JP2017173291A - Substance analysis method and analysis device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a substance analysis method and an analysis device with which it is possible to measure an absorption rate with high accuracy.SOLUTION: The substance analysis method comprises: a first step of estimating, within a frequency band that includes at least one absorption line of an object substance included in a measurement sample, a tentative reference signal from a sample signal of a low absorption rate section on the basis of a sample signal that corresponds to the intensity of light having passed through the measurement sample; a second step of subtracting the tentative reference signal from the sample signal and calculating the distribution of tentative absorption rate; a third step of determining a wavenumber section in which the tentative absorption rate is less than or equal to a prescribed value, and re-estimating a reference signal on the basis of a sample signal of the wavenumber section; and a fourth step of subtracting the re-estimated reference signal from the sample signal and calculating the distribution of absorption rate.SELECTED DRAWING: Figure 3

Description

  Embodiments described herein relate generally to a material analysis method and an analysis apparatus.

  Gas has an intrinsic absorption spectrum for infrared.

  By measuring the infrared intensity transmitted through the sample gas, the gas absorption coefficient and gas concentration can be calculated.

  Infrared rays having a wavelength of 3 μm or more can be emitted from the quantum cascade laser. However, the laser beam causes output fluctuations and wavelength fluctuations. If the measurement value is corrected each time using the reference signal, an error due to characteristic variation can be reduced.

  When performing gas analysis over a long period of time, such as environmental measurement, it may be difficult to measure the reference signal for each measurement.

JP 2003-232732 A

  Provided are a material analysis method and an analysis apparatus capable of measuring an infrared absorption rate with high accuracy.

  The substance analysis method according to the embodiment has a low absorption rate based on a sample signal corresponding to the intensity of light transmitted through the measurement sample in a wave number band including at least one absorption line of the target substance included in the measurement sample. A first step of estimating a provisional reference signal from a sample signal in a section; a second step of calculating a distribution of provisional absorption rate by dividing the provisional reference signal from the sample signal; and And a third step of re-estimating a reference signal based on a sample signal in the wave number interval, and dividing the reference signal re-estimated in the third step from the sample signal to And a fourth step of calculating the distribution.

It is a schematic diagram which illustrates the gas analyzer used for the gas analysis method concerning 1st Embodiment. 2A to 2C are graphs for explaining the effect of tuning the infrared wave number to the absorption line, FIG. 2A is a waveform diagram of the operating current of the QCL, and FIG. FIG. 2C is a graph showing a change in wavelength, and FIG. 2C is a graph showing a detector output. It is a flowchart of the gas analysis method concerning a 1st embodiment. 4 (a) shows the detector output of the first sample signal, FIG. 4 (b) shows the first absorption rate, FIG. 4 (c) shows the first determined wave number interval, and FIG. It is a graph showing the reference estimated signal determined in the second. It is a graph explaining the gas analysis method concerning a comparative example. 6A to 6C are graphs for explaining the measurement principle of the isotope ratio, FIG. 6A is a graph showing the detector output dependence with respect to time, and FIG. 6B is the wave number. FIG. 6C is a graph showing the absorption coefficient dependency after fitting to the wave number. FIG. 7A is a graph showing the absorption coefficient dependency on the wave number with the pressure as a parameter, and FIG. 7B is a graph showing the absorption coefficient dependency on the wave number with the temperature as a parameter. It is a graph explaining the determination method of a fitting range. It is a graph explaining the reference signal estimation method concerning 3rd Embodiment. It is a graph explaining the reference signal estimation method concerning 4th Embodiment.

Embodiments of the present invention will be described below with reference to the drawings.
FIG. 1 is a schematic view illustrating a gas analyzer used in the gas analysis method according to the first embodiment.
As shown in FIG. 1, the gas analyzer 110 includes a cell unit 20, a light source unit 30, a detection unit 40, and a control unit 45.

  A sample gas 50 is introduced into the cell unit 20. That is, the sample gas 50 is introduced into the space 23 s provided in the cell unit 20. The sample gas 50 includes carbon dioxide (including isotopes), hydrogen sulfide, acetone, ammonium, and the like.

  The light source unit 30 causes the infrared rays 30L to enter the space 23s. Moreover, the light source part 30 changes the wavelength of the infrared rays 30L by the drive part 30b.

