JP7458253B2 - Calibration method for elemental analyzer, elemental analyzer, and program for elemental analyzer - Google Patents

Description

The present invention relates to a method for calibrating an elemental analyzer that gasifies and extracts elements contained in a sample, analyzes the extracted gas based on infrared absorption method, and detects the mass of the element contained in the sample. It is something.

For example, for research and development and quality control of materials such as metals and ceramics, the mass % concentration of each element to be analyzed contained in a sample is measured.

The elemental analyzer used for such measurements consists of a heating furnace that gasifies and extracts each element to be analyzed contained in a pre-weighed solid sample, and an NDIR (Non Dispersive Infrared) gas analysis system. A detection mechanism that detects the uncorrected mass of each element contained in the extracted gas using the method), a correction calculation unit that converts the uncorrected mass to the corrected mass using a calibration curve created in advance, and An analysis value calculation unit that calculates the mass % concentration of the corrected mass with respect to the weighed value as a final analysis value.

The calibration curve that is initially set is created by analyzing a standard sample with a known theoretical content of the element to be analyzed, and from the relationship between the uncorrected mass detected by the detection mechanism and the theoretical content, as a proportional equation y = Ax, where x is the uncorrected mass, y is the corrected mass, and A is the gradient coefficient.

However, when an elemental analyzer is operated for a long period of time, the output of the NDIR changes over time, resulting in measurement errors with respect to the actual content and concentration of elements contained in the sample. In such cases, it is possible to recreate the calibration curve using standard samples, but this requires a lot of labor and, because elemental analysis is a destructive measurement, consumes a large amount of standard sample. For this reason, instead of recreating the calibration curve, a simpler method is used to calibrate the calibration curve.

As a method for calibrating a calibration curve of an elemental analyzer, there is an αβ correction method defined in JIS, etc. (see Non-Patent Document 1). Specifically, the method for calibrating an elemental analyzer using infrared absorption method is as follows.

First, the reference time control sample is analyzed at a time when no change occurs in the detection mechanism over time, and the mass % concentration C0 [wt%] calculated by the analysis value calculation unit is recorded.

Next, the calibration control sample is analyzed at the time when a change over time occurs in the detection mechanism, and the mass % concentration C1 [wt%] calculated by the analysis value calculation section is recorded.

Finally, calculate the calibration coefficient α = C0/C1, and use the equation y = A(αx) as the calibration curve after calibration.

However, the calibration method for elemental analyzers as described above calibrates the calibration curve based on the amount of change in mass% concentration [wt%], which is the analysis value, and it is difficult to detect actual changes over time. It does not calibrate the content [mg] of the element to be analyzed, which is the uncorrected mass detected by the mechanism.

Furthermore, such a calibration method can only be applied to a linear equation that passes through the origin. For this reason, such a calibration method cannot be used when the calibration curve is defined as a polynomial in accordance with the characteristics of the detection mechanism.

Furthermore, only one type of control sample is used to calibrate the calibration curve. For this reason, it is not possible to improve calibration accuracy by preparing multiple types of control samples with different theoretical contents or concentrations and performing calibration that takes into account changes in the characteristics of the entire calibration curve.

JIS G 1211-3

The present invention was made with the intention of solving the above-mentioned problems all at once, and it enables the content of the element to be analyzed detected by the detection mechanism itself to be calibrated, and provides a calibration curve that can be calibrated. An object of the present invention is to provide a method for calibrating an elemental analyzer, which is not limited to a proportional type and is capable of improving the calibration accuracy of a calibration curve compared to the conventional method.

That is, the method for calibrating an elemental analyzer according to the present invention includes a heating furnace that gasifies and extracts the element to be analyzed contained in a sample, and a heating furnace that gasifies and extracts the element to be analyzed contained in the sample, and a heating furnace that analyzes the extracted gas based on an infrared absorption method to extract the element contained in the sample. a detection mechanism configured to detect the uncorrected mass of the element contained in the sample and exhaust the gas after analysis, the uncorrected mass detected by the detection mechanism when the sample is a standard sample, and the standard a calibration curve storage unit that stores a calibration curve showing the relationship between the theoretical content of an element to be analyzed contained in a sample; and a correction calculation unit that converts the pre-correction mass to a post-correction mass based on the calibration curve. A method for calibrating an elemental analyzer comprising Calculating the reference content normalized to a value per target weighing value based on the reference time, and detecting the correction by the detection mechanism when the sample is a calibration control sample at the time of calibration after a predetermined period has passed from the reference time. A content at the time of calibration is calculated by normalizing the pre-mass to a value per the target weighing value based on the actual measured weight value of the control sample at the time of calibration, and the content at the time of calibration is calculated based on the reference content and the content at the time of calibration. The method is characterized in that a calibration coefficient for calibrating the calibration curve is calculated.

Furthermore, the elemental analyzer according to the present invention includes a heating furnace that gasifies and extracts the elements to be analyzed contained in the sample, and a heating furnace that gasifies and extracts the elements to be analyzed contained in the sample, and a heating furnace that analyzes the extracted gas based on an infrared absorption method to extract the elements contained in the sample. a detection mechanism configured to detect a pre-correction mass and exhaust gas after analysis; a detection mechanism configured to detect a pre-correction mass and exhaust gas after analysis; a calibration curve storage unit that stores a calibration curve showing the relationship between the theoretical content of the element to be analyzed and the theoretical content of the element to be analyzed; a correction calculation unit that converts the uncorrected mass to a corrected mass based on the calibration curve; Reference content that is obtained by normalizing the uncorrected mass detected by the detection mechanism when the sample is a reference time control sample to a value per target weight value based on the actual measured weight value of the reference time control sample. a reference content calculation unit that calculates the pre-correction mass detected by the detection mechanism when the sample is a calibration control sample at the time of calibration after a predetermined period has elapsed from the reference time; a calibration content calculation unit that calculates a content at the time of calibration normalized to a value per the target weighing value based on the actually measured weighing value; and a calibration curve based on the reference content and the content at the time of calibration. A calibration coefficient calculation unit that calculates a calibration coefficient for calibrating the .

With such a device, it is possible to calibrate the uncorrected mass itself detected by the detection mechanism, which actually changes over time. Therefore, it is not necessary to assume that the calibration curve is a proportional equation as in the past, and it can be defined as a function in various forms.

