JP5169005B2 - Spectrophotometer and measurement signal correction method - Google Patents

Description

  The present invention is a spectrophotometer such as a Fourier transform infrared spectrophotometer, and the influence of absorption by unnecessary components other than the sample from the signal measured by the spectrophotometer, particularly water vapor and carbon dioxide present in the atmosphere. The present invention relates to a measurement signal correction method for removing the influence of absorption of light.

  In the measurement with a spectrophotometer such as a Fourier transform infrared spectrophotometer (FTIR), in general, a sample spectrum obtained by performing measurement with a sample in the measurement optical path and a sample removed from the measurement optical path. By performing background correction using the background spectrum obtained by performing measurement (so-called blank measurement), the absorbance and transmittance of the sample are calculated. In an infrared spectrophotometer using light in the infrared region, an absorption peak due to water vapor or carbon dioxide in the atmosphere existing on the measurement optical path overlaps with an absorption spectrum of the sample, which becomes one of the backgrounds. However, if the measurement atmosphere changes between the sample measurement and the background measurement (blank measurement), the above-mentioned unnecessary components such as water vapor and carbon dioxide may be used even if a calculation process is performed to divide the sample spectrum by the background spectrum. There is a problem in that the influence of absorption by is not completely removed.

  On the other hand, Patent Document 1 discloses a measurement signal correction method that automatically removes the influence of unnecessary components as described above from a sample spectrum with high accuracy and without relying on manual work. In this method, by first obtaining the outer contour of the background spectrum, a spectrum from which absorption peaks due to unnecessary components are removed is obtained, and an unnecessary component peak pattern representing the transmittance spectrum of the unnecessary components is obtained from this spectrum and the original background spectrum. calculate. Then, the expansion / contraction rate for adjusting the unnecessary component peak pattern to the level of the sample spectrum is calculated, and the sample spectrum is corrected by using the unnecessary component peak pattern expanded / contracted by the expansion / contraction rate, so that it is superimposed on the sample spectrum. The effect of unnecessary components is removed.

  However, even with the conventional measurement signal correction method described above, the accuracy of removing absorption peaks due to unnecessary components may decrease. That is, since sample measurement and background measurement are performed at a certain interval, if atmospheric fluctuation occurs on the optical path during this time, this is included in the unwanted component peak pattern as a fluctuation component, and after removal correction processing The fluctuation component is superimposed on the sample spectrum. In particular, when the peak of the change in transmittance due to water vapor or carbon dioxide in the unnecessary component peak pattern is small, the influence of the fluctuation component as described above appears relatively large, so the influence of the unnecessary component is removed. The accuracy of correction may be reduced.

Japanese Patent No. 3767490

  The present invention has been made in order to solve the above-mentioned problems, and the object of the present invention is to perform correction for removing the influence of absorption due to unnecessary components in the atmosphere in consideration of the influence of atmospheric fluctuations. An object of the present invention is to provide a spectrophotometer and a measurement signal correction method capable of improving the accuracy of removal correction, and by extension, obtaining the absorbance and transmittance of a sample with higher accuracy.

The spectrophotometer according to the first invention made to solve the above problems is as follows.
a) Spectroscopic measurement means for acquiring background spectral data in a state where there is no sample in the measurement optical path and acquiring sample spectral data in a state where there is a sample in the measurement optical path;
b) an unnecessary component spectrum calculating means for calculating an unnecessary component spectrum reflecting absorption by an unnecessary component in the atmosphere superimposed on the spectrum from the background spectrum data;
c) an expansion / contraction rate calculating means for calculating an expansion / contraction rate for adjusting the unnecessary component spectrum to the level of the sample spectrum data;
The constants for removing the influence of the spectral variations due to fluctuations of the d) air, the expansion ratio and the reference to the unwanted component spectra overlap the sample spectral data and said background spectrum data after assuming the constants Constant determining means for determining based on the validity of the results when correction is performed so as to remove the influence of absorption by unnecessary components in the atmosphere,
e) After correcting the unnecessary component spectrum with the constant determined by the constant determining means, the unnecessary component spectrum in the atmosphere is superposed on the sample spectrum data using the corrected unnecessary component spectrum and the expansion / contraction rate. Unnecessary component removal correction means for performing correction correction of the influence of absorption by the component;
It is characterized by having.