  In this example, the light source unit 30 includes a QCL (Quantum Cascade Laser) 30a. As the QCL 30a, for example, a distributed feedback (DFB) type QCL is used. The QCL 30a may be an edge emitting type or a surface emitting type.

  For example, the detection unit 40 detects the infrared ray 30L that has passed through the space 23s in a state where the sample 50 is introduced into the space 23s. The detection unit 40 detects the intensity I of the transmitted light that has passed through the space 23s. For the detection unit 40, an element having sensitivity in the infrared region is used. For example, a thermopile or a semiconductor sensor element (for example, InAsSb) is used.

When the intensity of incident light is I ₀ , the absorbance is A, the transmittance is T _R , the absorption coefficient (cm ⁻¹ ) is α, and the optical path length of the cell unit 20 is L, the absorbance A is expressed by the formula (1). Is done.

When the molar concentration is ε, the absorption coefficient (cm ⁻¹ ) is expressed by Equation (2).

  Further, the molar concentration c of the sample gas is represented by the formula (3).

  Based on the measured transmitted light intensity I, the control unit 45 calculates an absorption coefficient α, an absorption rate B, a molar concentration c, an absorption spectrum based on the HITRAN database, and the like.

FIG. 2A to FIG. 2C are graphs for explaining the effect of tuning the wave number of infrared rays to an absorption line. 2A is a waveform diagram of the operating current of the QCL, FIG. 2B is a graph showing the change in wavelength, and FIG. 2C is a graph showing the detector output.
In FIG. 2A, the vertical axis represents the current J of the QCL 30a, and the horizontal axis represents time t. In FIG. 2B, the vertical axis represents wavelength and the horizontal axis represents time t. The wave number is the reciprocal of the wavelength. In FIG. 2C, the vertical axis represents the detector output Sg, and the horizontal axis represents time t. The current J supplied to the QCL 30a changes monotonously as the time t increases. As shown in FIG. 2A, when the current is a triangular wave, the wavelength λ increases with time within one period (FIG. 2B), the infrared wavelength becomes λ1 at time t1, and the infrared wavelength at time t2 becomes λ2 (> λ1). In FIG. 2, the right half is when the sample gas containing the target gas is introduced into the cell, and the left half is when the reference gas not containing the target gas is introduced.

  For example, the wavelength λ1 corresponds to the first absorption line, and the wavelength λ2 corresponds to the second absorption line. In the vicinity of each absorption line, in the right half where the sample gas containing the target gas is introduced into the cell, the infrared light is absorbed by the sample gas 50, so that the infrared detector output transmitted through the sample gas 50 is the target gas. The reference gas that does not contain is reduced compared to the left half introduced into the cell. In addition, if a triangular wave is repeatedly generated for one sample gas 50 and an average value of detector outputs is obtained, measurement accuracy can be improved.

In the light source unit 30, the wavelength λ is changed in a short time (for example, about 100 ms or less). For example, when the center value of the wavelength λ of infrared rays is set to be not less than 4.345 micrometers and not more than 4.384 micrometers, carbon dioxide absorption lines are included. In this case, the wave number band is 2281 cm ⁻¹ or more and 2301 cm ⁻¹ or less.

In the gas analysis, (I / I ₀ ) used in the equation (1) matches the ratio of the detector output with and without the absorption. Therefore, a “reference signal” when there is no absorption is required for the “sample signal” that is the detector output that has passed through the sample gas. In the prior art, as described above, this reference signal is obtained by measuring a reference gas that does not include the target gas. However, in the present invention, since it is obtained from the sample signal, measurement with the reference gas is not required.

FIG. 3 is a flowchart of the gas analysis method according to the embodiment.
4A shows the detector output of the sample signal, FIG. 4B shows the provisional absorption rate, FIG. 4C shows the determined wave number interval, and FIG. 4D shows the determined reference estimation signal. FIG.
In FIG. 4A, the vertical axis represents the detector output (V), and the horizontal axis represents time (s). In FIG. 4B, the vertical axis represents the absorption rate, and the horizontal axis represents time (s). In FIG.4 (c), a vertical axis | shaft is an absorptivity and a horizontal axis is time (s).

