JP2810803B2 - Time-resolved Fourier transform spectroscopy - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method in which a stimulus is periodically applied to a measurement object which repeatedly shows the same response to a stimulus, and the output of a detector using a rapid scan interferometer is given a predetermined delay time after the stimulus. Time-resolved Fourier transform that measures the spectrum state in the reaction process of the measurement target by acquiring an interferogram of the difference between the predetermined delay time and the normal state by performing sampling, etc., and obtaining a spectrum by performing a Fourier transform. Related to spectrometry.

[0002]

2. Description of the Related Art In the process of periodically applying a stimulus to a sample by means of electricity, a laser or other means and recovering from the stimulus, the need to measure the state of the reaction is, for example, when evaluating the characteristics of a liquid crystal, or in other cases. In various fields. As a measuring method, there is a time-resolved Fourier transform spectrometry using FT-IR (Fourier transform infrared spectrophotometer). In this method, a wide wavenumber range is set to a high SN.
Since it can be measured by ratio, it has been developed and used conventionally,
It can be classified into those using a rapid scan interferometer and those using a step scan interferometer.

[0003] The FT-IR uses a Michelson interferometer comprising a semi-transmissive mirror, a movable mirror and a fixed mirror to obtain an interferogram by moving the movable mirror. There is a condition that the characteristics must be constant during the measurement, and if the characteristics of the sample change, information that is different from the original information will be output when the sample is subjected to Fourier transform. Therefore, in the time-resolved Fourier transform spectrometry, the period of the applied stimulus is
The condition is that the time is longer than the time at which the reaction ends. However, in the case of applying a periodic stimulus, if the stimulus is applied irrespective of the movement of the movable mirror, it is necessary to match the stimulation.
Therefore, conventionally, a stimulus is applied in synchronization with the reference signal of the interferometer.

Under such a circumstance, the applicant has already applied a stimulus periodically to a measurement object which repeatedly shows the same response to the stimulus, and sampled the output of a detector using a rapid scan interferometer. This is a time-resolved spectroscopic measurement method in which an interferogram for a certain delay time from a stimulus is acquired, and a Fourier transform is performed to measure a reaction state by a spectrum. The sample in the excited state is obtained by controlling the delay time from the stimulus by sampling the detector output at a half cycle of the stimulus repetition cycle and extracting a component on the stimulus repetition frequency from the sampled signal. Interferogram of the difference between the interferogram obtained from A time-resolved Fourier transform spectroscopy method characterized by obtaining a difference spectrum between the measured object at the controlled delay time and the measured object in a normal state by Fourier-transforming the interferogram of the difference. It was proposed (Japanese Patent Application No. 1-335748). In this method, the stimulus can be given asynchronously with the reference signal of the interferometer, so that the constraint on the stimulus can be greatly eased. Efficiency can be improved, and a stimulus generator, a variable delay circuit, a gate circuit,
By adding a band pass filter and lock-in amplifier, data sampling circuit and data editing software etc.
It has the merit that it can be implemented without requiring a significant change in the system, and has the merit of being unaffected by changes in the state of the apparatus and the measurement environment, which are the merit of the difference measurement method.

The present applicant has also made it possible to simultaneously measure transient phenomena at different times when a stimulus is applied to a sample in a short time without repeating the same measurement by changing the delay time for the same sample. In order to do this,
No. 35748 is multi-channeled and the same detector output signal using a rapid scan interferometer is sampled separately at a plurality of different delay times, and a component on the stimulus repetition frequency is extracted from each sampled signal. By obtaining the interferogram of the difference for each delay time, to obtain the difference spectrum between the measurement target at the corresponding delay time and the measurement target in the normal state by Fourier transforming the interferogram of each difference, It has also been proposed to simultaneously measure the reaction states of a plurality of measurement objects that exhibit the same response to a stimulus repeatedly at a plurality of delay times (Japanese Patent Application No. 2-259357).