In addition, the reference content and the calibration content are values that are standardized to the target weight value based on the actual measurement weight value, so even if there is variation in the actual measurement weight value, the target weight value Can be calibrated with a value equivalent to the value. Therefore, it is possible to obtain reliable calibration coefficients that eliminate the influence of variations in actually measured weight values.

Furthermore, since the reference content and the calibration content are standardized values, multiple types of control samples can be specified and used, and the calibration accuracy of the calibration curve can be improved.

In order to calibrate the change in the pre-correction mass due to a change over time in the detection mechanism, the calibration curve may be a function in which the pre-correction mass is an input variable x and the post-correction mass is an output variable y. It is sufficient that the calibration coefficient is one that at least calculates α by which the input variable x is multiplied.

In order to be able to obtain the calibration coefficient based on a pair of the reference content and the content at the time of calibration, the calibration coefficient obtained by analyzing one type of reference time control sample one or more times at the reference time is required. The mass before correction, or the average value of each mass before correction is calculated as one of the reference contents, and the mass before correction is obtained by analyzing one type of calibration control sample once or multiple times during the calibration. Alternatively, the average value of each mass before correction may be calculated as one content at the time of calibration, and the value of the ratio of the content at the time of calibration to the reference content may be calculated as the calibration coefficient α. In addition, when using the standard content and the calibration content by averaging the uncorrected mass obtained from multiple analyzes, the influence of random errors caused by various factors can be alleviated, and further accuracy can be achieved. It is possible to obtain the calibration coefficients that are likely to be the same.

In order to obtain a more reliable calibration coefficient by taking into account all the pairs of the reference content and the content at the time of calibration, the theoretical content or theoretical concentration of the element to be analyzed is determined at the reference time. The above-mentioned mass before correction obtained by analyzing each of a plurality of different types of reference control samples once or multiple times, or the average value of each mass before correction for each type, is calculated as a plurality of reference contents, and the above-mentioned calibration is performed. The mass before correction obtained by analyzing multiple types of standard control samples and the corresponding multiple types of calibration control samples one or more times, or the average value of each type of mass before correction. The calibration coefficient α may be calculated as a plurality of the calibration contents, and the calibration coefficient α may be calculated by the least squares method based on a plurality of pairs of the reference content and the calibration content.

For example, as a calibration method corresponding to a conventionally used proportional type calibration curve, the calibration curve before calibration is a proportional type y=Ax having a slope A, and the calibration coefficient α is composed of a plurality of pairs of calibration coefficients α. Calculating the slope of a regression line passing through the origin of the set of the reference content and the content at the time of calibration by the least squares method, and storing the formula y=A(αx) in the calibration curve storage unit as a calibration curve after calibration. can be mentioned.

In order to be able to take into account intercepts that occur in the calibration curve due to sample matrix effects and blank sample effects, the calibration curve before calibration is a linear function y=Ax+B with slope A and intercept B, and the calibration coefficient As α, the slope of the regression line of the plurality of pairs of the reference content and the calibration content is calculated, and as the calibration coefficient β, the intercept of the regression line is calculated, and the equation y= Any method is sufficient as long as A(αx+β)+B is stored as a calibration curve after calibration.

Even if the calibration curve before calibration is a polynomial of the input variable x, calibration can be made possible by calculating the slope of the regression line of the pairs of the reference content and the content at the time of calibration as the calibration coefficient α, calculating the intercept of the regression line as the calibration coefficient β, and storing the equation in which the input variable is converted to (αx + β) in the calibration curve storage unit as the calibration curve after calibration.

As a specific calibration method when the calibration curve is a polynomial, the calibration curve before calibration is a quadratic function y=Ax ² +Bx+C having a quadratic coefficient A, a linear coefficient B, and a constant term C. , the reference content, which is the average value of each uncorrected mass obtained by analyzing one type of standard control sample multiple times, and each correction obtained by analyzing one type of calibration control sample multiple times. The value of the ratio of the contents at the time of calibration, which is the average value of the pre-mass, is calculated as the calibration coefficient α, and the formula y=A(αx) ² +B(αx)+C is stored in the calibration curve storage unit as the calibration curve after calibration. Examples include things that can be memorized.

In order to be able to calibrate the calibration curve obtained by subtracting the blank sample influence from the mass before correction, the calibration curve before calibration has a slope of A, and the element to be analyzed contained in the sample is gasified when the calibration curve is created. y=A(x-D), which has the uncorrected mass or its average value D output by the detection mechanism when the detection mechanism has not extracted the sample, and gasifies the element to be analyzed contained in the sample during the calibration. The uncorrected mass or its average value D1 that is output by the detection mechanism when the detection mechanism has not extracted is the value obtained by subtracting D from each uncorrected mass obtained by analyzing one type of reference control sample multiple times. The ratio of the standard content, which is the average value of It is sufficient if the value is calculated as the calibration coefficient α and the equation y=A(α(x−D1)) is stored in the calibration curve storage unit as the calibration curve after calibration.

In order to display how much of the target element was contained in the sample before being analyzed in a form that is easy to use in material analysis, quality control, etc., the corrected mass converted by the correction calculation section is Any device may be used as long as it further includes an analysis value calculation section that calculates the mass % concentration of the element contained in the sample by dividing it by the weighed value.

In order to be able to easily use the calibration method of an elemental analyzer according to the present invention by simply updating the program in an existing elemental analyzer, it is possible to use heating to gasify and extract the element to be analyzed contained in the sample. a furnace; a detection mechanism configured to analyze the extracted gas based on infrared absorption to detect the uncorrected mass of the element contained in the sample; and exhaust the gas after the analysis; and the sample. a calibration curve storage unit that stores a calibration curve indicating the relationship between the uncorrected mass detected by the detection mechanism and the theoretical content of the element to be analyzed contained in the standard sample when is a standard sample; , a correction calculation unit that converts the uncorrected mass into a corrected mass based on the calibration curve, and the program is used in an elemental analyzer, where the sample is a reference time control sample at a reference time. a reference content calculation unit that calculates a reference content in which the uncorrected mass detected by the detection mechanism is normalized to a value per target weight value based on the actual measured weight value of the control sample at reference time; When the sample is a calibration control sample at the time of calibration after a predetermined period has elapsed since then, the uncorrected mass detected by the detection mechanism is calculated based on the actual weight value of the calibration control sample per the target weight value. a calibration content calculation unit that calculates a content at the time of calibration standardized to a value; and a calibration coefficient calculation unit that calculates a calibration coefficient for calibrating the calibration curve based on the reference content and the content at the time of calibration. What is necessary is to use a program for an elemental analyzer that is characterized by causing a computer to perform the functions of and.