  The wave number of infrared rays is changed within the wave number band including the absorption line of at least one target gas contained in the sample gas 50, and the change in the transmitted light intensity I of the infrared ray that has passed through the sample gas 50 is measured to obtain a sample signal ( First step S100 and FIG. 4 (a)).

Samples in a plurality of sections where the absorptance is always low (a time range in which the absorptance is always low even when the wavelength variation of the laser light is taken into account, “low” means, for example, 5 × 10 ⁻⁵ or less) A provisional reference signal is estimated from the interpolation of the signal level, for example, by a quadratic curve (second step S102).

  The provisional reference signal is divided from the sample signal to calculate the provisional absorption rate (third step S104 and FIG. 4B).

A plurality of wave number intervals of a predetermined absorption rate (for example, 5 × 10 ⁻⁵ ) or less are determined, and the reference signal is re-estimated from interpolation of, for example, a quadratic curve of the level of the sample signal in the wave number interval (fourth step) S106 and FIG. 4 (d)).

  The reference signal re-estimated from the sample signal is divided to calculate the absorption rate (fifth step S108).

  If there is a sample gas to be measured next, the process returns to the first step S100, and if not, the process ends (sixth step S110).

FIG. 5 is a graph for explaining a gas analysis method according to a comparative example.
The vertical axis represents the detector output (V), and the horizontal axis represents time t. In the comparative example, wave number intervals F11 and F12 for estimating the reference signal by interpolation are fixed on the time axis. On the other hand, the characteristics of QCL fluctuate as shown in FIG. 5 on the time axis due to changes over time and changes in the environment (temperature, pressure, etc.) inside and outside the gas cell. For this reason, if the wave number bands F11 and F12 are fixed in the range on the time axis, the reference signal is correctly estimated in the sample signal A1, but the estimated reference signal is less than the original reference estimation signal in the sample signal A2. Thus, the error in calculating the absorption rate B increases.

On the other hand, in the gas analysis method according to the first embodiment, the wave number at which the reference signal is estimated by interpolation in the wave number section in which the absorption rate B is lower than a predetermined value (for example, 5 × 10 ⁻⁵ or less). The interval is determined for each sample gas. For this reason, since the wave number section with a high absorption rate B is not included in the interpolation range, the estimation accuracy of the reference signal is not impaired, and the measurement accuracy of the absorption rate B can be increased.

  Next, consider a case where there are at least two absorption lines generated by an isotope element in the wave number section. In the gas analysis method according to the second embodiment, a bimodal absorption spectrum curve is separated into two spectra by fitting, and two isotope ratios are calculated.

FIGS. 6A to 6C are graphs for explaining the measurement principle of the isotope ratio. 6A is a graph showing the detector output dependency with respect to time, FIG. 6B is a graph showing the absorption coefficient dependency with respect to the wave number, and FIG. 6C is an absorption coefficient after fitting with respect to the wave number. It is a graph showing a dependency.
In FIG. 6A, the vertical axis represents the detector output, and the horizontal axis represents time. In FIG. 6B, the vertical axis represents the absorption coefficient, and the horizontal axis represents the wave number (cm ⁻¹ ). In FIG.6 (c), a vertical axis | shaft is an absorption coefficient and a horizontal axis is a wave number (cm ^<-1> ).

In this figure, the isotopes are ¹³ CO ₂ and ¹² CO ₂ . As shown in FIG. 6A, the reference estimation signal is calculated by the gas analysis method according to the first embodiment. As shown in FIG. 6B, the absorption coefficient distribution is calculated by dividing the intensity distribution of the reference estimation signal from the intensity distribution of the detector output. As shown in FIG. 6C, the absorption coefficient spectra of ¹³ CO ₂ and ¹² CO ₂ are separated by fitting with theoretical values.

  Next, a gas absorption model will be described in order to perform absorption spectrum fitting. In the gas absorption model, the absorption spectrum is calculated using the value of the HITRAN database. The calculation is performed for each absorption line (3190 lines) in consideration of pressure dependency and temperature dependency.

  The absorption coefficient αi is calculated by equation (4).

  Each absorption line intensity S (T) is calculated by the equation (5).

  The spectrum shape function f (ν, T, p) is calculated by the equation (6).

