JPH0536748B2 - - Google Patents

Description

[Detailed description of the invention]

INDUSTRIAL APPLICATION FIELD This invention relates to a flow injection analysis method, which is a method for quantitatively analyzing a target component in an original sample by detecting optical specificity such as absorbance or fluorescence intensity. It is something. Conventional technology As a method for quantitatively analyzing target components in a sample,
The spectrophotometric method has been widely used for a long time, but in recent years, the flow injection analysis method (hereinafter referred to as "FIA") has been developed as a method to automate this spectrophotometric method. It is becoming popular in various fields such as environmental analysis. FIA adds optical specificity to the target component by adding a reagent to the flow of the original sample to cause the target component in the original sample to continuously react with the reagent, thereby developing a specific color. The target component is quantified by continuously passing the flow through a flow cell to detect absorbance, etc. In this way, FIA generally uses a visible-ultraviolet spectrometer to detect visible-ultraviolet absorbance, but conventional general FIA systems are configured to detect only one specific wavelength. There is usually one. Problems to be solved by the invention As mentioned above, the conventional FIA system is configured to detect only one specific wavelength.
Only one component can be quantified at a time. Therefore, when it is desired to quantify each component in an original sample containing multiple components, it is necessary to perform the analysis in two or more times. If it were possible to simultaneously quantify each component of an original sample containing multiple components, the number of fields in which analysis can be applied would dramatically increase, and analysis efficiency would also improve. The benefits are thought to be significant, such as the ability to simultaneously quantify multiple components even in valuable samples that are difficult to analyze. Also, as mentioned above, only one specific wavelength is detected.
The problem with the FIA system is that analysis errors tend to increase due to interference from contaminants.
In other words, in FIA, the target component is detected by adding a reagent that reacts with the target component to be quantified, giving the target component optical specificity.
In reality, components other than the target component (contaminant components) often react and affect the absorbance, and therefore, the presence of contaminants results in analytical errors. Among the above-mentioned problems, in order to solve the problem of not being able to simultaneously quantify each component in a sample containing multiple components, it is possible to analyze multiple components simultaneously using the methods described in A to D below. It is being considered that A: Method using PH gradient. B: A method that utilizes the difference in reaction rates. C: A method in which the merging zone method is used and the reaction reagents are changed and quantified separately. D: A method in which the flow path is divided into two or more and analysis is performed using separate reaction systems. However, in methods A and B, the analysis accuracy is significantly lower, and in methods C and D, the flow path system is extremely complicated, resulting in high costs, and the flow path cleaning, maintenance, and inspection are extremely complicated. Therefore, the reality is that neither method is practical. In addition, attempts have been made to solve the problem of reduced quantitative accuracy due to interference with contaminants by changing analytical conditions such as reaction reagents.
In this case, another problem occurred that made it difficult to consider and set the analysis conditions, and it was difficult to completely eliminate the influence of contaminants simply by changing the analysis conditions. This invention has been made against the background of the above circumstances, and aims to provide a flow injection analysis method that enables the simultaneous and independent determination of two or more components in an original sample containing complex components. This is the basic purpose. In addition, this invention
Another purpose is to remove the influence of impurities and improve the accuracy of quantitative analysis of target components. Means for Solving the Problems The flow injection analysis method of the present invention is
A multi-element detector is used as a photodetector to detect the optical specificity of a target component in a sample, such as absorbance or fluorescence intensity, and simultaneous detection at multiple wavelengths is performed to obtain detected values at multiple wavelengths. (spectrum)
When the sample contains multiple components as analysis target components, pre-stored standard spectral information for each of the multiple components is added to the observed spectrum of the sample. This method is characterized in that a concentration profile on the time axis separated for each component is obtained by performing a calculation for adaptation using least squares error evaluation at each time on the time axis. Function As described above, in the flow injection analysis method of the present invention, a multi-element detector is used as a photodetector to detect the optical specificity (for example, absorbance) of a sample, and detection at multiple wavelengths is simultaneously performed. Because of this, it is also possible to simultaneously obtain spectral information, i.e., changes in the absorbance of the target component in the sample depending on the wavelength. Of course, at the same time as the normal flow injection analysis method, detection is also performed at certain time intervals according to the time course, so it is possible to detect the time course of spectral information such as the absorbance of the target component in the sample. . In the case of conventional spectrophotometers, it was possible to obtain spectra such as absorbance by wavelength scanning, but scanning takes time, so the time interval between the obtained spectra is a few minutes at the shortest. This is extremely long and cannot be put to practical use at all for flow injection analysis, which determines the concentration profile of each component over time. On the other hand, in the case of the present invention, which simultaneously detects multiple wavelengths using a multi-element detector, the time interval between spectra can be significantly shortened to about 0.1 seconds at the minimum, and spectral data that is almost continuous over time can be obtained. It is possible. As mentioned above, with conventional flow injection systems, only data observed at a single wavelength (concentration profile of the target component over time at a single wavelength) can be obtained, so the degree of freedom is 1. In other words, only information on the concentration of a single component could be obtained, and if there were impurities that affected the absorbance, information that included the interference caused by the impurities could only be obtained. On the other hand, in the case of the method of the present invention, since data is obtained as a time course of a spectrum, the method theoretically has the maximum number of degrees of freedom that allow the spectrum to be divided by wavelength. In this invention, by applying a data processing method that effectively utilizes this feature, it is possible to quantify multiple components simultaneously, and to quantify only the target component by removing the influence of interference caused by contaminants. It is said that That is,