Moreover, the measurement signal correction method of the spectrophotometer according to the second invention is a measurement signal correction method used for the spectrophotometer according to the first invention,
a) an unnecessary component spectrum calculation step for calculating an unnecessary component spectrum reflecting absorption by an unnecessary component in the atmosphere superimposed on the spectrum from background spectrum data acquired in a state where there is no sample in the measurement optical path;
b) an expansion / contraction rate calculating step for calculating an expansion / contraction rate for adjusting the unnecessary component spectrum to a level of sample spectrum data acquired in a state where the sample is present in the measurement optical path;
The constants for removing the influence of the variations due to fluctuations of c) air, superimposed on the sample spectral data and said background spectrum data using said unwanted component spectrum and the stretch ratio in terms of assuming the constants A constant determination step that is determined based on the validity of the result when correction is performed so as to remove the influence of absorption by unnecessary components in the atmosphere,
d) After correcting the unnecessary component spectrum with the constant determined in the constant determining step, the unnecessary component spectrum in the atmosphere superimposed on the sample spectrum data using the corrected unnecessary component spectrum and the expansion / contraction rate is used. An unnecessary component removal correction step for performing correction correction of the influence of absorption by the component,
It is characterized by including.

  That is, in the spectrophotometer according to the first invention adopting the measurement signal correction method according to the second invention, the unnecessary component spectrum calculated by the unnecessary component spectrum calculating means is contracted at the expansion ratio calculated by the expansion ratio calculating means. Or, by adjusting the level by stretching and removing the influence of unnecessary components superimposed on the sample spectrum, the waveform of the unnecessary component spectrum is corrected in consideration of the factor of atmospheric fluctuation, and the correction is made. The unnecessary component spectrum is used to correct the sample spectrum.

However, it is not possible to directly determine the extent of the influence of fluctuations caused by atmospheric fluctuations. Therefore, here, it is assumed that there is a fluctuation in the spectrum waveform caused by the fluctuation of the atmosphere, and a constant for removing the influence of the fluctuation is considered. The constant determination means assumes the value of the above constant and corrects the spectrum so as to eliminate the influence of absorption by unnecessary components in the atmosphere. There is determined the constants as appropriate as the value of the constant.

As a more specific aspect, the constant determination means performs the correction when the correction is performed so as to remove the influence of absorption by unnecessary components in the atmosphere superimposed on the sample spectrum data and the background spectrum data. The constant may be determined so as to minimize the sum of squares of the difference in transmittance with respect to the reference spectrum of both later spectra. Here, as the reference spectrum, for example, a spectrum obtained by removing a peak appearing in the unnecessary component spectrum from the background spectrum, that is, an absorption peak due to the unnecessary component can be used.

Although the constants determined in this way are simply obtained, at least the effect of fluctuations in atmospheric fluctuation that occurred during the time interval between sample measurement and background measurement is Thus, the transmittance and absorbance of the sample obtained can be reduced most. Note that the peak of the unnecessary component spectrum calculated as described above may be larger or smaller than the actual one due to fluctuations caused by atmospheric fluctuations. Therefore, the constant can take both positive and negative values.

Therefore, according to the spectrophotometer according to the first invention and the measurement signal correction method according to the second invention, interference caused by atmospheric fluctuations at the time of sample measurement and at the time of background measurement is reduced. It may also be removed correcting the influence of absorption due to the unwanted components such as water vapor and carbon dioxide in the atmosphere Te than with high accuracy. Thereby, calculation accuracy, such as the transmittance | permeability and light absorbency by a sample, can be improved.

  Hereinafter, FTIR which is one embodiment of the spectrophotometer according to the first invention to which the measurement signal correction method according to the second invention is applied will be described with reference to the drawings. FIG. 1 is a schematic configuration diagram of FTIR according to this embodiment.

