The present invention relates to a spectrophotometer that employs infrared spectroscopy such as, for example, Fourier-transform infrared spectroscopy, and to a spectrophotometry method.
[Technical Background]
[0002]
Conventionally, as is shown in Patent Document 1, an FTIR (Fourier-transform infrared spectroscopy) method is used in order to measure concentrations of measurement target components contained in a test sample such as, for example, exhaust gas or the like. [0003]
A spectrophotometer that uses this FTIR method is provided with a measurement cell into which a test sample is introduced, a light irradiation unit that irradiates infrared light onto the measurement cell, and a photodetector that detects an intensity of light transmitted through the measurement cell. This spectrophotometer calculates an optical absorption spectrum of exhaust gas using light intensity signals obtained by the photodetector, and calculates the concentration of the measurement target component from the absorbance of this optical absorption spectrum. Here, when calculating a concentration from the absorbance of the optical absorption spectrum, a calibration curve comparing the absorbance of the optical absorption spectrum with the concentration of the measurement target component shown by this absorbance is used. [0004]
In the above-described spectrophotometer, during the measurement of a test sample, the temperature of the measurement cell is adjusted so that it remains constant. If the temperature of the measurement cell differs from the temperature of the measurement cell at the time when the calibration curve was created, then discrepancies occur in the concentrations obtained from the calibration curve. [0005]
As is shown in Patent Document 2, a spectrophotometer that corrects the calibration curve at each temperature has been developed in order to reduce discrepancies caused by these temperature changes. More specifically, using the same sample, this spectrophotometer measures a spectrum in advance at a reference temperature and at a different temperature from the reference temperature, and thereby determines a difference

spectrum thereof. By then modifying the difference spectrum that has been multiplied by a coefficient in accordance with the temperature changes in the sample such that it is set to the measurement spectrum, the difference spectrum is converted into an equivalent spectrum to the spectrum measured at the reference temperature. The spectrophotometer then corrects the calibration curve calculation results obtained from temperature changes in the sample.
[Documents of the prior art] [Patent documents]
H9-101257 2005-331386
[0006]
[Patent document 1] Japanese Unexamined Patent Application (JP-A) No
[Patent document 2] Japanese Unexamined Patent Application (JP-A) No
[Disclosure of the Invention]
[Problems to be Solved by the Invention]
[0007]
The present invention was therefore conceived in order to comprehensively solve the above-described problems, and was conceived not with the idea of correcting the calibration curve, but with the idea of correcting concentrations obtained from the calibration curve using temperature. [0008]
Here, using an equation of state for a gas, because the concentrations of measurement target components decrease as the absolute temperature thereof increases, correcting measurement discrepancies in concentrations obtained from a calibration curve using a correction formula in which the post-correction concentration increases as the absolute temperature increases may be considered. [0009]
The inventors of the present application measured the concentrations of CO2, CO, NO, and C3H8 when the temperature of a measurement cell was changed from 100 °C to 190 °C. The calibration curve used at this time was created using the measurement cell when the temperature thereof was 180 °C. [0010]
The changes in the concentration of each gas component at this time are shown in

FIG. 3. It was found that, as is shown in FIG. 3, the CO2 concentration could not be corrected by means of a correction formula that uses an equation of state for a gas that changes proportionally to the absolute temperature. It was also found that the NO concentration and the C3H8 concentration exhibited the same type of behavior depending on the temperature range. [0011]
When the test samples have high concentrations, because the optical absorption spectrum becomes saturated (i.e., peaks out), it is thought that the aforementioned behaviors may be due to the fact that the absorption of a wavelength band or wavenumber band having a small degree of absorbance located at the edge regions of the absorption band in the optical absorption spectrum is used. More specifically, in simulation results, in a central region of the absorption band in the optical absorption spectrum, the intensity decreases as the temperature increases according to conventional theory. However, in edge regions at a distance from the central regions, the results show that the intensity increases as the temperature increases. It is thought that the reason for this phenomenon is that, as is shown in FIG. 4, the existence probabilities when several rotation states having different energy levels are present change depending on the temperature. [0012]
In this manner, it is a principal object of the present invention to flexibly deal with and correct concentration changes that are caused by changes between the temperature when a correction calibration curve is created and the temperature when a test sample is measured, without having to rely on corrections made using an equation of state for a gas.
[Means for Solving the Problem]
[0013]
In other words, a spectrophotometer according to the present invention is a spectrophotometer that measures concentrations of measurement target components contained in a test sample from an optical spectrum obtained by irradiating light onto the test sample, and is provided with a concentration calculation unit that calculates concentrations of the measurement target components from the optical spectrum using a calibration curve, and a concentration correction unit that, using a temperature correction formula that corresponds to a wavelength region or a wavenumber region in which concentrations of the measurement target components are being determined, corrects concentration changes in the measurement target components that accompany temperature