The spectrum observed at each time can be regarded as the sum of the individual spectra of multiple components (two or more target components, or target components and impurity components) contained in the sample (i.e., a composite spectrum). Therefore, the observed spectrum can be separated into the spectra of each component by performing an operation that fits the standard spectral information of the target component stored in advance to the observed spectrum using the method of least squares. Therefore, by drawing and quantifying the time profile of the concentration of each component from the spectrum separated into each component at each time, it is possible to simultaneously quantify multiple components and remove the effects of contaminants. DETAILED DESCRIPTION OF THE INVENTION FIG. 1 shows an example of a flow path system of a flow injection system for carrying out the flow injection analysis method of the present invention. This flow path system will be described with reference to an example in which the target component in the sample S is once reduced by the reducing agent A and then reacted with the coloring agent R to cause the target component to develop color. In FIG. 1, water 1 as a carrier is sent by a liquid pump P1 through a damper 2 to a sample injector 3 as a sample injection port, and a sample S is injected and mixed in this sample injector 3. The sample liquid into which S is injected and mixed is sent to the reduction reaction coil 5. On the other hand, the reducing agent A is also sent to the reduction reaction coil 5 via the damper 6 by the liquid sending pump P2. In this reduction reaction coil 5, the sample liquid containing the sample S and the reducing agent A are mixed, and the reduction reaction of the sample S proceeds. The sample liquid after the reduction reaction is sent to the reaction coil 7. On the other hand, the coloring agent R is passed through the damper 8 by the liquid feeding pump P3 to the reaction coil 7.
sent to. In this reaction coil 7, the sample liquid containing the reduced sample and the coloring agent R are mixed, and a coloring reaction of the sample proceeds. The sample liquid in which this coloring reaction has substantially completed is transferred to the reaction coil 7.
from there to the flow cell 9. This flow cell 9
As will be explained in detail later, a multi-wavelength simultaneous detection device D consisting of a multi-element detector (multi-wavelength visible and ultraviolet detector) simultaneously detects the absorbance of the sample liquid at multiple wavelengths, and its time course is also measured. Detected. The output of this photodetector D is sent to a data processing section 10 such as a computer, and the data processing result by the data processing section 10 is displayed on a display section 11 such as an X-Y blotter. FIG. 2 shows an example of a double beam type multi-wavelength simultaneous detection device D used in the device shown in FIG. Light from a multi-wavelength light source 12 such as a white light source is split into two beams by a beam splitter 13, and each beam is divided into a reference chamber 14 and a sample chamber 15.
pass through. The aforementioned flow cell (not shown) is provided in the sample chamber 15, and the liquid containing the reacted sample passes through this flow cell. On the other hand, the reference chamber 14 is provided with an empty cell or a cell containing a standard substance. The beams that have passed through the reference chamber 14 and the sample chamber 15, that is, the reference beam and the sample beam, are alternately chopped by a chopper 16, and then split into multiple channels of different wavelengths (for example, 32
Further, the sample light and reference light of each wavelength are inputted through a beam mixer 18 to a multi-element detector 19 consisting of, for example, a 32-channel photodiode array. For example, 32 unit detection elements 2 constitute this multi-element detector 19.
0, light of each wavelength (each channel) that is split into multiple wavelengths is input alternately with the reference light and sample light, and a signal corresponding to the sample light intensity normalized by the output of the reference light is output. be done. That is, the multi-element detector 19 simultaneously outputs 32 channels of signals corresponding to the sample light intensity in each channel (each wavelength) of which the observation wavelength region is divided into 32 channels. 3A, B, and C show the detection output from the multi-element detector 19 in a three-dimensional film, that is, the wavelength axis (channel axis), the time axis, and the detected value axis (absorbance axis, light intensity axis, etc.). It is stored as three-dimensional information, and it is also stored in advance with standard spectral information obtained through experiments for the components expected to be included in the sample (components for analysis), and the observed sample The three-dimensional information about the sample and pre-stored standard spectrum information corresponding to the known components contained in the sample are processed to separate the observed data into concentration time profiles for each target component. A data processing unit 10 consisting of a computer and the like is shown. In particular, FIG. 3A shows a circuit section 10A for reading out 32 channels of signals detected by a 32 channel photodiode array as a multi-element detector 19, and FIG. A processing control section 10B is shown for digitizing the obtained values and processing them as three-dimensional information corresponding to a time axis and a wavelength channel axis and for performing various controls,
Furthermore, FIG. 3C shows information about the sample that has been observed and stored by reading out the stored information from the information storage section (floppy disk section) in which standard spectral waveform information of various components prepared in advance is stored. Data station section 1 that calculates three-dimensional information