[0006] Further, the present applicant has disclosed the above-mentioned Japanese Patent Application No. Hei.
Instead of controlling and sampling the delay time in the method of 35748, it was also proposed that the light source be a pulsed light source that emits light with a controlled delay time from the stimulus in a half cycle of the stimulus repetition cycle. (Japanese Patent Application No. 2-82128
issue). Also in this case, the present applicant can simultaneously measure transient phenomena at different times when a stimulus is applied to a sample in a short time without repeating the same measurement by changing the delay time for the same sample. In order to make the method described above, the method of Japanese Patent Application No. 2-82128
Light is emitted from the pulse light source at a plurality of different delay times per half of the stimulus repetition period, and detector output signals for each delay time are separately extracted, and the stimulus repetition frequency is calculated from each separately extracted signal. The interferogram of the difference with respect to each delay time is obtained by extracting the component on which the measurement is performed, and the Fourier transform of the interferogram of each difference is performed, so that the measurement object at the corresponding delay time and the measurement object in the normal state are obtained. It has also been proposed to obtain a difference spectrum and simultaneously measure the reaction states of a plurality of measurement targets exhibiting the same response repeatedly to a stimulus at a plurality of delay times (Japanese Patent Application No. 2-259358).

[0007]

However, according to the proposals by the present applicant, in each case, the repetition period of the stimulus is twice the maximum frequency f _max of the interferogram signal by the sampling theorem. It was assumed that it was shorter than the reciprocal. Assuming that the repetition period of the stimulus is τ, the above is τ <１ ／ f _max or f _max <１ ／
This means that τ was assumed.

The present invention has been made in view of such a situation. It is an object of the present invention to provide the above-described time-resolved Fourier transform spectroscopy method proposed by the present applicant by using the time-resolved cycle of the interferogram signal. In the case of a measurement longer than the reciprocal of twice the maximum frequency f _max , that is, to be applicable to a measurement object having a longer response cycle to a stimulus.

[0009]

According to the first time-resolved Fourier transform spectrometry of the present invention which achieves the above object, a stimulus is periodically applied to a measurement object which repeatedly responds to the stimulus, and a rapid scan is performed. Sampling the output of a detector using an interferometer to obtain an interferogram for a certain delay time from the stimulus, and performing Fourier transform to measure the reaction state of the measurement object that shows the same response to the stimulus repeatedly by spectrum. Fourier spectrometry, where f / 2 is the modulation frequency of the interferometer and 1 / τ is the repetition frequency of the stimulus, and m / 2τ <f <(m + 1) / 2
In a time-resolved Fourier spectroscopy method applied when a detector output signal is present only at τ (m is a positive integer), a stimulus is repeatedly given to a measurement object at a cycle longer than twice the reaction cycle, and a detector output is provided. Is sampled by controlling the delay time from the stimulus in a half cycle of the stimulus repetition cycle, and n / 2τ <f <(n +
1) / 2τ (n is n = m + 2k or n = − (m + 1) +
Zero or positive integer represented by 2k. k is an odd number) by taking out frequency components and subjecting the signal to phase correction to take out an interferogram of a difference between an interferogram obtained from a sample in an excited state and an interferogram obtained from a sample in a normal state. The method is characterized by obtaining a difference spectrum between the spectrum of the measurement target at the controlled delay time and the spectrum of the measurement target in the normal state by Fourier transforming the difference interferogram.

[0010] In the second time-resolved Fourier transform spectrometry, a stimulus is periodically given to a measurement object which repeatedly shows the same response to the stimulus, and the output of a detector using a rapid scan interferometer is sampled. A time-resolved Fourier spectroscopy method that obtains an interferogram for a certain delay time from a stimulus by performing a Fourier transform and measures the response state of a measurement object that repeatedly shows the same response to the stimulus by a spectrum. When the frequency is f and the repetition frequency of the stimulus is 1 / τ, m / 2τ <f <
In a time-resolved Fourier spectroscopy method applied when a detector output signal is present only at (m + 1) / 2τ (m is a positive integer), a pulse light source emits a pulse from a stimulus at a half cycle of a stimulus repetition cycle. Light is emitted by controlling the delay time, and n / 2τ <f <(n +
1) / 2τ (n is n = m + 2k or n = − (m + 1) +
Zero or positive integer represented by 2k. k is an odd number) by taking out frequency components and subjecting the signal to phase correction to take out an interferogram of a difference between an interferogram obtained from a sample in an excited state and an interferogram obtained from a sample in a normal state. The method is characterized by obtaining a difference spectrum between the spectrum of the measurement target at the controlled delay time and the spectrum of the measurement target in the normal state by Fourier transforming the difference interferogram.