The program for the elemental analysis device may be distributed electronically or may be recorded on a program recording medium such as a CD, DVD, or flash memory.

In this way, the calibration method for an elemental analyzer according to the present invention directly calibrates the uncorrected mass detected by the detection mechanism, and can achieve more accurate calibration than the conventional method of calculating a calibration coefficient based on the ratio of mass percent concentrations, which are the analytical values before and after calibration. In addition, since the reference content and the content at the time of calibration are values normalized to a value per target weighing value based on the actual weighing value, variations in the actual weighing value are less likely to affect the calibration.

FIG. 1 is a schematic diagram showing an elemental analyzer according to a first embodiment of the present invention. FIG. 2 is a functional block diagram showing the configuration of a control calculation mechanism according to the first embodiment. 5 is a flowchart showing operations related to analysis of a control sample at a reference time for calibrating the elemental analyzer of the first embodiment. 4 is a flowchart showing operations related to analysis of a control sample and updating of a calibration curve during calibration of the elemental analyzer of the first embodiment. 10 is a flowchart showing an operation related to the analysis of a control sample at a reference time for calibrating an elemental analyzer of a second embodiment. 7 is a flowchart showing operations related to analysis of a control sample and updating of a calibration curve during calibration to calibrate the elemental analyzer of the second embodiment. 13 is a flowchart showing an operation related to the analysis of a control sample at a reference time for calibrating the elemental analyzer of the third embodiment. 13 is a flowchart showing operations related to analysis of a control sample and updating of a calibration curve during calibration of the elemental analyzer of the third embodiment. 12 is a flowchart showing operations related to analysis of a control sample and updating of a calibration curve during calibration to calibrate the elemental analyzer of the fourth embodiment. 12 is a flowchart showing operations related to analysis of a control sample at a reference time for calibrating the elemental analyzer of the fifth embodiment. 12 is a flowchart showing operations related to analysis of a control sample and updating of a calibration curve during calibration to calibrate the elemental analyzer of the fifth embodiment.

An elemental analysis apparatus 100 according to a first embodiment of the present invention will be described with reference to each figure.

<Description of configuration>
The elemental analyzer 100 heats and melts, for example, a metal sample or a ceramic sample (hereinafter simply referred to as a sample) contained in a graphite crucible, and measures the amount of elements contained in the sample by analyzing the sample gas generated during the process. In the first embodiment, the elements to be measured are O (oxygen), H (hydrogen), and N (nitrogen) contained in the sample.

Specifically, as shown in FIG. 1, the elemental analyzer 100 includes a heating furnace 2 in which a sample accommodated in a crucible MP is heated, gasifies and extracts the element to be analyzed, and a carrier for the heating furnace 2. It includes an introduction channel L1 through which gas is introduced, and an outlet channel L2 through which a mixed gas of carrier gas and sample gas is led out from the heating furnace. More specifically, the elemental analyzer 100 includes a heating furnace, each device provided in the introduction channel L1 or the derivation channel L2, and a control calculation mechanism that controls each device and performs calculation processing of measured concentrations, etc. It is configured by COM.

Each part will be explained in detail.

A gas cylinder serving as a carrier gas supply source 1 is connected to the base end of the introduction channel L1. In the first embodiment, He (helium) is supplied from the supply source 1 into the introduction channel L1.

The heating furnace 2 is configured so that a graphite crucible MP containing a sample is held between a pair of electrodes, and a current is passed directly through the crucible MP to heat the crucible MP and the sample. O, H, and N, which are the elements to be analyzed, are extracted from the sample in the form of CO gas, H ₂ gas, and N ₂ gas, respectively.

Next, each device provided on the outlet channel L2 will be explained.

As shown in FIG. 1, a dust filter 3, a CO detector 4, an oxidizer 5, a _CO2 detector 6, an _H2O detector 7, a removal mechanism 8, and an _N2 detector 9 are arranged in this order from the upstream side on the outlet flow path L2.

Each part will be explained in detail.

The dust filter 3 filters and removes dust such as soot contained in the mixed gas.

The CO detection unit 4 detects CO (carbon monoxide) contained in the mixed gas that has passed through the dust filter 3 and measures its concentration. It is configured to output a voltage signal according to the concentration of CO. The CO detection unit 4 operates effectively when the oxygen contained within the sample is high in concentration due to its measurement accuracy. Specifically, it is preferable to measure CO with a concentration of 150 ppm or more.

The oxidizer 5 oxidizes CO and CO ₂ contained in the mixed gas that has passed through the CO detection unit 4, and also oxidizes H ₂ to H ₂ O (water) to generate water vapor.

The _CO2 detector 6 detects _CO2 in the mixed gas that has passed through the oxidation reactor 6 and measures its concentration, outputting a signal according to the concentration of _CO2 in the mixed gas by NDIR. From the viewpoint of measurement accuracy, the _CO2 detector 6 operates effectively when the oxygen contained in the sample is at a low concentration (for example, less than 150 ppm).

The H ₂ O detector 7 detects H ₂ O in the mixed gas that has passed through the _CO2 detector 6 to measure its concentration, and outputs a voltage signal according to the concentration of H ₂ O in the mixed gas by NDIR.

The removal mechanism 8 adsorbs and removes CO ₂ and H ₂ O contained in the mixed gas.

The _N2 detection unit 9 is a TCD (thermal conductivity detector), and detects N2, which is a predetermined component contained in the mixed gas, based on changes in the thermal conductivity of the mixed gas and the flow rate of the supplied mixed _gas. Measures the concentration in a mixed gas. In other words, since the mixed gas supplied to the _N2 detection unit 9 consists almost only of the carrier gas and _N2 , the concentration of _N2 contained in the mixed gas corresponds to the change in the measured thermal conductivity. value. Note that the mixed gas that has passed through the N ₂ detection section 9 is not recovered but is exhausted. Therefore, the elemental analysis of the first embodiment is a destructive analysis in which the sample disappears.

The control calculation mechanism COM is, for example, a so-called computer equipped with a CPU, memory, A/D converter, D/A converter, and various input/output means, and programs stored in the memory are executed and various devices cooperate. Thus, the analysis operation or calibration operation of the elemental analyzer 100 is realized. Specifically, as shown in the functional block diagram of FIG. 2, the control calculation mechanism COM includes at least a reception section C1, a pre-correction mass calculation section C2, a calibration curve storage section C3, a correction calculation section C4, an _N2 content calculation section C5, It functions as an analysis value calculation unit C6, a reference content calculation unit C7, a content calculation unit during calibration C8, a calibration data storage unit C9, a calibration coefficient calculation unit C10, and a calibration curve update unit C11.

