JPH1194735A - Quantitative analyzer for sample characteristic using spectrophotometry and multivariate analysis and analysis method using it - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a quantitative analyzer for sample characteristic which is modified to simplify calibration work and an analysis method using the analyzer. SOLUTION: A reference instrument 1 for calibration and each predictive quantitative instrument are separately formed and a wavelength shifting means 4 is installed to the spectral optical path 3 of the reference instrument 1 so as to add a waveform shifting condition to spectrophotometry at the time of calibration. The spectral optical path 3 of each predictive quantitative instrument is not provided with the waveform shifting means 4, but designed equally to the reference instrument 1, so that the calibrated matrix obtained by means of the reference instrument 1 can be used as that of each predictive quantitative instrument. Therefore, the original quantitative value difference between each predictive quantitative instrument can be compensated by making the fine wavelength error correspond to a prescribed characteristic item as the instrumental error of each predictive quantitative instrument.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for quantitatively analyzing sample characteristics using spectrophotometry and multivariate analysis, and an analytical method using the apparatus.

[0002]

2. Description of the Related Art A conventional quantitative analyzer of sample characteristics using a spectrophotometric measurement and a multivariate analysis method includes a calibrator for obtaining a spectral response matrix with a known characteristic amount for a given measurement condition, and a spectrometer for the same. The quantitative device that predicts the characteristic amount using the calibration matrix obtained by analyzing the response matrix was configured in the same body (see FIG. 8).

[0003]

In the conventional quantitative analyzer as described above, for example, when the characteristic quantity to be predicted is the concentration of a plurality of components in the solution, the spectral absorbance vector of the solution sample whose concentration is unknown is determined. Because the spectrophotometer (of the predictive quantification device) and the spectrophotometer (of the calibration machine) for obtaining the spectral absorbance response matrix whose concentration is known are the same body, I had to calibrate. Therefore, a great deal of labor is required in the calibration process, which reduces the productivity of the apparatus and increases the cost.

[0004] The present invention has been made in view of such circumstances,
It is an object of the present invention to provide an apparatus for quantitatively analyzing sample characteristics and an analysis method using the apparatus for quantitatively analyzing a sample, which facilitates calibration work.

[0005]

According to the present invention, means for solving the above-mentioned problems are constituted as follows. That is, in the invention according to the first aspect, in a quantitative analysis apparatus for sample characteristics using spectrophotometry and multivariate analysis, a calibration reference machine for performing spectrophotometry with a wavelength shift condition is added to each predictive quantitative It is formed separately from the apparatus, and the instrumental differences of each of the predictive quantification devices are configured to be aggregated under the wavelength shift condition of the spectrophotometric measurement at the time of calibration to perform the compensation. I have.

According to a second aspect of the present invention, in the apparatus for quantitatively analyzing the characteristics of a sample using spectrophotometry and multivariate analysis, a calibration reference machine and each predictive quantitative apparatus are formed separately. And a wavelength shift means is provided in the spectroscope of the calibration reference device, and a wavelength shift condition is added to the spectrophotometric measurement at the time of calibration, while the spectroscope of each of the predictive quantification devices,
Without providing the wavelength shift means, the optical system and the optical system of the calibration reference device and the same design, the calibration matrix obtained by the calibration reference device, by the calibration matrix of each of the predictive quantitative device, It is characterized in that a minute wavelength error as a machine difference of each prediction and quantification device is made to correspond to a predetermined characteristic item, and machine difference of an original quantitative value is compensated.

According to a third aspect of the present invention, in each of the predicting and quantifying devices according to the first or second aspect of the present invention, a wavelength shift means for adding the wavelength shift condition is set in a spectral optical path. It is characterized by being provided so as to be retractable or detachable.

According to a fourth aspect of the invention, the characteristic of the sample to be analyzed in any one of the first to third aspects of the invention is a concentration of each component of the multi-component aqueous solution. .

According to a fifth aspect of the present invention, the spectrophotometric measurement in any one of the first to third aspects uses spectral absorbance data ranging from a near ultraviolet region to a near infrared region. I have.