Indicates 0C. In the data processing unit 10 shown in FIGS. 3A to 3C, the detected value (in this case, sample light intensity) in each channel of the simultaneously observed wavelength region divided into 32 channels is output from the multi-element detector 19. The signal is sent to the main amplifier 23 via the preamplifier 22 corresponding to each channel, is logarithmically amplified by the logarithmic amplifier 23A in the main amplifier 23, and is synchronously detected by the synchronization switch 23B. Note that the synchronization switch 23B is controlled by a synchronization signal from a synchronization signal generation circuit 23C. The outputs of the main amplifiers 23 of each channel synchronously detected in this way are sequentially read out by an analog switch 24 consisting of a multiplexer, passed through a range switching circuit 25, and are digitized by an A/D converter 26 in FIG. 3B. and is read into the main computer 28 via the interface 27. Note that the analog switch 24 is connected to the main computer 2.
Switching is controlled by a control signal from 8. The main computer 28 converts the detected signal (logarithmically amplified value) into a signal corresponding to the time axis and wavelength axis (channel), that is, three-dimensional information, and the three-dimensional information is transferred to the data/command bidirectional transfer interface 29, 29' to the external computer section, ie, data station section 10C shown in FIG. 3C. As mentioned above, the detected value is usually detected as the light intensity of the sample and logarithmically amplified, but in normal analysis it is often used as its reciprocal, that is, the logarithmic absorbance, and in that case, it is converted to the reciprocal as appropriate. In the following explanation, the detected value will be expressed as absorbance. In addition, in FIG. 3B, 30 is a multi-channel DAC output interface, which outputs raw data that is not subjected to calculations in cases such as when an external computer section (data station section) is not connected.
31 is a front panel on which command displays and instruction switches are provided, and 32 is for temporarily transmitting at least a part of the output signal from the multi-element detector 19. This is a memory for storing information and reading it out as needed. On the other hand, the data station section (external computer section) 10C in FIG.
9', a computer 34, a floppy disk unit 35, a CRT 36, a printer 37, and an output interface 38. The output interface 38 is connected to the XY plotter 11. The floppy disk section 35 constitutes an information storage section for storing standard spectral waveform information of various target components expected to be included in the sample. also stores the waveforms and the like of each component from which the observed three-dimensional information is separated. Then, the computer 34 processes the standard spectrum waveform information read from the floppy disk unit 35 and the three-dimensional information about the observed sample using the least squares error evaluation method, and calculates the concentration separated for each target component. A time profile is derived and the separated concentration-time profile is displayed on the XY plotter 11. Next, a method of the present invention for quantitative analysis using the above-mentioned apparatus will be explained. FIG. 4 shows an example of three-dimensional information detected by the multi-element detector 19 for a sample containing a plurality of components as analysis target components. Here, the X-axis in the horizontal plane represents time t, the Y-axis represents wavelength λ, and the Z-axis in the height direction represents absorbance. In this example, the peaks of the time profile (i.e., the waveform in the plane formed by the time axis and the absorbance axis) for concentration (actually absorbance or logarithmic absorbance) are not separated for each component, and the combined concentration of multiple components - It is a time profile. Of course, the spectral waveform (waveform in the plane formed by the wavelength axis and the absorbance axis) is also a composite spectral waveform of a plurality of components. FIG. 5 is a diagram schematically showing the relationship between the spectral waveform, concentration-time profile, and the three-dimensional information in an easy-to-understand manner. Note that this figure illustrates a case where the sample contains two components as analysis target components, and the explanation will be made below with reference to this figure. Here, it is assumed that there is no interference between the respective components of the multi-component spectrum, and that linearity exists between the amount of each component (composition ratio) and the spectrum intensity information (including the absorbance of the logarithmically converted value). First, the multiple components contained in the observed sample are
Although the type itself is known, if the peaks of the concentration-time profile of the observed sample composed of those known components (therefore, the combined concentration-time profile of multiple known components) overlap, it is difficult to separate them. An arithmetic processing method according to the present invention for determining the concentration-time profile for each component will be explained. In this case, the standard spectral waveform (pre-memorized) of each component contained in the observed sample is used.
is read out, the standard spectrum of each component is multiplied by each component concentration (unknown amount), and the composite value is observed.The least squares method is applied to the composite spectrum (wavelength axis - absorbance axis information of three-dimensional information) of the sample. By fitting, we find the component ratio of each component at a certain observation time, and by performing the same calculation at each time on the observation time axis, we find the separated concentration-time profile of each sample component. . In other words, if the observed sample contains m known components and the observed wavelength region is divided into n channels (32 channels in the above example) at equal intervals, the horizontal axis of the spectrum at a certain time t ( The vertical axis according to the j-th standard among the m standard spectra (previously stored standard spectra corresponding to each of the m components) at the i-th equidistant point out of the n equidistant dots on the wavelength axis) Axis value (absorbance)
Rij. Let the component quantity Xj of the j-th component be an unknown quantity, and multiply the value Xj by the vertical axis value Rij of the corresponding standard spectrum to calculate the i-th equinox point of the channel on the wavelength axis of the observed composite spectrum of the sample. It is sufficient to perform an operation to match the vertical axis value (absorbance) Si. To do this with least squares error estimation,
The value of Q in the following equation (1) may be minimized. Q= _o 〓 ⁱ⁼¹ [Si− _n ^〓 j= ¹ The normal equation obtained by finding αQ/αXn and setting it to 0 is the following equation (2).