differences between a temperature when the calibration curve is created and a temperature
when the concentrations are measured.
[0014]
Moreover, a spectrophotometry method according to the present invention is a spectrophotometry method in which concentrations of measurement target components contained in a test sample are measured from an optical spectrum obtained by irradiating light onto the test sample, comprising a concentration calculation step in which concentrations of the measurement target components are calculated from the optical spectrum using a calibration curve, and a concentration correction step in which concentration changes in the measurement target components that accompany temperature differences between a temperature when the calibration curve is created and a temperature when the concentrations are measured are corrected using a temperature correction formula that corresponds to a wavelength region or a wavenumber region in which concentrations of the measurement target components are being determined. [0015] According to the above-described invention, because concentrations of measurement target components are corrected using a temperature correction formula that corrects an amount of change in a temperature when a concentration is measured compared to a temperature when a calibration curve was created, concentrations of measurement target components can be corrected without the calibration curve having to be corrected. Moreover, by creating a temperature correction formula in advance for each measurement target component, corrections can be made so as to correspond to any increase or decrease behavior in the concentration of the measurement target components that are generated by increases or decreases in the absolute temperature. Furthermore, as a consequence of this, correcting concentrations of measurement target components can be performed in real time. [0016]
In an optical spectrum there are wavelength regions or wavenumber regions where absorbance signals generated by a plurality of components are mutually superimposed, and there are also wavelength regions or wavenumber regions where peak intensity becomes saturated. In order to reduce the effects of these, and perform accurate measurements of concentrations, the concentration calculation unit is formed so as to calculate concentrations of the measurement target components using predetermined wavelength regions or wavenumber regions in the optical spectrum. At this time, in order to perform temperature corrections accurately, it is desirable that the concentration

correction unit use a temperature correction formula that corresponds to the wavelength region or wavenumber region in which the concentrations of the measurement target components are being determined. [0017]
The wavelength region or wavenumber region used when concentrations of measurement target components are calculated by the concentration calculation unit vary depending on the type of the measurement target components and the measurement range of the measurement target components. Because of this, in order for temperature correction to be performed accurately, it is desirable that the concentration correction unit update the temperature correction formula in accordance with the type of the measurement target components or the measurement range of the measurement target components. [0018]
It is also desirable that the concentration calculation unit calculate concentrations of a plurality of measurement target components using multivariate analysis, and that the concentration correction unit correct the concentration of each measurement target component using the temperature correction formula set for each one of the plurality of measurement target components. [0019]
It is also desirable that, in order to favorably correct concentrations of measurement target components that exhibit behavior that cannot be corrected by means of correction that employs an equation of state for a gas, the temperature correction formula output progressively smaller values as the temperature increases. At this time, the temperature correction formula is used to correct changes in energy levels that are caused by the temperature of the measurement target components. [0020]
When the test sample is a gas, changes in energy levels that are caused by the temperature of the measurement target components tend to occur easily, so that the effects obtained when the present invention is applied are even more conspicuous. [0021]
As a specific embodiment for calculating the temperature correction formula automatically, it is desirable that there be further provided a standard spectrum acquisition unit that acquires respective optical spectra of a plurality of temperatures from standard test samples having known concentrations, and a correction formula creation unit that calculates the temperature correction formula from the standard spectra of the plurality of