We will now go into detail about each part. First, we will explain the configuration related to normal analysis.

The receiving unit C1 receives a mode setting indicating in which mode the elemental analyzer 100 is to be operated from the user. The control calculation mechanism COM is configured to change its operation depending on the accepted mode setting. Mode settings include, for example, a calibration curve creation mode, a normal analysis mode, a reference data acquisition mode, a calibration curve calibration mode, and the like. The receiving unit C1 also receives data such as weighing values, etc., which are necessary for calculations in various modes.

The pre-correction mass calculation unit C2 calculates the mixed gas of each gas component from the voltage signals output from the CO detection unit 4, the CO ₂ detection unit 6, and the H ₂ 0 detection unit 7, which detect the concentration of each gas based on NDIR. At the same time, the flow rate of the mixed gas is obtained from a flow rate sensor (not shown). The pre-correction mass calculation unit C2 calculates the pre-correction mass of O or H, which is the element to be analyzed, contained in the sample from the integrated value of the concentration and flow rate at each time. Note that the CO detection unit 4, the CO ₂ detection unit 6, the H ₂ 0 detection unit 7, and the pre-correction mass calculation unit C2 constitute a detection mechanism DMDM.

The calibration curve storage unit C3 stores a calibration curve that indicates the relationship between the uncorrected mass detected by the detection mechanism DM and the theoretical content of the element to be analyzed contained in the standard sample when the sample is a standard sample. do. In the first embodiment, the calibration curve is a proportional equation y=Ax in which the input variable is the mass before correction x, the output variable is the mass after correction y, and the slope coefficient A. Here, the mass before correction x is the mass of O or H calculated from the output of each detection unit 4, 6, and 7, and the gradient coefficient A is a different value for each element.

The correction calculation unit C4 converts the pre-correction mass of O or H into the post-correction mass of O or H based on the respective calibration curves.

The N ₂ content calculation unit C5 calculates the mass of N contained in the sample from the integrated value of the N ₂ gas concentration detected by the N ₂ detection unit 9 and the flow rate of the mixed gas.

The analysis value calculation section C6 calculates the measured weight value of the sample received by the reception section C1, the corrected mass of O or H output from the correction calculation section C4, and the mass of N calculated from the _N2 content calculation section C5. Based on this, the mass % concentration [wt%] of each element in the sample is calculated. This calculated value is output to the user as a final analysis value.

Next, things related to calibration of the calibration curve will be explained. Each section described below functions, for example, when the user operates the elemental analyzer 100 in a reference data acquisition mode or a calibration curve calibration mode by setting a mode.

The reference content calculation unit C7 calculates reference data indicating a state before a change over time occurs in the detection mechanism DM, which is necessary when calibrating the calibration curve when set to the reference data acquisition mode. More specifically, the reference content calculation unit C7 calculates the analysis detected by the detection mechanism DM when a reference time control sample is installed in the heating furnace 2 at the reference time and elemental analysis of the reference time control sample is performed. The mass before correction of O or H, which is the target element, and the target weight value and the actual weight value of the reference control sample accepted from the user are acquired. Further, the reference content calculation unit C7 calculates a reference content in which the pre-correction mass of the reference time control sample is normalized to a value per target weight value based on the actually measured weight value of the reference time control sample. In the first embodiment, the uncorrected mass obtained from the detection mechanism DM when analyzing the reference control sample is multiplied by m0 [mg], the target weight value by M [g], and the actual measured weight value by M0 [g]. When the symbol is * and the division symbol is /, the standardized reference content can be expressed as m0*M/M0.

The calibration content calculation unit C8 calculates data indicating the state after a change over time occurs in the detection mechanism DM, which is necessary when calibrating the calibration curve when set to the calibration curve calibration mode. More specifically, the calibration content calculation unit C8 calculates that the calibration control sample is installed in the heating furnace 2 after a predetermined period of time has elapsed from the reference time, and elemental analysis of the calibration control sample is performed. In this case, the uncorrected mass of O or H, which is the element to be analyzed detected by the detection mechanism DM, and the target weight value and actual weight value of the calibration control sample accepted from the user are acquired. Note that the calibration control sample used during calibration is, for example, a sample cut from the same base material as the reference control sample used at the reference time, and has approximately the same composition, or a sample that is guaranteed to have the same composition. be. Further, the target weight value is the same value as the target weight value of the reference time control sample analyzed at the reference time. Further, the calibration content calculation unit C8 calculates the calibration content in which the pre-correction mass of the calibration management sample is normalized to a value per target weight value, based on the actually measured weight value of the calibration management sample. In the first embodiment, when a calibration control sample is analyzed, the mass before correction obtained from the detection mechanism DM is m1 [mg], the target weight value is M [g], and the actual weight value is M1 [g]. In this case, the normalized calibration content can be expressed as m1*M/M1.

The calibration data storage unit C9 stores the corresponding reference content and content at the time of calibration. For example, the reference content measured at the time of creating the calibration curve is stored at that time.

The calibration coefficient calculation unit C10 calculates a calibration coefficient for calibrating the calibration curve based on the reference content obtained at the reference time and the content at the time of calibration obtained at the time of calibration. In the first embodiment, the calibration coefficient calculation unit C10 calculates the ratio of the content at the time of calibration to the reference content as the calibration coefficient α. That is, the calibration coefficient α can be expressed as α=(m0*M/M0)/(m1*M/M1).

The calibration curve updating unit C11 calibrates the calibration curve y=Ax before calibration stored in the calibration curve storage unit C3 based on the calibration coefficient α. In the first embodiment, the equation y=A(αx) obtained by multiplying the input variable x by α is stored in the calibration curve storage unit C3 as a new calibration curve.

<Explanation regarding calibration>
The operation related to the calibration of the elemental analyzer 100 configured in this way, the user's operations, etc. will be explained with reference to the flowcharts of FIGS. 3 and 4.

First, with reference to Figure 3, we will explain the procedure for calculating the standard content from the reference time control sample, assuming that the reference time is immediately after the calibration curve is created based on the standard sample.