FIG. 7A is a graph showing the absorption coefficient dependence on the wave number with pressure as a parameter, and FIG. 7B is a graph showing the absorption coefficient dependence on the wave number with temperature as a parameter.
7A and 7B, the vertical axis represents the absorption coefficient (cm ⁻¹ ), and the horizontal axis represents the wave number (cm ⁻¹ ).

  As shown in FIG. 7A, when the pressure is changed to 0.4 atm, 0.5 atm, and 0.6 atm, the fluctuation of the wave number at which the absorption coefficient peaks is small, but the minimum value of the absorption coefficient varies. Further, as shown in FIG. 7B, when the temperature is changed to 296K, 300K, and 304K, the fluctuation of the wave number at which the absorption coefficient peaks is small, but the maximum value of the absorption coefficient varies. By using such a gas absorption model and fitting an absorption spectrum obtained from a measured value and a theoretical value, for example, the absorption spectra of two isotopes are separated to improve the accuracy of the isotope ratio. be able to.

FIG. 8 is a graph illustrating an example of a method for determining the fitting range.
The spectrum curve fitting range of the absorption rate B is, for example, 13 or more of the absorption lines of ¹³ CO ₂ and ¹² CO ₂ , and the signal intensity ratio SNR is 10 or more of each peak intensity. Can be a region.

Expressed in the isotope ratio of CO ₂ (the abundance ratio ^{_{= 13 CO 2/12 CO 2}} )%, becomes large number of decimal places. For this reason. It is convenient to use the delta notation in which the degree of deviation from the natural isotope ratio is expressed in permil (‰). The natural isotopic ^ratio, and ^{_{(13 CO 2/12 CO 2}} ) ref = 1.12372%.

  Formula (7) represents the display formula by the delta notation.

For example, when environmental gas analysis is performed, the concentration ratio between ¹³ CO ₂ and ₁₂ CO ₂ can be represented by δ ¹³ C (‰). The tropospheric free air δ ¹³ C is −8 ‰, the fossil fuel concentration ratio δ ¹³ C is −28 ‰, and the soil δ ¹³ C is −25.7 ‰. _The concentration ratio δ ¹³ C by the _two generation sources is different. For this reason, it is possible to specify the generation source of CO ₂ by measuring the concentration ratio δ ¹³ C.

Similarly, the value of the concentration ratio δ ¹³ C varies depending on the volcanic activity level. For this reason, it can utilize for an eruption prediction by measuring density | concentration ratio (delta) ^{13C in the} vicinity of a crater.

  That is, in the second embodiment, the isotope ratio can be obtained with high accuracy by fitting the measured value of the absorption coefficient including two isotopes with the theoretical spectrum.

FIG. 9 is a graph for explaining the reference signal estimation method according to the third embodiment.
In the third embodiment, as shown in FIG. 9, when there are two isotope absorption spectra in the sample signal, the reference signals at the respective absorption wave numbers are individually received from the low absorption sections on both sides. Interpolate. In the first embodiment, since the reference estimation is performed by using, for example, one quadratic curve from all the low absorption sections in the sample signal, the error of the reference signal increases in each absorption wave number. Errors can be suppressed.

FIG. 10 is a graph illustrating the reference signal estimation method according to the fourth embodiment.
In the fourth embodiment, as shown in FIG. 10, when there are absorption spectra of a plurality of target gases in the sample signal, the wave number is associated with the peak time, and the theoretical value determined from, for example, the HITRAN database or the like. Only the low-absorption wavenumber interval is subdivided, and the reference signal is estimated by interpolating them. In the first embodiment, a range that is not theoretically low absorption is included in the interpolation interval, whereas in this embodiment, since interpolation is performed only from the lower absorption interval, the estimation accuracy of the reference signal Will improve.

  According to the embodiment of the present invention, a gas analysis method capable of measuring the absorption rate with high accuracy is provided. The reference estimation signal for correcting the measured value of the transmitted light intensity is estimated for each sample gas based on the intensity distribution of the absorptance. For this reason, it is not necessary to measure the reference signal using the reference gas, and the analyzer can be configured simply.

  Although several embodiments of the present invention have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.