temperatures. [0022]
The wavelength region or wavenumber region used by the concentration calculation unit is updated depending on whether or not interference components are present, the type of measurement target components, and the measurement range and the like. In order to make accurate corrections corresponding to these updates, it is desirable that, when the wavelength region or wavenumber region used for the concentration calculation performed by the concentration calculation unit is updated, the correction formula creation unit update the temperature correction formula so that it corresponds to the updated wavelength region or wavenumber region. [0023]
In a spectrophotometer, processing to update the calibration curve is performed at regular intervals. In order to update the correction formula so that it matches the update processing for the calibration curve, and thereby standardize the various processings, it is desirable that the correction formula creation unit calculate the temperature correction formula from the standard spectra obtained at the time the calibration curve was created.
[Effects of the Invention]
[0024]
According to the present invention which is formed in the above-described manner, it is possible, even for components that are difficult to correct simply by performing calculations using an equation of state for a gas, to correct temperature effects which differ for each component by flexibly dealing with concentration changes that are caused by changes between the temperature when a calibration curve was created and the temperature when a test sample was measured, without having to correct the calibration curve.
[Brief Description of the Drawings]
[0025]
[FIG. 1] FIG. 1 is a schematic view showing the structure of an infrared spectrophotometer according to the present embodiment.
[FIG. 2] FIG. 2 is a functional block diagram showing an information processing device of the same embodiment.
[FIG. 3] FIG. 3 depicts graphs showing relationships between the absolute

temperature and the concentration of each gas component.
[FIG. 4] FIG. 4 is a view showing changes in energy level that accompany changes in temperature.
[Best Embodiments for Implementing the Invention]
[0026]
Hereinafter, an embodiment of an infrared spectrophotometer according to the present invention will be described with reference to the drawings. [0027]
An infrared spectrophotometer 100 according to the present embodiment is an exhaust gas analyzer that, for example, measures, as time series data, concentrations of a plurality of components that are contained in exhaust gas, which is serving as a test sample, that is emitted from an internal combustion engine of an automobile or the like. [0028]
More specifically, as is shown in FIG. 1, this infrared spectrophotometer 100 collects, for example, either a portion of or all of the exhaust gas emitted from a tailpipe of an automobile using a test sample collecting unit 2, and then, without diluting it, introduces the exhaust gas collected by the test sample collecting unit 2 into a measurement cell 3, and then measures the respective concentrations of a plurality of components such as, for example, carbon monoxide (CO), carbon dioxide (CO2), nitric oxide (NO), nitrogen dioxide (NO2), nitrogen oxide (NOx), and water (H20) and the like that are contained in the exhaust gas using an FTIR method. A temperature control mechanism H such as a heater or the like that controls the temperature of the measurement cell 3 and of the gas inside the measurement cell 3 is provided peripherally to the measurement cell 3. The temperature of the measurement cell 3 is detected by a temperature sensor Tl. This temperature control mechanism H is controlled by a computing device COM based on temperatures detected by the temperature sensor Tl such that a predetermined temperature is obtained. [0029]
Moreover, in the infrared spectrophotometer 100, a suction pump 4 that is used to introduce the exhaust gas into the measurement cell 3 is provided on a downstream side of the measurement cell 3 on an exhaust gas line LI on which the measurement cell 3 is provided. In addition to these, valves 5 that adjust the flow rate of the exhaust gas, an orifice 51, a flowmeter 6 that measures the flow rate of the exhaust gas, and filters 7 that