The user places a reference control sample weighed to the target weight value into the crucible MP and places it in the heating furnace 2 (step S1). Next, the elemental analyzer 100 is operated to start elemental analysis (step S2). The reference content calculation unit C7 obtains the uncorrected mass m0 of H or O detected by the detection mechanism DM through elemental analysis (step S3), and also obtains the target of the reference time control sample accepted by the user in the reception unit C1. The weighed value M and the measured weighed value M0 are acquired (step S4). The reference content calculation unit C7 calculates the reference content m0*M/M0, which is the normalized mass before correction, based on each acquired value (step S5). The calculated reference content m0*M/M0 is stored in the calibration data storage section C9 (step S6).

Next, the time when a certain period of time has passed since the reference time and a change over time is occurring in the detection mechanism DM is defined as the calibration time. The procedure for calculating the calibration time content using the calibration time control sample at the calibration time and the procedure for calibrating the calibration curve will be explained with reference to Figure 4.

The procedure for calculating the calibration content corresponds to the procedure for calculating the reference content shown in FIG. 3 (steps S1 to S6). That is, as shown in FIG. 4, the user places the calibration control sample, which has been weighed to obtain the target weight value, in the crucible MP and places it in the heating furnace 2 (step S7). Next, the elemental analysis device 100 is operated to start elemental analysis (step S8). The calibration content calculation unit C8 acquires the uncorrected mass m1 of H or O detected by the detection mechanism DM through elemental analysis (step S9), and also acquires the target weight value M and the actual weight value M1 of the calibration control sample accepted by the acceptance unit C1 from the user (step S10). The calibration content calculation unit C8 calculates the calibration content m1*M/M1, which is the uncorrected mass standardized based on the acquired values (step S11). The calculated calibration content m1*M/M1 is stored in the calibration data storage unit C9 (step S12).

Next, the calibration coefficient calculation unit C10 calculates the ratio of the content at the time of calibration to the reference content (m0*M/M0)/(m1*M/M1) as the calibration coefficient α (step S13). Finally, the calibration curve updating unit C11 updates the calibration curve stored in the calibration curve storage unit C3 to y=A(αx) (step S14).

<Explanation of effects>
In such a method of calibrating the elemental analyzer 100 of the first embodiment, the detection mechanism DM based on the infrared absorption method, which changes over time, detects the concentration rather than the mass % concentration [wt%] calculated as the analysis value. Since the calibration curve is calibrated based on the mass before correction [mg], a more reliable calibration coefficient α can be obtained.

Further, regarding the variation in the actual measured weight value, since the detected mass before correction is normalized to a value per the target weight value, it is possible to prevent the reference content and the content at the time of calibration from being affected.

Next, a method for calibrating the elemental analyzer 100 according to the second embodiment of the present invention will be described with reference to FIGS. 5 and 6. In the following description, members corresponding to those described in the first embodiment will be given the same reference numerals.

<Explanation regarding calibration>
The calibration method of the elemental analysis apparatus 100 of the second embodiment differs from the calibration method of the first embodiment in that one type of reference time control sample with the same theoretical content or theoretical concentration is analyzed multiple times to calculate the reference content, and in that one type of calibration time control sample with the same theoretical content or theoretical concentration is analyzed multiple times to calculate the calibration time content.

First, the procedure for calculating the reference content will be explained with reference to the flowchart of FIG. 5.

The user puts a standard control sample weighed to the target weight value into the crucible MP and places it in the heating furnace 2 (step ST1). Next, the elemental analyzer 100 is operated to start elemental analysis (step ST2). The reference content calculation unit C7 acquires the uncorrected mass m0i of H or O detected by the detection mechanism DM by elemental analysis (step ST3), and also acquires the mass m0 _i of the reference time control sample received from the user in the reception unit C1. A target weight value M and an actual weight value _M0i are acquired (step ST4). Here, the suffix i indicates that it was obtained during the i-th analysis. The reference content calculation unit C7 calculates a value m0 _i *M/M0 _i , which is the normalized mass before correction, based on each acquired value (step ST5).

Further, if the target number of analyzes has not been completed (step ST6), the same type of reference time control sample is weighed again to the target weight value M, and the operations of steps ST1 to ST5 are repeated. When the analysis of the target number of reference time control samples is completed, the reference content calculation unit C7 calculates the average value of the normalized values of the uncorrected masses obtained in each analysis as the reference content (step ST7). . That is, the reference content can be expressed as AVE (m0 _i *M/M0 _i ) (here, AVE ( ) indicates the average value of the values in ( )). The calculated reference content is stored in the calibration data storage section C9 (step ST8).

Next, the procedure for calculating the content at the time of calibration and the procedure for calibrating the calibration curve will be described with reference to the flowchart of FIG. 6.

The procedure for calculating the content at the time of calibration corresponds to the procedure for calculating the content at the reference time (steps ST1 to ST8) shown in FIG. That is, the user puts the calibration control sample weighed to the target weight value into the crucible MP and places it in the heating furnace 2 (step ST9). Next, the elemental analyzer 100 is operated to start elemental analysis (step ST10). The calibration content calculation unit C8 acquires the uncorrected mass _m1i of H or O detected by the detection mechanism DM by elemental analysis (step ST11), and also obtains the calibration control sample accepted by the user in the reception unit C1. The target weighing value M and the actual measured weighing value _M1i are obtained (step ST12). Here, the suffix i indicates that it was obtained during the i-th analysis. The calibration content calculation unit C8 calculates a value obtained by normalizing the mass before correction, m1 _i *M/M1 _i , based on each acquired value (step ST13).

Furthermore, if the target number of analyses has not been completed (step ST14), the same type of calibration control sample is weighed again to reach the target weighing value M, and the operations of steps ST9 to ST13 are repeated. When the target number of analyses of the calibration control sample are completed, the reference content calculation unit C7 calculates the average value of the normalized values of the uncorrected masses obtained in each analysis as the reference content (step ST15). In other words, the calibration content can be expressed as AVE( _m1i *M/ _M1i ). The calculated calibration content is stored in the calibration data storage unit C9 (step ST16).

Finally, the calibration coefficient calculation unit C10 calculates the value of the content at the time of calibration with respect to the reference content as the calibration coefficient α (step ST17). That is, the calibration coefficient can be expressed as α=AVE(m0 _i *M/M0 _i )/AVE(m1 _i *M/M1 _i ). Further, the calibration curve updating unit C11 uses the obtained calibration coefficient α to store y=A(αx) in the calibration curve storage unit C3 as a new calibration curve.