  In this FTIR, infrared light emitted from the light source 1 is incident on an interferometer 2 such as a Michelson interferometer, and the interferometer 2 has an interferogram whose amplitude periodically varies with time, that is, an interferogram. Is generated, and the sample 4 is irradiated as the measurement light 3. The measurement light 3 transmitted through the sample 4 or reflected by the sample 4 is detected by the detector 5, and the detection signal is input to the data processing unit 6. The data processing unit 6 includes a measurement spectrum data acquisition unit 61, an unnecessary component removal correction unit 62, and an absorption spectrum calculation unit 63 as functions.

  The measurement spectrum data acquisition unit 61 converts a time element into a wave number (or frequency) element by performing a Fourier transform operation on the detection signal obtained by the detector 5 so that the horizontal axis represents wave number and the vertical axis represents energy (intensity). ) Is created. This spectrum is a spectrum including the background. The unnecessary component removal correction unit 62 performs correction to remove the influence due to absorption of unnecessary components such as water vapor and carbon dioxide, which is caused by the difference between the sample measurement and the background measurement. Further, the absorption spectrum calculation unit 63 uses the spectrum in which the influence of the unnecessary component is removed and corrected as described above, removes the influence of the background and purely reflects the absorption by the sample (or transmittance spectrum). Create

  The various data processing functions in the data processing unit 6 can be realized by executing dedicated control / processing software installed in a personal computer.

FTIR of the present embodiment is characterized in signal correction to be executed by the unnecessary component removing correcting unit 62 of the data processing unit 6. 2 is a flowchart showing a processing procedure centering on this signal correction, FIG. 3 is a flowchart showing a specific processing procedure of step S5 in FIG. 2, and FIGS. 4 to 9 are diagrams for explaining the signal correction operation. It is a waveform diagram.

  First, the sample measurement in which the sample 4 is inserted into the optical path of the measurement light 3 and the background measurement in which the sample 4 is removed from the measurement optical path are performed, and the measurement spectrum data acquisition unit 61 reflects the absorption by the sample 4 Spectral data and background spectral data without absorption by the sample 4 are acquired (step S1). FIG. 4 is a diagram showing an example of these two spectra. Since unnecessary components having the property of absorbing infrared light, such as water vapor and carbon dioxide, are present in the measurement optical path, peaks due to absorption of such unnecessary components are superimposed on both spectra.

  These two pieces of spectrum data are given to the unnecessary component removal correction unit 62, and the unnecessary component removal correction unit 62 executes a correction process for removing the influence of the unnecessary component in the following procedure. First, the outer contour of this spectrum is calculated based on the background spectrum data (step S2). Specifically, the outer contour can be obtained by the following procedure, for example.

  First, the curve of the background spectrum is traced in the wave number increasing direction (or decreasing direction), and the peak top points of the background spectrum are detected by searching for points where the gradient changes from positive to negative. Next, the value of a point between adjacent peak top points is replaced with an interpolated value using the values of the peak top points on both sides. Then, the detection and interpolation of the peak top points are repeated until the interval between all peak top points becomes equal to or greater than a specific value. About this specific value, according to the waveform shape of the absorption peak of the unnecessary component to be removed and corrected, an appropriate value is obtained experimentally so that the absorption peak is filled by interpolation, and this is set in advance. Good.

FIG. 5 is a diagram showing the background spectrum shown in FIG. 4 and the outer contour obtained therefrom by the above procedure. This outer contour can be regarded as a waveform obtained by removing absorption peaks due to unnecessary components in the atmosphere from the background spectrum. Thus, when the outer contour of the background spectrum is obtained in this way, an unnecessary component peak pattern is calculated by dividing the background spectrum data by the value of the outer contour at each wave number position (step S3). That is, when the background spectrum is BG and the outer contour of the background spectrum is BOL, the peak pattern PP of the unnecessary component is calculated by the following equation (1).
PP = BG / BOL (1)
FIG. 6 is a diagram showing a peak pattern of unnecessary components obtained based on FIG. This unnecessary component peak pattern shows the absorption characteristics due to unnecessary components superimposed on the background spectrum.

