JPH01259226A - Fourier transform spectroscope - Google Patents

Abstract

PURPOSE:To eliminate stray light due to Fourier transformation by interpolation from a discrete spectrum data sequence of each equal wave number interval after the Fourier transformation to a discrete spectrum data sequence of each wavelength of a desired equal wavelength interval. CONSTITUTION:An interferometer 1 form an interference graphic form on the photodetection surface of a photodiode array detector 2 as a spatial light-shade distribution. The light-shade distribution of this interference graphic pattern is converted by the photodiode array detector 2 into an electric signal, which is inputted to a microcomputer 4 through an A/D converter 3. This signal is processed by a fast Fourier transforming program 5 stored in the microcomputer 4 into the discrete spectrum data sequence of each equal wave number interval. This spectrum data sequence is processed by an equal wavelength interval conversion program 6. This program 6 correct a sensitivity difference of each wavelength after converting lateral axis data intervals into equal wavelengths.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a Fourier transform spectrometer, and particularly to a Fourier transform spectrometer suitable for measuring spectra in the visible range.

[Conventional technology]

Conventionally, regarding Fourier transform spectrometers in the visible range, Applied Optics 23, 269 (1984), pp. 269 to 273 (Appi.

Opt, 23, 269 (1984) pp269-27
3) and Applied Spec-Roscopy 40°6
91 (1986) pp. 691-695 (Ap
pl,, 5pectrosc, 40, 691 (1
986) pp. 691-695).

As shown in FIG. 3, the discrete spectral data string obtained as a result of Fourier transform by a Fourier transform spectrometer is a sample of the spectral distribution of the object to be measured at regular wave number intervals.

On the other hand, conventional spectra in the visible range are generally measured with the horizontal axis as the wavelength, so in order to be compatible with these data, the horizontal axis of the spectrum obtained as a result of Fourier transform must be We need to convert the axis to wavelength. For every equal wave number interval ΔY, that is, SI1=ΔSIXi...(1
) Here, j; 0, 1, 2. A discrete spectral data sequence sampled for each wavenumber shim, which is a sequence of...
a1)), it is conventional to draw a spectral curve with the wavelength λi corresponding to each ν1 on the horizontal axis, a(,,) on the vertical axis, and the horizontal axis as the wavelength as shown in Figure 4. It has been done since then. However, although the spectral curve obtained by the above-mentioned conventional technique appears to be measured with the horizontal axis as the wavelength, the actual spectral curve is Δ λ1 = λ1 + 1 − λω ν++1 ya1 ν1 + Δ乍 乍1 The data string has non-constant wavelength intervals Δλi for each wavelength (wave number) represented by . For this reason, there is a difference in the density of the number of sampling points on the short wavelength side and long wavelength side of the drawn spectral curve, so the expression as a polygonal line that lacks smoothness on the long wavelength side is noticeable, and differential spectra, etc. During data processing, because the data strings are not at equal wavelength intervals, they cannot be treated in the same way as spectrum data strings measured at equal wavelength intervals using a conventional diffraction grating spectrometer, with the horizontal axis being the wavelength.
There was a problem with data compatibility.

Furthermore, in a diffraction grating spectrometer, since the width of the input and output slits is finite, the wavelength of the light that passes through the output slit and comes out is not uniform but has a finite spread. This wavelength width is called a bandpass, but in general, visible-range diffraction grating spectrometers perform wavelength scanning with a fixed slit width, so the bandpass in units of wavelength is approximately constant over the entire measurement wavelength range. On the other hand, in the spectrum obtained as a result of Fourier transform in a Fourier transform spectrometer, the bandpass in units of wave numbers is constant over the entire measurement wavelength range. Now, let Δda be the bandpass in wavenumber units, Δλ be the bandpass when the unit is converted to wavelength, shi be the wavenumber of the light to be considered, and λ be the wavelength. ...(3) is obtained, If Δy2 is ignored as a minute quantity, then the following holds true, and it can be seen that in the entire measurement wave number range, when Δda is constant, Δλ, which is converted into a wavelength, changes for each wavelength λ of the light being considered. Therefore, in spectra where the vertical axis is the energy intensity, such as fluorescence spectra and emission spectra, there are two types of spectra: one measured under conditions where the bandpass is constant using the wavenumber as a unit, and the other when the bandpass is constant using the wavelength as the unit. With the originally measured spectrum, it is not possible to match the intensity on the vertical axis by simply converting the reciprocal relationship between the wave number and wavelength on the horizontal axis, and the waveforms are not equal. For example, according to equation (4), when Δν is constant, λ=3
Δλ differs by a factor of 4 between 50nm and λ=700nm, and if the resolution of the instrument is sufficiently small for the measured spectrum, this will be different between 350nm and 700nm.
This means that there is a sensitivity difference of 4 times in m.

[Problem to be solved by the invention]

The above-mentioned conventional technology does not take these points into consideration, and there is a problem in that the vertical axis intensity information of the measured spectrum is not compatible with the spectrum measured with a diffraction grating spectrometer.

An object of the present invention is to convert a discrete spectral data string at equal wave number intervals obtained as a result of Fourier transform into a discrete spectral data string at equal wavelength intervals, and to convert By providing a means for correcting the sensitivity difference for each wavelength caused by the path changing for each wavelength, we provide a Fourier transform spectrometer that is compatible with the measurement length spectral data string by a conventional diffraction grating spectrometer. It's about doing.