According to a sixth aspect of the present invention, the wavelength shift means provided in the spectroscope of the calibration reference device according to any one of the first to fifth aspects of the present invention is arranged around an axis parallel to the diffraction grating axis. It is made of a parallel flat plate that is rotatable and has optical characteristics that are substantially transparent within the spectral range, and is arranged in the optical path at a position near the rear of the entrance slit or near the front of the exit slit.

According to a seventh aspect of the present invention, there is provided an analysis method in which calibration can be performed unitarily using a calibration reference machine formed separately from each of the predictive quantification devices to be calibrated. In addition, the spectrum data under the condition that a certain wavelength shift is added considering the range of the machine difference of each predictive quantification device is also acquired, and the reference calibration matrix is obtained from the resulting spectral response matrix. After storing the calibration matrix as a calibration matrix common to each of the prediction and quantification devices, each of the prediction and quantification devices measures an unknown characteristic quantity, and the resulting spectral absorbance vector is stored. By performing a matrix operation with the calibration matrix, the concentration of each component and the aggregated wavelength error of the aircraft used are obtained, and as a result, the concentration value of each component that compensates for the instrumental error aggregated into the minute wavelength error is obtained. This It is characterized in.

A calibration matrix obtained by performing a spectrophotometric measurement to which a wavelength shift condition is added by a calibration reference machine is used as a calibration matrix for prediction in each predictive quantitative device, thereby obtaining a difference between the individual predictive quantitative devices. Can be made to correspond to another characteristic item (predetermined characteristic item), and instrumental difference compensation of quantitative values corresponding to other characteristic items can be performed. In other words, since the calibration reference devices are configured separately and independently, the calibration matrix can be unified, and the calibration process for each prediction and quantification device becomes unnecessary, and the calibration work is greatly simplified.

[0013]

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of a quantitative analysis apparatus for sample characteristics using spectrophotometry and multivariate analysis and an analysis method using the quantitative analysis apparatus according to the present invention will be described in detail. The basic concept of this quantitative analyzer is that, as shown in FIG. 1, the calibration reference machine 1 is configured separately and separately from each of the predictive quantification apparatuses 2,. The calibration matrices are used as calibration matrices of the predictive quantification devices 2... To unify the calibration matrices. By performing the spectrophotometric measurement at the time of calibration with the wavelength shift condition added (see FIGS. 2 and 3), the individual machine differences of the respective predictive quantification devices 2 are aggregated under the wavelength shift condition, The main feature is that the machine difference compensation of all the predictive quantitative devices 2,... Can be unitarily performed by a single calibration reference machine 1.

More specifically, as shown in FIG. 2, the configuration of the calibration reference device 1 includes a light source 11, a lens 13, an entrance slit 15, a concave spherical mirror 20, a plane diffraction grating 25, a concave spherical mirror 30, and an exit slit 45. Czerny-Turner spectrometer 75 comprising the above-described entrance slit 15 and concave spherical mirror 20
Wavelength shift means 4, switching mirror 50, liquid cell 55, cell compensator 56, condensing mirror 6
2, 63, a detector 70, an amplification and AD converter 82, a storage operation means 85, a display 87 and the like.

The above-mentioned wavelength shifting means 4 is capable of rotating a parallel plane plate 41 having substantially transparent optical characteristics in the spectral region around an axis 42 parallel to the axis 27 (diffraction grating axis) of the plane diffraction grating 25. Thus, a desired wavelength shift condition can be added at the time of spectrophotometric measurement in the calibration process.

On the other hand, each of the predictive quantitative devices 2 to be calibrated,
.. Do not include only the wavelength shifting means 4 described above, the other optical systems have the same design as the optical system of the calibration reference machine 1, and are not shown. Note that the wavelength shifting means 4
May be provided in the spectroscopic optical path of the predictive quantitative device 2 so as to be able to be set / retracted or detached.