remove dust particles from the exhaust gas, and the like are also provided on the exhaust gas line LI. Moreover, a calibration gas supply line L2 that supplies the measurement cell 3 with zero gas or span gas that is used to calibrate a photodetector 9, and a purge gas line L3 that is used to purify the exhaust gas line LI or the measurement cell 3 are also connected to the exhaust gas line LI or the measurement cell 3. [0030]
Furthermore, the infrared spectrophotometer 100 is also provided with a light irradiation unit 8 that irradiates interfered infrared light onto the measurement cell 3, and the photodetector 9 that detects an intensity of the light that is emitted after being transmitted through the measurement cell 3. The computing device COM of the infrared spectrophotometer 100 calculates an optical absorption spectrum in the optical spectrum of the exhaust gas using light intensity signals obtained by the photodetector 9, and calculates the concentrations of a plurality of measurement target components from the absorbance of a predetermined wavelength region or wavenumber region in this optical absorption spectrum. [0031]
As is shown in FIG. 2, the computing device COM of the infrared spectrophotometer 100 is provided with a calibration curve storage unit 10 that stores calibration curve data showing relationships between absorbances of measurement target components and concentrations of measurement target components, a concentration calculation unit 11 that calculates concentrations of measurement target components from optical absorption spectra and calibration curves shown by the calibration curve data, and a concentration correction unit 12 that corrects concentrations of measurement target components using temperature correction formulas. [0032]
The calibration curve storage unit 10 stores calibration curve data created at a predetermined temperature (for example, at a reference temperature). Relationships between the concentrations of each component (formed by a plurality of representative values, for example, in the case of CO, concentrations of 2%, 4%, 6%, 8%, and the like) and the respective absorbances thereof are determined on this calibration curve. This calibration curve data can be created by a calibration curve creation unit 16 (described below). Note that the calibration curve data may be in the form of an arithmetical expression, or may be in tabular form. [0033]

The concentration calculation unit 11 acquires light intensity signals from the photodetector 9, and also acquires calibration curve data from the calibration curve storage unit 10, and then calculates optical absorption spectra from the light intensity signals, and then uses multivariate analysis to calculate the concentrations of each measurement target component from the absorbance in a predetermined wavelength region or wavenumber region of the relevant optical absorption spectrum, and from the calibration curve shown by the calibration curve data. [0034]
The concentration correction unit 12 uses a temperature correction formula set for each relevant measurement target component to perform temperature correction on the concentrations of each measurement target component obtained by the concentration calculation unit 11 based on temperatures detected by the temperature sensor Tl. The correction formula data showing the temperature correction formula is stored in a correction formula storage unit 17. [0035]
Here, temperature correction formulas are used to correct changes in the concentrations of the measurement target components that accompany temperature differences between the temperature of the measurement cell 3 at the time when the calibration curve was created, and the temperature of the measurement cell 3 at the time when the concentration was measured. Temperature correction formulas are determined in accordance with the wavelength region or the wavenumber region in which the concentrations of the measurement target components are being determined. The temperature correction formulas of the present embodiment are determined from a polynomial in the form of a temperature-concentration relational formula (for example, the formula given below in Equation 2). [0036]
More specifically, the following formulas can be considered as examples of temperature-concentration relational formulas. Namely, (1) a formula in which the concentration increases as the absolute temperature increases, (2) a formula in which the concentration decreases as the absolute temperature increases, and (3) a formula in which, within a predetermined range, the concentration increases as the absolute temperature increases, while once that predetermined range is exceeded, the concentration decreases as the absolute temperature increases. [0037]