[0011]

In each of the first method and the second method, the modulation frequency of the interferometer is f, and the repetition frequency of the stimulus is 1 /.
When τ is set, it can be applied to a case where the detector output signal exists only in m / 2τ <f <(m + 1) / 2τ (m is a positive integer). In the case of a measurement longer than the reciprocal of twice the maximum frequency f _max , that is, a time-resolved Fourier transform spectrometry method can be applied to a measurement object whose response cycle to a stimulus is longer than a sample that was conventionally applicable. Become. In addition, for example, when m = 1, it becomes possible only by replacing the band-pass filter with a low-pass filter as compared with the conventional method.

As in the conventional method, the stimulus can be given asynchronously with the reference signal of the interferometer, so that the restriction on the stimulus can be greatly reduced and the apparatus which is an advantage of the difference measurement method. And the advantage of being unaffected by changes in the measurement environment.

[0013]

As described above, in the time-resolved Fourier transform spectrometry method proposed by the present applicant, the modulation frequency f of each time-resolved signal of the spectrometer is determined by the repetition frequency of the stimulus 1 / τ (τ: (The stimulus repetition period) has been dealt with only in a range smaller than half of the period.
On the other hand, according to the present invention, m / 2τ <f <(m + 1) / 2
The case where a signal exists only in the range of τ (m is a positive integer) is handled.

An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 shows a basic configuration of a time-resolved Fourier transform spectrometer for implementing one embodiment of the time-resolved Fourier transform spectrometer according to the present invention when m = 1.
In FIG. 1, 1 is a light source, 2 is an interferometer, and 3 is a sample (in this case, the sample 3 changes its transmittance by stimulation, but may change its reflectance or the like). , 4
Is a detector, 5 is a stimulus generator, 6 is a 1/2 frequency divider, 7 is a timer, 8 is a variable delay circuit, 9 is a preamplifier, 10 is a gate circuit, 11 is a low-pass filter, 12 is a main amplifier, 13 Denotes an AD converter, and 14 denotes a CPU. However,
A band pass filter is used for 11 depending on the value of m.

In FIG. 1, a sample 3 is an object to be measured for a reaction period τ ′ in which a response repeatedly shows the same response to a stimulus, and a timer 7 is provided with a clock signal having a period τ / 2 (≧ reaction period τ ′). (FIG. 3C). 1 /
The ２ frequency divider 6 divides the frequency of the clock signal of the timer 7 by 、, and supplies a signal of the divided cycle τ to the stimulus generator 5. The stimulus generator 5 is generated by the 周 frequency divider 6 asynchronously with a reference signal of the interferometer 2 (a sampling signal for A / D-converting an interferogram into a computer and performing Fourier transform). A stimulus (trigger) having a period τ as shown in FIG. 3D is given to the sample 3 based on the signal. The variable delay circuit 8 has the configuration shown in FIG.
As shown in (F), a trigger which is delayed from the clock signal of the timer 7 by a predetermined time Δτ is generated and the gate circuit 1
0 is controlled. The gate circuit 10, like the AD converter 13, has a gate width sufficiently narrow with respect to the period τ / 2, and passes a signal only during the presence of a trigger, as shown in FIG. The low-pass filter 11 extracts a low-frequency component of the spectrum obtained from the output of the gate circuit 10 (a range of f <1 / 2τ, as described later), and obtains the transmittance and the transmittance of the sample 3 at a constant delay time Δτ from the stimulation. The signal is converted into an analog signal related to a difference spectrum of transmittance in a normal state.

Sample 3 used in the system of the present invention
Is a measurement object whose response repeatedly shows the same response to the stimulus.