With the above-described configuration, the white light emitted from the light source 11 is condensed and passed through the entrance slit 15, becomes a parallel light beam at the spherical mirror 20, and enters the diffraction grating 25, and the diffracted light 28 is reflected by the spherical mirror 30. , Its reflected light 32
Forms an image on the exit slit 45. At this time, the diffraction grating 2
5 rotates 29 about an axis 27 parallel to the grating groove, and the wavelength of light passing through the exit slit 45 is scanned.

For example, if the sample to be quantitatively analyzed is a multi-component aqueous solution and the concentration of each component is a characteristic amount, the spectral shift obtained by adding the wavelength shift amount by the wavelength shifting means 4 in addition to the conventional characteristic amounts (see FIG. 8). Measurement conditions (see FIG. 1) for performing photometric measurement are set, a spectral response matrix composed of a group of spectral absorbance data is obtained, and a calibration matrix is obtained using known values of characteristic amounts (each component concentration and wavelength shift amount). Can be calculated.

At this time, a standard aqueous solution having a known characteristic amount is introduced into the liquid cell 55 of the calibration reference machine 1, and the transmittance of the sample is obtained by alternately comparing the transmission with the cell compensator 56 in the reference optical path. Is converted into an absorbance by the storage operation means 85.

On the other hand, in the predictive quantitative device 2, the calibration reference machine 1
To store the calibration matrix obtained in the above, measure and obtain the spectral (absorbance) vector whose characteristic quantity is unknown for each analysis, and perform another matrix operation in the operation storage means (CPU) 85 different from the calibration. Component concentration consisting of multiple characteristic quantities
The wavelength machine difference vector is calculated, and the characteristic amount display device 87 displays each component concentration.

Next, considering the principle of spectrophotometric measurement, the horizontal axis of the spectral data to be used is an extraction wavelength or an extraction point number, and the vertical axis is a spectral photometric amount or transmittance or a transmittance or a transmittance corresponding to each point on the horizontal axis. Absorbance. The original data processed by the multivariate analysis method is a group of absorbance values corresponding to the extraction point numbers, and does not include wavelength information.

The instrumental differences in a spectrophotometer include vertical axis errors caused by light source intensity, detector sensitivity, linearity of an amplifier, and deviations from a true wavelength caused by mechanical accuracy of a spectroscopic optical system. A horizontal axis error, which is a minute wavelength error, can be considered.

The former vertical axis error can be relatively easily corrected by two-beam spectroscopy using a sample beam and a reference beam, and a digital storage operation means, and is practically used. On the other hand, the latter horizontal axis error cannot be completely removed, but measures have been taken to keep it within an allowable range corresponding to the desired quantitative accuracy. However, there has been a limit to the improvement of the quantitative accuracy, and the improvement of the wavelength accuracy has always been accompanied by high costs.

Therefore, in the conventional spectrophotometric measurement using the multivariate analysis method, as shown in FIG. 8, calibration and prediction are performed by the same spectrophotometer, and the spectrophotometer is used in a state where the instrumental difference does not influence. In these devices, absolute wavelength accuracy is unnecessary, and if only wavelength reproducibility is secured, considerable quantitative accuracy is also guaranteed.

Therefore, the extracted wavelength points of each body were slightly different, and the values of the elements of the obtained calibration matrix were completely different. Therefore, it is impossible to use a single calibration matrix for each machine, and the calibration work performed for each machine is complicated, requires equipment and time, and productivity is not high.

The present invention has been proposed in order to solve such difficulties. In the multivariate analysis method, a factor (principal component) having a large variate is selected from the multivariate data.
Principle Component Analysis (Principle Component Analysis, P
CA) and Partial Least Square
s, PLS) can also be applied. For the spectrophotometric measurement of the multi-component aqueous solution, spectral absorbance data from the near ultraviolet region to the near infrared region can be used.

The analysis method of the present invention will be explained in order and easily (see the flowchart of FIG. 4). The wavelength error of each airframe is regarded as another (predetermined) characteristic item which causes a change in the absorbance value. In the calibration work using the machine 1, a condition in which a certain wavelength shift is added in consideration of the variation range of the machine difference between the individual predictive quantification devices together with the known combination conditions of other characteristic amounts (including no wavelength shift). )) To obtain and acquire the spectral data under (S1).