For example, the temperature-concentration relational formula for CO2 is a formula in which the concentration increases steadily as the temperature increases. Moreover, the temperature-concentration relational formula for CO is a formula in which the concentration decreases steadily as the temperature increases. Furthermore, the temperature-concentration relational formula for NO is a formula in which the concentration switches from a steady increase as the temperature increases to a steady decrease. Note that because it may be thought that the change in concentration arising from the change in temperature in the case of CO2 is caused by a change in the energy level that is generated by the change in temperature, the temperature correction formula for CO2 can be described as a formula to correct the change in the energy level that is generated by the change in temperature. [0038]
In order to create this temperature correction formula, the computing device COM of the infrared spectrophotometer 100 is further provided with a standard spectrum acquisition unit 13 that acquires optical absorbance spectra for each of a plurality of temperatures from standard test samples having known concentrations, and a correction formula creation unit 14 that creates temperature correction formulas from the standard spectra of the plurality of temperatures. [0039]
The standard spectrum acquisition unit 13 acquires a plurality of sets of spectrum data that show the respective standard spectra of a plurality of temperatures of a standard test sample at the time when a calibration curve was created or the like, and stores these sets of spectrum data in a standard spectra storage unit 15. Here, the standard spectrum data may be obtained as a result of the standard spectrum acquisition unit 13 acquiring light intensity signals from the photodetector 9 and then calculating the standard spectrum data, or may be obtained as a result of a separate functional block such as the concentration calculation unit 11 acquiring the light intensity signal from the photodetector 9 at the time when the calibration curve was created, and then calculating the standard spectrum data, which is then received by the standard spectrum acquisition unit 13. [0040]
The correction formula creation unit 14 creates correction formula data showing temperature correction formulas that correspond to the respective wavelength region or wavenumber region used by the concentration calculation unit 11 to calculate the concentrations of the plurality of measurement target components. Additionally, when the

wavelength region or wavenumber region used in the calculations for the respective measurement target components by the concentration calculation unit 11 is updated, the correction formula creation unit 14 creates updated correction formula data showing temperature correction formulas that correspond to the relevant updated wavelength region or wavenumber region. [0041]
Next, processing to calculate a temperature correction formula which is performed in conjunction with the creation of the calibration curve in the infrared spectrophotometer 100 will be described. [0042]
A standard test sample (i.e., calibration gas) having a known concentration is introduced into the measurement cell 3 whose temperature has been adjusted to a predetermined reference temperature (for example, 100 °C). In this state, infrared light is irradiated onto the measurement cell 3 from the light irradiation unit 8, and light that has been transmitted through the measurement cell 3 is detected by the photodetector 9. The standard spectrum acquisition unit 13 or the like then calculates a standard spectrum from a light intensity signal output from the photodetector 9, and stores this in the standard spectra storage unit 15. Here, the temperature of the measurement cell 3 into which the calibration gas has been introduced is then raised from the reference temperature by, for example, 10 °C each time, and the standard spectrum data at each temperature is also acquired. [0043]
Next, the calibration curve creation unit 16 of the spectrophotometer 100 decides the wavelength region or wavenumber region to be used when calculating the concentration of each measurement target component from the standard spectrum data obtained at the reference temperature, and creates calibration curve data showing a calibration curve using the absorbance in the relevant wavelength region and wavenumber region. [0044]
Furthermore, the concentration calculation unit 11 then determines the concentrations of the measurement target components at each temperature in the relevant wavelength region and wavenumber region from the calibration curve data and the standard spectrum data obtained at the other temperatures. In addition, the correction

formula creation unit 14 creates correction formula data showing a temperature correction formula from the concentrations the measurement target components at each temperature obtained from the standard spectrum data, and from the known concentrations of the measurement target components of the calibration gas. [0045]
Hereinafter, the temperature correction formula used by the concentration correction unit 12 will be described. The temperature correction formula used by the concentration correction unit 12 can be shown by the following. [0046] [Equation 1]
,""r uy T + a2T2 + a{ T + *;0
Here, Ccomp is the post-correction concentration of the measurement target components, and Cunk is the pre-correction concentration of the measurement target components. The coefficients ao ~ a.3 are fitting parameters. [0047]
This formula (Equation 1) can be determined in the following manner. [0048]
If fitting is performed using graphs (these are the graphs shown in FIG. 3 in which the horizontal axis shows temperature while the vertical axis shows concentration) of the concentrations of measurement target components at each temperature obtained using a calibration gas having a known concentration, then the following relational expression (i.e., a temperature - concentration relational expression) is obtained. Note that in the following a third-order equation is used, however, it is also possible for a fourth-order equation or greater, or a first-order or second-order equation to be used. [0049] [Equation 2]
— = a3T +