As can be seen from FIG. 6, the unnecessary component peak pattern obtained here represents the transmittance of the unnecessary component at each wave number. Therefore, since the value is a ratio, in order to remove the peak derived from the unnecessary component superimposed on the sample spectrum using this unnecessary component peak pattern, it is necessary to match the signal to the value of the sample spectrum. is there. Therefore, the expansion / contraction rate is considered as a ratio for adjusting the unnecessary component peak pattern to the magnitude of energy at each wave number of the sample spectrum, and the expansion / contraction rate F is calculated by the following equation (2) (step S4).
F = Log (PW / BS) / Log (PP) (2)
Here, PW is a sample spectrum and BS is a reference spectrum. Here, the outer contour of the above-described background spectrum is defined as a reference spectrum. In addition, the said expansion / contraction rate is good to set it as the median value (median value) of the individual expansion / contraction rate in each peak which appears in the spectrum area | region containing the peak corresponding to an unnecessary component. Although not described in detail here, a known method such as that described in Patent Document 1 can be used as a method for calculating the expansion / contraction rate.

  The unnecessary component peak pattern and the expansion / contraction rate are obtained as described above. This unnecessary component peak pattern is calculated based on the background spectrum. If there is no influence of atmospheric fluctuation, the spectrum (sample spectrum) obtained by sample measurement performed at a different time from the background measurement. For the above, it is possible to correct and eliminate the influence of unnecessary components by using the above-mentioned unnecessary component peak pattern and expansion / contraction rate. However, if there are fluctuations due to atmospheric fluctuations between the sample measurement and the background measurement, the waveform shape of the unnecessary component peak pattern is slightly different from that calculated above during sample measurement. There is a fear. Therefore, here, the waveform shape of the unnecessary component peak pattern is corrected in consideration of fluctuations due to atmospheric fluctuations.

First, the following equation (3) is considered as an equation for removing and correcting the influence of unnecessary components included in both the sample spectrum and the background spectrum.
SP ′ = SP / [PP / (1 + Δ)] ^F (3)
Here, SP is sample spectrum data or background spectrum data, SP ′ is spectrum data after correction, and Δ is a constant for removing the influence assuming atmospheric fluctuations. When the atmospheric fluctuation is not taken into account or when there is no influence of the atmospheric fluctuation, the constant Δ can be set to zero. In other words, the estimation of the degree of fluctuation of the influence of atmospheric fluctuation is replaced here with the work of calculating the constant Δ.

Therefore, the constant Δ is determined as follows. Now, if a certain value is temporarily set as the constant Δ, the sample spectrum and the background spectrum after the unnecessary component removal correction can be obtained by the equation (3). Here, the transmittance error ΔTr is defined by the following equation (4).
ΔTr = (SP ′ / BS) −1 (4)
That is, the transmittance error ΔTr is the amount of change from 100% transmittance of the corrected sample spectrum and background spectrum with respect to the reference spectrum BS. As described later, when determining the transmittance and absorbance of the final sample, it is necessary to perform background correction by dividing the corrected sample spectrum by the reference spectrum. In order to obtain the transmittance and absorbance, it is desirable that the influence of unnecessary components be removed equally in both the corrected sample spectrum and the corrected background spectrum. Therefore, the sample spectrum after correction according to the equation (3) when a certain constant Δ is set is PW ′, and the background spectrum after correction according to the equation (3) when the same constant Δ is set is BG ′. If ΔTr obtained by the equation (4) is ΔTr1 and ΔTr2, respectively, a constant Δ is searched so that the sum of squares of ΔTr1 and ΔTr2 is minimized (step S5). That is, the constant Δ is obtained by the least square method.

An example of a specific procedure will be described in detail with reference to FIG. When the change range of the constant Δ is, for example, +0.05 to −0.05 (or may be +0.01 to −0.01 or the like), first, the temporary value of the constant Δ is −0, which is the lower limit value of the change range. .05 (step S11). Next, using this constant Δ and the expansion / contraction rate F, correction for removing the influence of unnecessary components contained in the sample spectrum is performed by equation (3) (step S12). Next, the temporary value of the same constant Δ is used, and the expansion / contraction rate F is set to 1, and the influence of unnecessary components included in the background spectrum is removed and corrected by the equation (3) (step S13). Here, the reason why the expansion / contraction rate F is set to 1 is that the unnecessary component peak pattern is created based on the background spectrum data, and therefore there is no need for expansion / contraction for matching the signal level.