<Explanation of effects>
In this way, in the method of calibrating the elemental analyzer 100 of the second embodiment, the reference content and the content at the time of calibration can be calculated using values obtained by analyzing control samples with the same theoretical content or theoretical concentration multiple times. Since it is calculated as an average value, it is possible to reduce the influence of random errors caused by various factors and obtain a more reliable calibration coefficient α.

Next, a modification of the second embodiment will be described. In this modification, the calibration curve before calibration is a quadratic function y=Ax ² +Bx+C having a quadratic coefficient A, a first coefficient B, and a constant term C. Even when the calibration curve is a quadratic function as described above, it can be calibrated by multiplying each input variable x by the calibration coefficient α obtained by the calibration method of the second embodiment. That is, the calibration curve after calibration is stored in the calibration curve storage unit C3 as the formula y=A(αx) ² +B(αx)+C.

Next, a method for calibrating the elemental analyzer 100 of the third embodiment will be described with reference to the flowcharts of FIGS. 7 and 8. Note that steps that are the same as those in the calibration method of the second embodiment are given the same numbers, and steps that are different are given -1. In addition, different numbers are given to steps that did not exist in the calibration method of the second embodiment.

The calibration method of the elemental analyzer 100 of the third embodiment differs from the second embodiment in that a reference content is calculated by analyzing multiple types of reference time control samples (first reference time control sample, second reference time control sample, ... nth reference time control sample) with different theoretical content or theoretical concentration multiple times, and a calibration content is calculated by analyzing multiple types of calibration time control samples (first calibration time control sample, second calibration time control sample, ... nth calibration time control sample) with different theoretical content or theoretical concentration multiple times. It also differs from the second embodiment in that the calibration coefficient calculation unit C10 calculates the calibration coefficient α by the least squares method using multiple pairs of reference content and calibration time content.

First, the procedure for calculating the reference content will be explained with reference to the flowchart of FIG. 7.

<Explanation regarding calibration>
The procedure for calculating the average uncorrected mass normalized by analyzing a certain type of reference control sample (jth reference control sample) multiple times in steps ST1 to ST7 as the reference content _m0ij *M/ _M0ij is the same as in the second embodiment, and therefore will not be described again, where i indicates the value obtained in the i-th analysis of a certain type of reference control sample (jth reference control sample), and j is an identifier indicating the type of reference control sample.

The calculated reference content is stored in the calibration data storage section C9 together with an identifier indicating the type of the reference time control sample (step ST8-1). If the analysis has not been completed for all types of reference time control samples (step STA), the type of reference time control sample is changed (step STB), and steps ST1 to ST8-1 are repeated. Finally, AVE (m0 _i1 *M/M0 _i1 ), AVE (m0 _i2 *M/M0 _i2 ), AVE (m0 _i3 *M/M0 _i3 ), etc. are stored in the calibration data storage unit C9. Ru.

Next, the procedure for calculating the content during calibration and the procedure for calibrating the calibration curve will be explained with reference to the flowchart in Figure 8.

The procedure for calculating the content at the time of calibration corresponds to the procedure for calculating the content at the reference time (steps ST1 to STA) shown in FIG. Further, from step ST9 to step ST15 in FIG. 8, a procedure for calculating the average value of the normalized mass before correction by analyzing a certain type of calibration control sample multiple times as the reference content m1 _ij *M/M1 _ij is the same as the second embodiment, so the explanation will be omitted. Here, i indicates a value obtained in the i-th analysis of a certain type of calibration control sample (jth calibration control sample), and j is an identifier indicating the type of calibration control sample. In addition, the types of calibration control samples (calibration 1st control sample, calibration 2nd control sample, ... calibration nth control sample) during calibration are the calibration control samples used at the reference time (calibration control sample 2). 1 control sample, 2nd control sample at the time of calibration, . . . nth control sample at the time of calibration). That is, the samples are selected so that there are pairs of reference control samples and calibration control samples that have the same theoretical content or theoretical concentration of the element to be analyzed.

The calculated calibration content is stored in the calibration data storage unit C9 together with an identifier indicating the type of the calibration control sample (step ST16-1). If analysis of all types of calibration control samples has not been completed (step STC), the type of calibration control sample is changed (step STD) and steps ST9 to ST16-1 are repeated. Finally, AVE(m1 _i1 *M/M1 _i1 ), AVE(m1 _i2 *M/M1 _i2 ), AVE(m1 _i3 *M/M1 _i3 ), ... are stored in the calibration data storage unit C9.

Finally, the calibration coefficient calculation unit C10 calculates a regression line passing through the origin of the plurality of pairs of reference contents and reference contents using the least squares method. The calibration coefficient calculation unit C10 outputs the slope of the regression line as the calibration coefficient α (step ST17-1). Further, the calibration curve updating unit C11 uses the obtained calibration coefficient α to store y=A(αx) as a new calibration curve in the calibration curve storage unit C3 (step ST18).

In this way, according to the method for calibrating the elemental analyzer 100 of the third embodiment, the calibration coefficient α can be determined based on various values of the mass before correction output from the detection mechanism DM. Even if the change over time differs depending on the size of the previous mass value, it is possible to multiply by the most appropriate calibration coefficient.

Next, a method for calibrating the elemental analyzer 100 of the fourth embodiment will be described with reference to the flowchart in FIG.

The calibration method of the elemental analyzer 100 of the fourth embodiment is that the calibration curve to be calibrated is not a proportional equation but a linear equation y=Ax+B, and that the calibration coefficient calculation unit C10 uses multiple pairs of reference contents and This embodiment differs from the third embodiment in that two calibration coefficients α and β are calculated from the content regression line.

<Explanation regarding calibration>
The procedure for calculating the reference content is the same as in the third embodiment and will therefore be omitted. Further, the procedure for calculating the content at the time of calibration in steps ST9 to STC in FIG. 8 is also omitted because it is the same as in the third embodiment.

As shown in FIG. 9, the calibration coefficient calculation unit C10 uses the least squares method to calculate the slope of the regression line as the calibration coefficient α, and also calculates the intercept of the regression line as the calibration coefficient β. (Step ST17-2). Then, the calibration curve updating unit C11 stores y=A(αx+β)+B as a new calibration curve in the calibration curve storage unit C3 (step ST18-2).

<Explanation of effects>
With such a method of calibrating the elemental analyzer 100 of the fourth embodiment, even if the calibration curve is defined as a linear function in consideration of blank effects, etc., the slope and intercept of the calibration curve can be calibrated. be able to.