【table】

[Table] 〓 ⁱ⁼¹ 〓
Then, equation (2) can be expressed as AY=B...(3). Then, the solution to equation (2) can be obtained in the form of Y=(A ^T A) ^-1 A ^T B (4) if expressed in a matrix using A, Y, and B. In order to obtain this solution, there are methods such as solving so-called simultaneous equations without using matrix operations, and directly performing matrix operations in equation (4). As can be seen from equation (4), (A ^T A) ^-1 is an m×m symmetric matrix, so Choleski's method or the like is used to calculate equation (4). In any case, the regular formula is m element 1
Since these are the following simultaneous equations, their solutions can be obtained using a conventionally known program. In this way, each component quantity Xj (j=1 to m) of m components at a certain time is determined. By performing such calculations one after another over the entire range of the observation time axis, the amount of each component at each time can be found over the entire range of the observation time axis. Therefore, the temporal change in the amount of each sample component at a certain wavelength, that is, the temporal profile of the concentration, is separated and determined. The above explanation is for the simultaneous quantification of two or more components in a sample containing two or more components. In the case of doing so, the same calculation process can be performed. That is, by processing the number m of the constituent components as 1, it is possible to obtain an accurate time-concentration profile of only the target component with the interfering influence of impurities removed. Example The flow injection analysis method of this invention is
An example in which a mixed sample containing iron ions (valence) and copper ions (valence) is simultaneously quantified will be described below. As a channel system, a device as shown in FIG. 1 was used, and as an optical system, a device as shown in FIG. 2 was used. As the multi-element detector 19, a photodiode array with 32 elements for visible and ultraviolet detection was used. Here, each element of the multi-element detector 19 outputs data with a wavelength width of 10 nm per element, so that the multi-element detector as a whole covers a range from 380 nm to 700 nm. The flow cell has an inner diameter of 1 mm and an optical path length of 5 mm.
I used the one from Water was used as a carrier liquid, and L-ascorbic acid was used as a reducing agent. This is to reduce the coexisting iron valence ions and copper valence ions, respectively, into iron valence ions and copper valence ions. On the other hand, color former (chelating reagent)
2,4,6-tris(2-pyridyl)-s-, which has specificity with the target metal and has a different absorption spectrum of the complex formed.
Furiazin (abbreviated as TPZ: a chelating reagent for iron ions) and Bathocuproinedisulfonic acid sodium salt (abbreviated as Bath: a chelating reagent for copper ions) were used. Under the above conditions, a mixed sample of iron ions () and copper ions () was prepared, and the concentrations of both iron ions and copper ions were 6 × 10 ^-5 mol, 8 × 10 ^-5 mol, and 15 × 10 ^-5 moles, 20×10 ^-5 moles in sequence
Figure 6 shows a three-dimensional FIA chart obtained by injecting the sample into the sample injector every minute and analyzing it. That is, the chart in FIG. 6 can be rephrased as information showing the time course of the absorbance spectra of samples containing various concentrations of iron complexes and copper complexes. Perform calculation processing to fit the standard spectrum of the iron complex and the standard spectrum of the copper complex using the method of least squares to the spectrum at a certain time in this chart, and calculate the concentration of the iron complex and the copper complex at that time. Figure 7 shows the results obtained over all times. That is, FIG. 7 is obtained by independently separating the iron complex concentration time profile and the copper complex concentration profile from the combined data of FIG. 5. The results of plotting each peak value of the iron complex concentration profile and each peak value of the copper complex concentration profile shown in FIG. 7, obtained by the method of the present invention as described above, against the concentration at that time. is indicated by a circle in FIG. 8A and FIG. 8B. On the other hand, Figures 8A and 8 show the results of floating each peak value of each concentration profile with respect to the concentration when complexes of iron ions and copper ions were individually FIA analyzed according to the conventional method. △ in B