Then, the outer contour of the background spectrum is set as the reference spectrum BS, and the transmittance error ΔTr1 of the corrected sample spectrum is calculated using equation (4) (step S14). Similarly, the outer outline of the background spectrum is set as the reference spectrum BS, and the corrected transmittance error ΔTr2 of the background spectrum is calculated using equation (4) (step S15). Then, the sum of squares of the transmittance errors ΔTr1 and ΔTr2 is calculated (step S16), and it is determined whether or not this sum of squares is smaller than the sum of squares obtained so far (step S17). However, when the sum of squares is calculated for the first time, the sum of squares obtained so far does not exist, so this step S17 is passed and the process proceeds to step S18, and the constant Δ at that time is stored in the memory.

Next, it is determined whether or not the provisional value of the constant Δ at that time is the upper limit value of the change range (step S19). If not, the provisional value of the constant Δ is determined by a predetermined step, for example, 0.001. Increase it (step S20) and return to step S12. Therefore, by executing the processing of steps S12 to S16 for the temporary value of the constant Δ slightly increased, a square sum of the transmittance error is newly obtained. In step S17, it is determined whether or not the square sum of the transmittance error is smaller than the square sum obtained so far. If the newly obtained square sum is smaller than the previous value, In step S18, the temporary value of the constant Δ already stored in the memory is updated to the temporary value at that time. On the other hand, if the newly obtained sum of squares is not smaller than the previous value, step S18 is skipped, and the temporary value of the constant Δ is not updated.

In this way, the processing of steps S12 to S18 is repeated each time the predetermined step increases until the provisional value of the constant Δ reaches an upper limit value, for example, 0.05. As a result, a temporary value of the constant Δ that gives the double sum of is left in the memory. Therefore, this is determined as the value of the constant Δ (step S21).

The constant Δ determined as described above is the most appropriate value when it is attempted to remove the influence of fluctuation due to atmospheric fluctuations in both the background spectrum and the sample spectrum. Therefore, the waveform of the unnecessary component peak pattern is corrected using the determined constant Δ (step S6). That is, this is to calculate PP / (1 + Δ) in the equation (3). With this modification, the shape of the peak pattern shown in FIG. 6 slightly changes as shown in FIG.

Subsequently, using the unnecessary component peak pattern corrected as described above and the expansion / contraction rate calculated in step S4, correction for removing the influence of unnecessary components superimposed on the sample spectrum data and the background spectrum data is performed. Execute (Step S7). That is, the calculation according to the equation (3) is executed using the determined constant Δ. Thereby, as shown in FIG. 8, the peak due to absorption of water vapor, carbon dioxide, etc. in the atmosphere disappears in each spectrum.

  Thereafter, the absorption spectrum calculation unit 63 that has received the corrected sample spectrum data and background spectrum data calculates a transmittance spectrum or an absorbance spectrum that is purely derived from the absorption of the sample, and the result is displayed on the screen of the display unit, for example. (Step S8). FIG. 9 is a diagram showing a comparison between the transmittance obtained from the spectrum result after correction as described above and the transmittance directly calculated from the spectrum without correction.

  As described above, the FTIR of the present embodiment automatically removes the influence of absorption of water vapor and carbon dioxide in the atmosphere, and also takes into account the fluctuation factors due to atmospheric fluctuations, and thus a sample with higher accuracy than before. The spectrum of can be calculated.

  In addition, although the said Example is FTIR, it is natural that this invention can be applied to spectrophotometers other than that other than infrared spectrophotometers, for example, ultraviolet visible spectrophotometers other than infrared.