As a result, a reference calibration matrix is calculated from the obtained spectral response matrix (S2).

The calibration matrix obtained by the calibration reference machine 1 is
The storage means of each predictive quantification device is stored as a common calibration matrix (S3).

Each predicting and quantifying device calculates unknown characteristic quantities,
For example, the type of the component is known, but the concentration of each component is unknown, and the concentration of each component is determined by matrix operation of the resulting spectral absorbance vector and the stored common calibration matrix. And each characteristic quantity including the instrumental difference collected into the wavelength error quantity of the used machine (S4).

As a result, it is possible to obtain and display the concentration value of each component with high quantitative accuracy by compensating for the machine difference gathered in the minute wavelength error (S5).

As described above, according to the analysis method of the present invention, it is possible to unify the calibration matrix, which has been impossible in the past, and it is not necessary to perform the calibration process for each predictive quantitative device.
For this reason, the calibration work in the production stage is facilitated, the productivity is greatly improved, and re-calibration is required on the user side by including minute wavelength shifts due to aging in the machine difference compensation in S5. Without compensation, it is possible to automatically compensate and maintain and improve the reliability.

Further, the degree of the wavelength error of each predictive quantification device 2 can be statistically grasped, which can contribute to the improvement of production technology. Further, by comparing the amount of the wavelength error, which is the machine difference obtained for each prediction, with the initial value, it becomes possible to determine the reliability of each measurement and the service time such as readjustment.

FIG. 5 shows the configuration of the calibration reference device 1 in a different embodiment of the quantitative analyzer. In this case, the condenser mirrors 5 and 6, the liquid cell 7, the cell compensator 8, and the switching mirror 9 are provided on the light source 11 side. While the multi-detector array 4 is used instead of the exit slit.
0 is provided on the output side, the diffraction grating 25 is fixed at a fixed angle, and forms an image dispersedly on the short-wavelength light 31, the long-wavelength light 33 and the light receiving surface 42. Although the spectral scanning is performed electronically, the operation of the wavelength shift means 4 and other operations are the same as those in FIG. The switching mirror 9 swings around an axis 91 parallel to the axis 42 of the wavelength shift means 4. On the other hand, it is assumed that each predictive quantification device 2 excludes the above-described wavelength shift means 4 or that the wavelength shift means 4 is provided in the spectral optical path 3 so as to be settable / retractable or detachable.

FIG. 6 shows that the wavelength shifting means 4 is
5 shows a case where it is provided in the vicinity of the front, and the other configuration is the same as that of FIG. FIG. 7 shows a multi-detector array 40 attached to a sliding means 44 for sliding in the direction of a straight line 49 instead of the wavelength shifting means 4 so that the same wavelength shifting condition can be added. Other configurations are the same as those in FIG.

[0036]

As described above, according to the present invention,
A calibration reference machine for performing the spectrophotometric measurement with the wavelength shift condition added is formed separately from each of the predictive and quantitative devices, and the individual machine difference of each of the predictive and quantitative devices is referred to as the wavelength shift condition of the spectrophotometric measurement at the time of calibration. Since it is summarized below and the compensation is performed, the calibration process for each predictive quantification device becomes unnecessary, the minute wavelength error is compensated by each machine difference, the measurement accuracy and operability and reliability are improved, This makes it possible to shorten the product production period, simplify the production equipment, and reduce the cost. Further, the distribution of the minute wavelength error for each machine in the production process can be statistically analyzed and reflected on the improvement of production technology.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a basic configuration of an embodiment of an apparatus for quantitatively analyzing sample characteristics using spectrophotometry and multivariate analysis according to the present invention.

FIG. 2 is a configuration diagram of the calibration reference device.

FIG. 3 is a plan view of the wavelength shift means.

FIG. 4 is a flowchart showing a procedure of the calibration process.