30 light source, 30a QCL, 30L infrared, 40 detector, 50 sample gas, S absorption line intensity, B absorption rate, α, αi absorption coefficient, F1, F2, F3 wave number interval, SNR signal to noise ratio

The substance analysis method according to the embodiment has a low absorption rate based on a sample signal corresponding to the intensity of light transmitted through the measurement sample in a wave number band including at least one absorption line of the target substance included in the measurement sample. A first step of estimating a provisional reference signal from a sample signal in a section; a second step of calculating a distribution of provisional absorption by dividing the provisional reference signal from the sample signal; and fitting the provisional absorption to a theoretical spectrum And a third step of calculating the concentration of the target substance .

  The wave number of infrared rays is changed within the wave number band including the absorption line of at least one target gas contained in the sample gas 50, and the change in the transmitted light intensity I of the infrared ray that has passed through the sample gas 50 is measured to obtain a sample signal ( Step S100 and FIG. 4 (a)).

Samples in a plurality of sections where the absorptance is always low (a time range in which the absorptance is always low even when the wavelength variation of the laser light is taken into account, “low” means, for example, 5 × 10 ⁻⁵ or less) The provisional reference signal is estimated from the interpolation of the signal level by, for example, a quadratic curve (step S102).

  The provisional reference signal is divided from the sample signal to calculate the provisional absorption rate (step S104 and FIG. 4B).

The concentration of the target substance is calculated by fitting the provisional absorption rate with the theoretical spectrum (step 106).

  If there is a sample gas to be measured next, the process returns to step S100, and if not, the process ends (step S110).

Claims (10)

Based on the sample signal corresponding to the intensity of the light transmitted through the measurement sample in the wave number band including at least one absorption line of the target substance contained in the measurement sample, the provisional reference signal from the sample signal in the interval where the absorption rate is low A first step of estimating
A second step of calculating a provisional absorption distribution by dividing the provisional reference signal from the sample signal;
A third step of determining a wave number interval in which the temporary absorption rate is equal to or less than a predetermined value, and re-estimating a reference signal based on a sample signal of the wave number interval;
A fourth step of calculating an absorptance distribution by dividing the reference signal re-estimated in the third step from the sample signal;
A material analysis method comprising:   The material analysis method according to claim 1, wherein the wave number changes monotonously with time.   The material analysis method according to claim 1 or 2, wherein when there are at least two absorption lines generated by an isotope element in the wave number band, the absorption rate is separated into at least two spectra by fitting.   4. The fitting range of the spectrum curve of the absorptance is a spectral intensity region in which each peak intensity is about one-half or more of the respective absorption lines and the signal-to-noise ratio is about 10 or more. The substance analysis method described.   The material analysis method according to claim 1, wherein the light is emitted from a quantum cascade laser. An analyzer for analyzing the measurement sample based on a sample signal corresponding to the intensity of light transmitted through the measurement sample in a wave number band including at least one absorption line of the target substance contained in the measurement sample,
A provisional reference signal is estimated from a sample signal in a section with a low absorption rate, a provisional absorption ratio distribution is calculated by dividing the provisional reference signal from the sample signal, and a wave number section in which the provisional absorption ratio is a predetermined value or less An analysis apparatus comprising: a control unit that determines, re-estimates a reference signal based on a sample signal in the wave number interval, and calculates an absorption rate distribution by dividing the reference signal re-estimated from the sample signal. A light source unit that changes a wave number of light within a wavelength band including at least one absorption line of a target substance contained in a measurement sample;
A cell part into which the measurement sample is introduced;
A detector that collects data for calculating the absorption rate of the target substance by measuring the intensity of the light transmitted through the measurement sample;
With
The data is obtained by estimating a provisional reference signal from a sample signal in a section with a low absorption rate, dividing the provisional reference signal from the measurement sample to calculate a distribution of the provisional absorption rate, and the provisional absorption rate is equal to or less than a predetermined value. The light for calculating a distribution of absorption by dividing the reference signal re-estimated from the sample signal, re-estimating the reference signal based on the sample signal in the wave number interval An analytical device comprising a measurement value of said intensity.   The analyzer according to claim 7, wherein the light source unit includes a quantum cascade laser capable of emitting the light.   The analyzer according to claim 7 or 8, wherein the light source unit monotonously changes the wave number with time.   10. The controller according to claim 7, wherein, when there is at least two absorption lines generated by an isotope element in the wave number band, the control unit separates the absorption rate into at least two spectra by fitting. The analyzer described.