The schematic block diagram of FTIR which is one Example of the spectrophotometer which concerns on this invention. The flowchart which shows the process sequence centering on signal correction. The flowchart which shows the specific process sequence of step S5 in FIG. The figure which shows an example of a sample spectrum and a background spectrum. FIG. 4 is a diagram showing the background spectrum shown in FIG. 3 and the outer contour obtained therefrom. The figure which shows the unnecessary component peak pattern computed from the background spectrum and outer contour shown in FIG. The figure which shows the unnecessary component peak pattern corrected in consideration of the fluctuation | variation by atmospheric fluctuation. The figure which shows the spectrum after carrying out removal correction | amendment of the influence on absorption of water vapor | steam, carbon dioxide, etc. in air | atmosphere using the unnecessary component peak pattern after correction | amendment. The figure which shows the comparison of the transmittance | permeability calculated | required directly from the transmittance | permeability calculated | required from the result after removal correction | amendment, and the spectrum which is not correct | amended.

DESCRIPTION OF SYMBOLS 1 ... Light source 2 ... Interferometer 3 ... Measurement light 4 ... Sample 5 ... Detector 6 ... Data processing part 61 ... Measurement spectrum data acquisition part 62 ... Unnecessary component removal correction part 63 ... Absorption spectrum calculation part

Claims (4)

a) Spectroscopic measurement means for acquiring background spectral data in a state where there is no sample in the measurement optical path and acquiring sample spectral data in a state where there is a sample in the measurement optical path;
b) an unnecessary component spectrum calculating means for calculating an unnecessary component spectrum reflecting absorption by an unnecessary component in the atmosphere superimposed on the spectrum from the background spectrum data;
c) an expansion / contraction rate calculating means for calculating an expansion / contraction rate for adjusting the unnecessary component spectrum to the level of the sample spectrum data;
The constants for removing the influence of the spectral variations due to fluctuations of the d) air, the expansion ratio and the reference to the unwanted component spectra overlap the sample spectral data and said background spectrum data after assuming the constants Constant determining means for determining based on the validity of the results when correction is performed so as to remove the influence of absorption by unnecessary components in the atmosphere,
e) After correcting the unnecessary component spectrum with the constant determined by the constant determining means, the unnecessary component spectrum in the atmosphere is superposed on the sample spectrum data using the corrected unnecessary component spectrum and the expansion / contraction rate. Unnecessary component removal correction means for performing correction correction of the influence of absorption by the component;
A spectrophotometer comprising: When the constant determination means performs correction so as to remove the influence of absorption by unnecessary components in the atmosphere superimposed on the sample spectrum data and the background spectrum data, both of the corrected spectra are obtained. 2. The spectrophotometer according to claim 1, wherein the constant is determined so as to minimize a square sum of a difference in transmittance with respect to a reference spectrum. A measurement signal correction method for removing the influence of absorption due to unnecessary components in the atmosphere from a measurement signal obtained by spectroscopic measurement,
a) an unnecessary component spectrum calculation step for calculating an unnecessary component spectrum reflecting absorption by an unnecessary component in the atmosphere superimposed on the spectrum from background spectrum data acquired in a state where there is no sample in the measurement optical path;
b) an expansion / contraction rate calculating step for calculating an expansion / contraction rate for adjusting the unnecessary component spectrum to a level of sample spectrum data acquired in a state where the sample is present in the measurement optical path;
The constants for removing the influence of the variations due to fluctuations of c) air, superimposed on the sample spectral data and said background spectrum data using said unwanted component spectrum and the stretch ratio in terms of assuming the constants A constant determination step that is determined based on the validity of the result when correction is performed so as to remove the influence of absorption by unnecessary components in the atmosphere,
d) After correcting the unnecessary component spectrum with the constant determined in the constant determining step, the unnecessary component spectrum in the atmosphere superimposed on the sample spectrum data using the corrected unnecessary component spectrum and the expansion / contraction rate is used. An unnecessary component removal correction step for performing correction correction of the influence of absorption by the component,
A method for correcting a measurement signal of a spectrophotometer, comprising: In the constant determination step, when correction is performed so as to remove the influence of absorption by unnecessary components in the atmosphere superimposed on the sample spectrum data and the background spectrum data, both of the corrected spectra are obtained. 4. The method for correcting a measurement signal of a spectrophotometer according to claim 3, wherein the constant is determined so as to minimize a square sum of a difference in transmittance with respect to a reference spectrum.

Images