Therefore, the light source 1, the interferometer 2, the sample 3,
The output from the detector 4 of the FT-IR / spectrometer unit including the detector 4 and the preamplifier 9 is output from the gate circuit 10,
The signal is input to the main amplifier 12 of the FT-IR / signal processing unit via the low-pass filter 11, and is sampled by the AD converter 13 for Fourier transform. A timer τ / 2 clock signal is output from the timer 7 in two directions, and one of the signals is frequency-divided by で in the 分 frequency divider 6. A stimulus having a period τ is applied to the sample 3 from the generator 5. On the other hand, another cycle τ / 2 clock signal from the timer 7 is input to the variable delay circuit 8 to generate a trigger which is delayed from the clock signal of the timer 7 by a fixed time Δτ. Open 10. The sample 3 is set in the sample chamber of the FT-IR / spectrometer section, and is repeatedly stimulated by a stimulus signal from the stimulus generator 5 at a period τ. Period of this stimulus τ
Is generally longer than twice the reaction period τ ′ from the start of the transient of the sample 3 due to the stimulus to its complete decay, and the time width of the stimulus is It is necessary to be shorter than the decomposition time for measuring the transient. The spectrum at each decomposition time in the normal state of the sample 3 and at the time of a transient phenomenon is the minimum and maximum of the frequency (frequency of the interferogram) modulated by the interferometer by fmin.
, Fmax, the stimulus is repeated at a period τ that satisfies the following condition: τ> m / 2fmin (a) τ <(m + 1) / 2fmax (b) Here, m is a positive integer. That is, this is a case where the spectrum from the sample 3 exists only in the range of m / 2τ <f <(m + 1) / 2τ. FIG. 2 shows an analog interferogram and its spectrum at a delay time Δτ from the stimulus when m = 1. The same applies to the normal state. The modulation frequency f is represented by f = 2vσ, where σ is the wave number of the spectrum (1 / λ when the wavelength is λ), and v is the moving speed of the movable mirror of the interferometer 2, so that the above condition is satisfied. As described above, the emission spectrum region (wavelength range) from the light source 1 may be limited by an optical filter, or the moving speed of the movable mirror may be adjusted to satisfy the above condition. One of the features of this spectrometry is that it is not necessary to synchronize the stimulus repetition frequency 1 / τ with the movement of the movable mirror, as can be seen from the conditions described above.

Now, the principle in the case of performing spectroscopy using an apparatus having such a configuration will be described below. Stimulated sample 3
Is excited with a certain probability, and at the same time begins to decay back to its original state. At this time, the sample 3 absorbs light having a wavenumber of a characteristic band corresponding to the molecular structure in the transient state. The transmittance spectrum (FIG. 3B) of the sample 3 in the transient state due to the single stimulus as shown in FIG. 3A is expressed by T (σ, t) = T ₀ (σ) + T ₁ (σ, t) .. (1), T (σ, t) = T ₀ (σ) + T ₁ (σ, t) * Шτ (t)... (2) is obtained (FIG. 3E). T ₀ (σ) is the transmittance spectrum that is independent of the stimulus, T ₁ (σ, t) is the transmittance spectrum that changes with the stimulus, and Шτ (t) is the time interval τ where the Delak function δ (t) is equally spaced Com function that represents the repetition operation, and *
Is a convolution operation. At this time, the detector 4
The output signal F (x, t) from is given by equation (3). F (x, t) = ∫T (σ, t) B (σ) cos2πx σ dσ = F ₀ (x) + ∫ ｛T ₁ (σ, t) * Шτ (t)｝ B (σ) cos2 πx σ dσ · (3) F ₀ (x) = ∫T ₀ (σ) B (σ) cos2πx σ dσ (3 ′) where F ₀ (x) is the interferometer output of a spectral component unrelated to the stimulus. And the interferometer output for the normal state. t
Is the time associated with the sample stimulation, x is the optical path difference of interferometer 2,
σ represents the wave number of the spectrum, and B (σ) represents the background spectrum when the sample 3 is removed from the sample chamber of the FT-IR / spectrometer section. Note that x has a relationship of x = 2vt 'with the time variable t', but since the excitation of the sample 3 is not synchronized with the movement of the movable mirror of the interferometer 2, t and t 'are in phase. Has no correlation. That is, t and
The phase amount Δt = t′−t between t ′ takes a different value each time the movable mirror scans. For this reason, the second term of the equation (3) indicates that the interferogram shows a different value each time the movable mirror scans.

The output from the detector 4 is supplied to a preamplifier 9
Is input to the gate circuit 10 through the variable delay circuit 8
From the clock signal generated in step (1) with a gate signal delayed by a time Δτ. This cutout timing is shown in FIG.
It is shown in (F). Here, the time width of the gate signal is narrower than the decomposition time to be measured, but for simplicity, the time width is made infinitely small, and the sampling process is performed using Шτ (t) and Шτ (t-τ
/ 2). The former represents sampling in the excited state of the sample 3, and the latter represents sampling in the normal state. At this time, the output signal F ′ ₁ (x, t) in the excited state with the delay time Δτ obtained from the gate circuit 10
(The cutout signal of the interferogram represented by the dotted line in FIG. 3 (G)) is as shown in the equations (4a) and (4b).

