discharge;
For the plurality of spark discharges, creating a frequency distribution diagram of emission intensity data of the emission light of the other elements;
Extracting emission intensity data of bright line light of Ni and other elements only for the spark discharge from which emission intensity data having a numerical value larger than the valley of the frequency distribution diagram was obtained,
For the emission intensity data of the extracted Ni and other element emission line light, calculating the emission intensity of Ni emission line light / emission intensity of other element emission line for each spark discharge;
Calculating the average value of the calculated emission intensity of the emission line light of Ni / the emission intensity of the emission line light of other elements;
And a step of calculating a Ni content based on the calculated average value. The Ni emission spectroscopic analysis method according to claim 1 or 2, wherein the sample is an analysis sample sampled from molten steel, and the other element is Fe. A steelmaking method, wherein the Ni content in a sample is calculated using the method according to any one of claims 1 to 3, and the Ni input amount is adjusted based on the calculated Ni content.

Images