Next, a method for calibrating the elemental analyzer 100 of the fifth embodiment will be described with reference to FIGS. 10 and 11.

The calibration method of the elemental analysis apparatus 100 of the fifth embodiment differs from that of the third embodiment in that the calibration curve is defined as the equation y = A(x-D) with the blank content D subtracted, and that the value obtained by subtracting the blank content from the uncorrected mass is normalized by the weighing value.

<Explanation regarding calibration>
The calibration procedure of the fifth embodiment is partially different from the calibration procedure of the second embodiment shown in FIGS. 5 and 6. Here, only the different parts will be explained.

As shown in FIG. 10, before calculating the reference content, the blank content D is measured in advance, for example, when creating a calibration curve. That is, the user places only the crucible MP without a sample in the heating furnace 2 (step STE), and elemental analysis is started (step STF). The reference content calculation unit C7 obtains the uncorrected mass detected by the detection mechanism DM as the blank content D (step STG). Steps ST1 to ST4 are the same as those in the second embodiment, and as shown in step ST5-1, the reference content calculation unit C7 normalizes the value obtained by subtracting the blank content D from the uncorrected mass m0 _i to a value (m0 _i -D) * M/M0 _i per target weighing value M based on the actual weighing value M0 _i . In addition, the average value AVE ((m0 _i -D) * M/M0 _{i )} of the value (m0 _i -D) * M/M0 _i calculated in step ST5-1 is calculated as the reference content (step ST7-1).

As shown in FIG. 11, when the calibration content is calculated, the blank content D1 at the time of calibration is also measured. That is, the user places only the crucible without a sample in the heating furnace 2 (step STH), and elemental analysis is started (step STI). The calibration content calculation unit C8 obtains the uncorrected mass detected by the detection mechanism DM as the blank content D1 (step STJ). Steps ST10 to ST12 are the same as those in the second embodiment, and as shown in step ST13-1, the calibration content calculation unit C8 normalizes the value obtained by subtracting the blank content D1 from the uncorrected mass m1 _i to a value (m1 _i -D1)*M/M1 _i per target weighing value M based on the actual weighing value M1 i. In addition, the average value AVE ((m1 _i -D1)*M/M1 _{i )} of the value (m1 _i -D1)*M/M1 _i calculated in step ST13-1 is calculated as the calibration content (step ST15-1).

Finally, calibration coefficient calculation unit C10 calculates the ratio of the content at calibration to the reference content, AVE(( _m0i -D)*M/ _M0i )/AVE((m1i _- D1)*M/ _M1i ), as the calibration coefficient α (step ST17). Then, calibration curve update unit C11 uses the calibration coefficient α to store the equation y=A(α(x-D1)) as a new calibration curve in calibration curve memory unit C3.

<Explanation of effects>
As described above, the method for calibrating the elemental analyzer 100 of the fifth embodiment allows calibration even when the calibration curve is created by subtracting blank components. Note that the measurement of the blank content D is not limited to the case where elemental analysis is performed in a state where only the crucible MP is installed without a sample, and any value that is measured during elemental analysis in a state where the zero point is desired may be used. That is, when the heating furnace 2 performs impulse heating as in the fifth embodiment and elemental analysis is performed with flux added together with the sample, elemental analysis is performed with only the flux added in the crucible MP. The blank content D may be measured by performing the following steps. In addition, when a sample is high-frequency heated in a heating furnace and a combustion improver is used together with the sample, or when a combustion improver and pure iron are used together with the sample, elemental analysis of only the combustion improver without the sample, or The blank content D may be measured by elemental analysis of only the fuel and pure iron.

Other embodiments will be described.

The elements to be analyzed, the mass of which is detected by the detection mechanism using infrared absorption, are not limited to O and H. For example, the elements to be analyzed may be C (carbon) and S (sulfur). That is, the elemental analysis device is not limited to that shown in FIG. 1, and the detection mechanism may be configured to detect the uncorrected mass of C and S. More specifically, in the above-mentioned embodiments, in order to measure the elements to be analyzed, O and H, the sample in the graphite crucible is melted by impulse heating while an inert gas such as Ar or He is introduced into the heating furnace, and the component gas is extracted. In contrast, in order to measure the elements to be analyzed, C and S, the component gas may be obtained by burning the sample contained in the ceramic crucible by high-frequency induction heating while introducing a supporting gas into the heating furnace. In addition, in order to measure the elements to be analyzed, C and S, the sample contained in the quartz boat may be burned by electrical resistance heating while introducing a supporting gas into the heating furnace. As described above, even if the heating method and the type of gas introduced and the crucible are different, the calculation process of the mass using the calibration curve is almost the same, so the calibration curve can be calibrated by the calibration method according to the present invention.

The elemental analyzer may also be one that analyzes only one type of element using infrared absorption.

In addition, various modifications of the embodiments or combinations of parts of each embodiment may be made as long as they do not go against the spirit of the present invention.

100...Elemental analyzer 2...Heating furnace DM...Detection mechanism C3...Calibration curve storage section C4...Correction calculation section C6...Analysis value calculation section C7...Standard content calculation Part C8... Calibration content calculation part C10... Calibration coefficient calculation part

Claims (12)