F ′ ₁ (x, t) = Шτ (t−Δτ) [F ₀ (x) + ∫ ｛T ₁ (σ, t) * Шτ (t)｝ × B (σ) cos2 πx σ dσ] (4a) = Шτ (t−Δτ) ∫T (σ, Δτ) B (σ) cos2 πxσ dσ · (4b) Also, the output signal F ′ ₂ (x, t) in the normal state obtained from the gate circuit 10 ) (The cutout signal of the interferogram represented by the dashed line in FIG. 3 (G)) is as shown in equation (5).

F ′ ₂ (x, t) = Шτ (t−τ / 2−Δτ) F ₀ (x) (5) Note that the derivation of the equation (4a) into the equation (4b) is as follows. See below.
The integral part of equation (4b) is the time Δ since the sample 3 was stimulated.
Measurement target spectrum T in the transient state of sample 3 after τ
(σ, Δτ) B (σ) is an interferogram (analog interferogram), and the entire equation (4b) is a discrete data (digital interface) in which this analog signal is sampled at Шτ (t−Δτ). Program). That is, equation (4b) indicates that the measurement signal is time-resolved by the delay time Δτ. Similarly, (5)
F ₀ (x) in the equation is an interferogram (analog interferogram) of the spectrum T ₀ (σ) B (σ) to be measured when the sample 3 is in a normal state. The signal is discrete data (digital interferogram) in a form sampled at Шτ (t−τ / 2−Δτ). That is, the equation (5) also shows the delay time τ / 2
+ Δτ indicates that the measurement signal is time-resolved.
In both equations (4b) and (5), the sampling is a variable t
And x are asynchronous, indicating that the sampled position of the interferogram is different each time the moving mirror scans.

In order to know the output from the gate circuit 10 in detail, Шτ (t−Δτ) in the equation (4b) and Ш in the equation (5)
τ (t -τ / 2- Δτ) is the time t 'correlated with the movement of the movable mirror
And the spectrum of the signal is examined.

∫Шτ (t−Δτ) exp (−i2πft ′) dt ′ = (1 / τ) exp ｛−i2π (Δτ + Δt) f｝ ×｝ 1 / τ (f) (6) = ( 1 / τ) [δ (f) + exp ｛-i2π (Δτ + Δt) / τ｝ δ (f-1 / τ) + ・ ・ ・ ・ + exp ｛i2 π (Δτ + Δt) / τ｝ δ ( f + 1 / τ) + ・ ・ ・] ・ (6 ') ∫Шτ (t -τ / 2- Δτ) exp (-i2πft') dt '= (1 / τ) exp ｛-i2π (τ / 2 + Δτ + Δt) f｝ × Ш1 / τ (f)
・ ・ (7) = (1 / τ) [δ (f) + exp ｛-i2 π (τ / 2 + Δτ + Δt) / τ｝ δ (f-1 / τ) + ・ ・ ・ ・ + exp ｛ i2π (τ / 2 + Δτ + Δt) / τ｝ δ (f + 1 / τ) +...] · (7 ′) Equations (6) and (7) are comb functions having a phase term. That is, both the equations (4b) and (5) are Fourier-transformed to give 2π ×
(Δτ + Δt) / τ is obtained, and between the two, the spectra of the odd-order terms (the two terms of the equations (6 ′) and (7 ′)) have opposite phases. You can see that they have a relationship. That is, in the spectrum of the output from the gate circuit 10 in FIG. 3G, the first-order term represents the difference spectrum between the excited state and the normal state. Here, m / 2τ <f <
From the condition of (m + 1) / 2τ, the spectra of the side bands do not overlap. The above is shown in FIG. 4 for m = 1. 3A shows the spectrum of the digital interferogram in the excited state (FIG. 3G), and FIG. 3B shows the spectrum of the digital interferogram in the normal state (FIG. 3G).
2 (G) shows the spectrum, and FIG. 2 (C) shows the spectrum of the output from the gate circuit 10. The numbers written in each spectrum in these figures indicate that the bands are the sidebands of the carrier numbers having the same numbers.