FIG. 5 is a configuration diagram of a calibration reference machine in different embodiments of the identification amount analyzer.

FIG. 6 is a main part configuration diagram of another calibration reference machine;

FIG. 7 is a main part configuration diagram of still another calibration reference machine.

FIG. 8 is a block diagram showing a basic configuration of a conventional quantitative analyzer.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Calibration reference machine, 2 ... Predictive quantification device, 3 ... Spectral optical path, 4
... wavelength shift means, 15 ... entrance slit, 27 ... diffraction grating axis, 45 ... exit slit.

Claims (7)

[Claims] An apparatus for quantitatively analyzing sample characteristics using spectrophotometry and multivariate analysis, wherein a calibration reference machine for performing spectrophotometry with a wavelength shift condition is provided separately from each predictive quantitative apparatus. The spectrophotometric measurement is characterized in that the instrumental differences between each of the predictive and quantitative measuring devices are formed under the above-mentioned wavelength shift condition of the spectrophotometric measurement at the time of calibration and the compensation is performed. Quantitative analyzer for sample characteristics using a variable analysis method. 2. An apparatus for quantitatively analyzing sample characteristics using spectrophotometry and multivariate analysis, wherein a calibration reference machine and each predictive quantitative apparatus are formed separately, and wherein said calibration reference machine is used. In the spectrometer, a wavelength shift means is provided, and a wavelength shift is added to the spectrophotometric measurement at the time of calibration.On the other hand, the spectroscope of each of the predictive quantification devices, the optical system is provided without the wavelength shift means. The same design as the optical system of the calibration reference device, by using the calibration matrix obtained by the calibration reference device as the calibration matrix of each predictive quantitative device, the minute wavelength error as a machine difference of each predictive quantitative device The apparatus for quantitatively analyzing sample characteristics using spectrophotometry and multivariate analysis, wherein the apparatus is adapted to correspond to a predetermined characteristic item and compensates for the mechanical difference of the original quantitative value. 3. The apparatus according to claim 1, wherein each of said prediction and quantification devices is provided with a wavelength shift means for adding said wavelength shift condition to said spectrometer so as to be settable, retractable or detachable. An apparatus for quantitative analysis of sample characteristics using the spectrophotometric measurement and the multivariate analysis method according to item 2. 4. The spectrophotometric measurement and multivariate analysis method according to claim 1, wherein the characteristic of the sample to be analyzed is the concentration of each component of the multicomponent aqueous solution. Quantitative analyzer for sample characteristics using 5. The spectrophotometric measurement and multivariate analysis method according to claim 1, wherein said spectrophotometric measurement uses spectral absorbance data ranging from a near ultraviolet region to a near infrared region. Quantitative analyzer for sample characteristics using 6. A wavelength shift means provided on a spectroscope of the calibration reference device is a parallel flat plate rotatable around an axis parallel to a diffraction grating axis and having substantially transparent optical characteristics in a spectral region, The spectrophotometric measurement and the multivariate analysis method according to any one of claims 1 to 5, wherein the spectrophotometric measurement and the multivariate analysis method are arranged in an optical path at a position near the rear of the entrance slit or at a position near the front of the exit slit. Quantitative analyzer for sample characteristics. 7. An analysis method in which calibration can be performed centrally using a calibration reference machine formed separately from each of the prediction and quantification devices to be calibrated. The spectral data under the condition of adding a certain wavelength shift in consideration of the variation range of the machine difference is also acquired, and a reference calibration matrix is obtained from the resulting spectral response matrix. After storing as a common calibration matrix in the predictive quantitative device, by each predictive quantitative device,
By measuring the unknown characteristic amount, and by performing a matrix operation of the resulting spectral absorbance vector and the stored calibration matrix, the concentration of each component and the aggregated wavelength error of the aircraft used are determined, As a result, a method of analyzing the characteristics of a sample using a spectrophotometric measurement and a multivariate analysis method using a spectrophotometric measurement and a multivariate analysis method, wherein a concentration value of each component compensating for an instrumental error aggregated into a minute wavelength error is obtained. .