[Means to solve the problem]

The above object is a means of converting a discrete spectral data string at equal wave number intervals obtained as a result of Fourier transform into a discrete spectral data string at each wavelength at each desired equal wavelength interval by interpolation using an interpolation function. I tried to achieve this by configuring the following. In addition, the sensitivity difference for each wavelength due to the bandpass expressed as a wavelength unit changing for each wavelength is expressed as the square of the wavelength. This is achieved by providing a means for correcting by multiplying by an inversely proportional coefficient.

[Effect]

FIG. 2 shows a processing flow of a computer program that achieves the functions of the present invention. In a Fourier transform spectrometer,
Let (a,) be the discrete spectral data string obtained as a result of the Fourier transform, let al be the photometric value in wave number type 1, and let ν1+1>ν. Let the wave number data interval be Δy as shown in the following formula.

A%l:V++1 'l't
. . (5) It is assumed that the following relationship exists. The spectral data string to be obtained at equal wavelength intervals is assumed to be (bz), b is the photometric value at the long wavelength λ, and λ, +1>λ. Further, the desired wavelength range is from λ1 to λN. Let the wavelength data interval be Δλ as shown in the following equation. Δλ is also a constant that does not depend on J, and is set as follows: Δλ=λJ+1−λ4 (7) For example, a 5ine function is used as the interpolation function. The 5ine function is a function in the form of the following formula f (x).

Here, when 5x is an integer other than O, the value of the function is O, and g(X) of the following equation is used as an interpolation function.

First, in the first half of the processing flow shown in FIG. 2, the horizontal axis is transformed by interpolation. For each λ quantity from λl to λN, the corresponding wave number ν is j−17λ1...(10
). Next, Y1≦V J< v ++s
−(11) Find Ya1, and set it to . Next, using the interpolation function g(x), the photometric value a'J at ν is obtained by calculating the sum of the products of the following equation.

Since g(x) oscillates and attenuates as IXI increases, from the value of X at which Ig(x)l becomes sufficiently small, (13
) The value of m in the range (-m, m) for calculating the sum of the products of the equations shall be determined in advance. In this way, a spectrum data string (a'j) for each equal wave number interval is obtained.

Next, from the relationship in equation (4), the spectrum data string (bJ> corrected for the sensitivity difference for each wavelength is expressed as b J '' a ' J /λJ2
...It is obtained as (14).

As a result, a spectral data string at equal wavelength intervals can be obtained, which is compatible with that conventionally measured with a diffraction grating spectrometer.

〔Example〕

The present invention will be explained in detail below using the embodiments shown in FIGS. 1 and 2.

FIG. 1 is a block diagram showing an embodiment of the Fourier transform spectrometer of the present invention. In FIG. 1, an interferometer 1 generates an interferogram as a spatial distribution of brightness and darkness on the light-receiving surfaces of the first to diode array detectors 2. In FIG. The brightness/darkness distribution of this interferogram is converted into an electrical signal by the photodiode array detector 2 and input into the microcomputer 4 through the A/D converter 3. This signal is a fast Fourier transform program 5 built into the microcomputer 4.
is processed, resulting in a discrete spectral data string at equal wave number intervals. This spectral data string is then processed by the equal wavelength interval conversion program 6. This equal wavelength interval program 6 converts the horizontal axis data interval from equal wave number to equal wavelength according to the processing flow diagram of FIG. 2, and then corrects the sensitivity difference for each wavelength. The resulting spectrum data string is displayed on the spectrum display section 7 or stored in the external storage device 8.

According to this example, while taking advantage of the advantages of a Fourier transform spectrometer, such as the absence of stray light and a bright optical system, it is also compatible with spectra with equal wavelength intervals measured with conventional grating spectrometers. This has the effect of obtaining a certain spectrum.

〔Effect of the invention〕

As explained above, according to the present invention, a discrete spectral data string at equal wave number intervals obtained as a result of Fourier transform is converted into a spectral data string at equal wavelength intervals by interpolation, and the wavelength is set as a unit. Since it is possible to correct the sensitivity difference for each wavelength due to the bandpass expressed as changing for each wavelength, it is possible to obtain a spectrum that is compatible with that measured with a conventional diffraction grating spectrometer. It has the effect of being

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the Fourier transform spectrometer of the present invention, FIG. 2 is a flow diagram showing an embodiment of the fast Fourier transform program and equal wavelength interval conversion program of FIG. 1, and FIG. 4 is a diagram showing a spectral data string sampled at equal wave number intervals, and FIG. 4 is a diagram in which the data string of FIG. 3 is plotted with the horizontal axis as the wavelength. 1 Interferometer, 2...Photodiode array detector, 3
・A/D converter, 4. Microcomputer, 5. High-speed Fourier transform program, 6. Equal wavelength interval conversion program, 7... Spectrum display unit, 8. External storage device.

Claims (1)

[Claims] 1. Comprising an interferometer and a detector that converts an interference pattern from the interferometer into an electrical signal, and converts the electrical signal from the detector into a digital signal and inputs it into a microcomputer. In the Fourier transform spectrometer constructed as follows, an interpolation function is used to convert the discrete spectral data string at each equal wave number interval obtained as a result of Fourier transform to the discrete spectral data string at each wavelength at each desired equal wavelength interval. A Fourier transform spectrometer, characterized in that it is provided with means for performing transformation by interpolation. 2. Fourier transform spectroscopy, which is equipped with an interferometer and a detector that converts the interference pattern from the interferometer into an electrical signal, and which converts the electrical signal from the detector into a digital signal and inputs it into a microcomputer. In the instrument, the sensitivity difference for each wavelength due to the bandpass expressed in units of wavelength changing for each wavelength is expressed as the square of the wavelength by the element at each wavelength of the spectrum data at equal wavelength intervals. A Fourier transform spectrometer, characterized in that it is provided with means for correcting by multiplying by an inversely proportional coefficient.