A heating furnace that gasifies and extracts the element to be analyzed contained in the sample, and a heating furnace that analyzes the extracted gas based on infrared absorption method to detect the uncorrected mass of the element contained in the sample, and extracts the gas after analysis. a detection mechanism configured to exhaust the sample, the uncorrected mass detected by the detection mechanism when the sample is a standard sample, and the theoretical content of the element to be analyzed contained in the standard sample. A method for calibrating an elemental analyzer, comprising: a calibration curve storage section that stores a calibration curve showing a relationship between ,
A reference content in which the uncorrected mass detected by the detection mechanism when the sample is a reference time control sample at the reference time is normalized to a value per target weight value based on the actual measured weight value of the reference time control sample. Calculate the amount,
When the sample is a calibration control sample at the time of calibration after a predetermined period has elapsed from the reference time, the pre-correction mass detected by the detection mechanism is set to the target weight value based on the actually measured weight value of the calibration control sample. Calculate the content at the time of calibration normalized to the permissible value,
A method for calibrating an elemental analyzer, comprising calculating a calibration coefficient for calibrating the calibration curve based on the reference content and the content at the time of calibration. The calibration curve is a function in which the pre-correction mass is an input variable x and the post-correction mass is an output variable y,
The method for calibrating an elemental analyzer according to claim 1, further comprising calculating at least α by which the input variable x is multiplied as the calibration coefficient. Calculating the pre-correction mass obtained by analyzing one type of reference time control sample one or more times at the reference time, or the average value of each pre-correction mass as one reference content,
Calculate the pre-correction mass obtained by analyzing one type of calibration control sample one or more times during the calibration, or the average value of each pre-correction mass as one content at the time of calibration,
3. The method for calibrating an elemental analyzer according to claim 2, wherein a value of the ratio of the content at the time of calibration to the reference content is calculated as the calibration coefficient α. The mass before correction obtained by analyzing multiple types of reference time control samples each having different theoretical content or theoretical concentration of the element to be analyzed at the reference time once or multiple times, or each mass before correction for each type. Calculate the average value of the plurality of reference contents,
The mass before correction obtained by analyzing each of the plurality of types of standard time control samples and the corresponding plurality of types of calibration control samples at the time of calibration one or more times, or the average of each type of mass before correction. Calculate the value as a plurality of contents at the time of calibration,
3. The method for calibrating an elemental analyzer according to claim 2, wherein the calibration coefficient α is calculated by a least squares method based on a plurality of pairs of the reference content and the content at the time of calibration. The calibration curve before calibration is a proportional equation y=Ax with a slope A,
Calculating the slope of a regression line passing through the origin of a plurality of pairs of the reference content and the calibration content as the calibration coefficient α by the least squares method;
5. The method for calibrating an elemental analyzer according to claim 4, wherein the equation y=A(αx) is stored in the calibration curve storage unit as a calibration curve after calibration. The calibration curve before calibration is a linear function y=Ax+B having a slope A and an intercept B,
Calculating the slope of a regression line of a plurality of pairs of the reference content and the calibration content as the calibration coefficient α, and calculating the intercept of the regression line as the calibration coefficient β;
5. The method for calibrating an elemental analyzer according to claim 4, wherein the equation y=A(αx+β)+B is stored in the calibration curve storage unit as a calibration curve after calibration. The calibration curve before calibration is a polynomial of input variable x,
Calculating the slope of a regression line of a plurality of pairs of the reference content and the calibration content as the calibration coefficient α, and calculating the intercept of the regression line as the calibration coefficient β;
5. The method for calibrating an elemental analyzer according to claim 4, wherein an equation obtained by converting an input variable into (αx+β) is stored in the calibration curve storage unit as a calibration curve after calibration. The calibration curve before calibration is a quadratic function y=Ax ² +Bx+C having a quadratic coefficient A, a linear coefficient B, and a constant term C,
The reference content is the average value of each pre-correction mass obtained by analyzing one type of standard control sample multiple times, and each pre-correction mass obtained by analyzing one type of calibration control sample multiple times. Calculate the ratio value of the content at the time of calibration, which is the average value of the mass, as the calibration coefficient α,
4. The method for calibrating an elemental analyzer according to claim 3, wherein the equation y=A(αx) ² +B(αx)+C is stored in the calibration curve storage unit as a calibration curve after calibration. The calibration curve before calibration has a slope A, and the detection mechanism outputs the mass before correction or its average value D when the analysis target element contained in the sample is not extracted by gasification at the time of creating the calibration curve. y=A(x-D),
As the uncorrected mass or its average value D1 output by the detection mechanism when the element to be analyzed contained in the sample is not gasified and extracted during the calibration,
The standard content is the average value of the values obtained by subtracting D from each uncorrected mass obtained by analyzing one type of standard control sample multiple times, and Calculate the value of the ratio of the content at the time of calibration, which is the average value of the values obtained by subtracting D1 from each obtained mass before correction, as the calibration coefficient α,
4. The method for calibrating an elemental analyzer according to claim 3, wherein the equation y=A(α(x-D1)) is stored in the calibration curve storage unit as a calibration curve after calibration. A heating furnace that gasifies and extracts the element to be analyzed contained in the sample;
a detection mechanism configured to analyze the extracted gas based on an infrared absorption method, detect the uncorrected mass of the element contained in the sample, and exhaust the gas after the analysis;
a calibration curve memory that stores a calibration curve indicating the relationship between the uncorrected mass detected by the detection mechanism and the theoretical content of the element to be analyzed contained in the standard sample when the sample is a standard sample; Department and
a correction calculation unit that converts the pre-correction mass into a post-correction mass based on the calibration curve;
A reference content in which the uncorrected mass detected by the detection mechanism when the sample is a reference time control sample at the reference time is normalized to a value per target weight value based on the actual measured weight value of the reference time control sample. a standard content calculation unit that calculates the amount;
When the sample is a calibration control sample at the time of calibration after a predetermined period has elapsed from the reference time, the pre-correction mass detected by the detection mechanism is set to the target weight value based on the actually measured weight value of the calibration control sample. a calibration content calculation unit that calculates the calibration content normalized to the permissible value;
An elemental analysis device comprising: a calibration coefficient calculation unit that calculates a calibration coefficient for calibrating the calibration curve based on the reference content and the content at the time of calibration. The elemental analysis according to claim 10, further comprising an analysis value calculation unit that calculates the mass % concentration of the element contained in the sample by dividing the corrected mass converted by the correction calculation unit by the weighed value of the sample. Device. A heating furnace that gasifies and extracts the element to be analyzed contained in the sample, and a heating furnace that analyzes the extracted gas based on infrared absorption method to detect the uncorrected mass of the element contained in the sample, and extracts the gas after analysis. a detection mechanism configured to exhaust the sample, the uncorrected mass detected by the detection mechanism when the sample is a standard sample, and the theoretical content of the element to be analyzed contained in the standard sample. A program for use in an elemental analyzer, comprising: a calibration curve storage unit that stores a calibration curve showing a relationship between hand,
A reference content in which the uncorrected mass detected by the detection mechanism when the sample is a reference time control sample at the reference time is normalized to a value per target weight value based on the actual measured weight value of the reference time control sample. a standard content calculation unit that calculates the amount;
When the sample is a calibration control sample at the time of calibration after a predetermined period has elapsed from the reference time, the pre-correction mass detected by the detection mechanism is set to the target weight value based on the actually measured weight value of the calibration control sample. a calibration content calculation unit that calculates the calibration content normalized to the permissible value;
A program for an elemental analyzer, characterized in that the program causes a computer to perform a function as a calibration coefficient calculation unit that calculates a calibration coefficient for calibrating the calibration curve based on the reference content and the content at the time of calibration. .