Indicated by a mark. From FIG. 8A and FIG. 8B, the results of simultaneous determination of each component according to the method of the present invention are in extremely good agreement with the results of individual determination of each component using the conventional method. It is clear that the method of the invention has actually made it possible to simultaneously quantify each component in a mixed sample containing the components. In addition, in the above example, if iron ions are considered as the target component for analysis and copper ions are considered as impurity components, accurate quantification of iron ions, which are the target components, can be achieved by removing the influence of copper ions as impurity components. It can be considered that this has become possible. That is, from the above results, it can be seen that the influence of impurity components has been removed and accurate quantification of only the target component has become possible. Effects of the Invention As is clear from the above explanation, according to the flow injection analysis method of the present invention, it is possible to simultaneously quantify two or more components even in a sample containing two or more components. With the method of the present invention, there is no risk of problems such as a significant decrease in analysis accuracy or a significant increase in complexity of the flow path system. As a result of being able to quantify multiple components simultaneously with high precision, the efficiency of quantitative analysis has been significantly improved compared to conventional methods, and it has become possible to analyze complex components even in minute amounts of samples, making it possible to analyze complex components using flow injection technology. The field of application of analysis can be greatly expanded compared to the past. Furthermore, according to the flow injection analysis method of the present invention, even when quantifying only a single target component, it is possible to remove the influence of impurities and accurately draw the concentration-time profile of only the target component. Therefore, quantitative accuracy can be significantly improved compared to conventional methods, and in this case, there is no need to set special analysis conditions for reaction reagents, etc., and high-precision analysis can be performed easily. can.

[Brief explanation of drawings]

FIG. 1 is a schematic block diagram showing an example of a flow path system of a flow injection analysis system to which the method of the present invention is applied, and FIG. A schematic diagram showing an example of an optical system of a simultaneous wavelength detection device; FIGS. 3A, B, and C are block diagrams showing an example of a data processing section used to implement the method of the present invention;
FIG. 4 is a waveform diagram showing an example of three-dimensional information (three-dimensional FIA chart) obtained from a multi-element detector using the method of the present invention, and FIG. FIG. 6 is a waveform diagram showing a three-dimensional FIA chart according to an embodiment of this invention, and FIG. 7 is a waveform diagram showing a three-dimensional FIA chart separated from the three-dimensional chart in FIG. FIG. 8 is a waveform diagram showing the concentration-time profile of each target component. In this example, each peak value of the concentration-time profile in FIG. Figure 8A is a diagram showing a comparison with the peak value.
is a graph shown for an iron complex, and FIG. 8B is a graph shown for a copper complex. 3... Sample injector, 5... Reduction reaction coil, 7... Reaction coil, 9... Flow cell, D... Multi-wavelength simultaneous detection device, 10... Data processing section, 11... Display section (X- Y Protester),
12... Light source, 15... Sample chamber, 17... Spectroscopic means, 19... Multi-element detector, 20... Unit detection element.

Claims (1)

[Claims] 1. A multi-element detector is used as a photodetector to detect the optical specificity of a target component in a sample, and simultaneous detection is performed at multiple wavelengths to obtain detected values at the multiple wavelengths. When the sample contains multiple components as analysis target components, pre-stored standard spectral information for each of the multiple components is added to the observed spectrum of the sample using a least squares error. A flow injection analysis method is characterized in that a concentration profile on the time axis separated for each component is obtained by performing a calculation that fits the method at each time on the time axis.