The output from the gate circuit 10 is led to a band-pass filter in the conventional system, but in the case of FIG.
4 (C) through the low-pass filter 11 having the characteristic shown in FIG. 5 (A).
An analog signal (an analog interferogram of the difference) having only the expression (the spectrum involving the two terms of the expression (7 ′)) is output (FIG. 5B). This signal is ∫ T ₁ (σ, Δ
τ) B (σ) cos2 πx σ dσ interferogram modulated at frequency 1 / τ. In the case of FIG. 4, it is also possible to pass only odd-numbered spectra using a band-pass filter. In the general case, n / 2τ <f
<(N + 1) / 2τ (n is n = m + 2k or n = − (m
+1) + 2k or zero or a positive integer. k is an odd number)
Using a filter that can pass through, the spectrum relating to the odd-order terms of the equations (6 ′) and (7 ′) can be extracted. By the way, since this signal contains the phase of the function of Δt, the form of the analog signal differs every time the moving mirror scans the interferometer. For this reason, the integration of the measured data must be performed after performing the phase correction for each data. That is, since this phase is restored to an analog signal output from the low-pass filter 11, a general FT-IR
In the same manner as in the above measurement, the A / D converter 13 performs sampling with the reference signal (FIG. 3 (I)) having the period τ ₀ of the interferometer 2 for Fourier transform, and then performs the phase correction with the CPU 14 and then the interface. The program is integrated as it is, and a Fourier transform process is performed later, or a Fourier transform is performed simultaneously with the phase correction to convert the spectrum to a spectrum, and then the integration is performed. In this case, tau ₀ is the minimum and maximum modulation frequency with the interferogram signal of the difference output from the low-pass filter 11 f _min, when the _{f max, τ 0> m '} / 2f min, τ 0 <(m It is necessary to satisfy the condition of '+1) / 2f _max . Here, m 'is zero or a positive integer.

From the above, for example, when m = 1, the difference spectrum T ₁ ｛1 / (2vτ) -σ, Δτ｝ B ｛1 / (2vτ)-
Since σ 得 is obtained, for example, the background spectrum B ｛1 /
(2vτ) -σ｝, and the transmittance spectrum T ｛1 in the transient state after a time Δτ after the sample 3 is stimulated.
/ (2vτ) -σ, Δτ｝ and the difference spectrum T ₁ ｛1 / (2v) between the transmittance spectrum T ₀ ｛1 / (2vτ) -σ｝ in its normal state
τ) -σ, Δτ｝ can be obtained. Hereinafter, similarly, by adjusting the delay time of the variable delay circuit 8 and changing the delay time Δτ of the trigger signal to the gate circuit 10, a time-series spectrum at a series of delay times can be obtained. In this method, since the wave number of the spectrum obtained as described above is folded or shifted, it is necessary for the CPU 14 to perform a process of restoring the wave number (wave number conversion).

As described above, according to the time-resolved Fourier transform spectrometry method of the present invention, the measurement object whose repetition period of the stimulus is longer than twice the reciprocal of the maximum frequency f _max of the interferogram signal, and It is possible to measure a difference spectrum between a transient state and a normal state with respect to a stimulus having a repetition cycle that is asynchronous with the sampling cycle.
Therefore, the present invention can be applied to those having a long response cycle τ ′ to a stimulus. In addition, since processing is performed in the form of a difference spectrum in this manner, the signal input to the A / D converter 13 is compressed, thereby compensating for the shortage of the dynamic range of the A / D converter 13. 13 can be prevented from deteriorating.

Here, from the above equation (4a), (4b)
The derivation to the expression will be described. F ′ (x, t) = Шτ (t−Δτ) [F ₀ (x) + ∫ ｛T ₁ (σ, t) * Шτ (t)｝ × B (σ) cos2 πx σ dσ] (4a) ( Шτ (t-Δτ), which represents the sampling process of equation 4a), is
Since it is a function independent of the variable σ, it can be incorporated into the integration of the second term. Here, Ш dependent on the time variable t
Only τ (t−Δτ) ｛T ₁ (σ, t) * {τ (t)} is extracted and the equation is transformed. Since the comb function can be expressed by the sum of the delta functions, Шτ (t) = Σδ (t-nτ) ・ ・ ・ (A1) Шτ (t -Δτ) ｛T ₁ (σ, t) * Шτ (t) こ の(A1)
Rewrite using the formula. Here, n is an integer.

Шτ (t-Δτ) ｛T ₁ (σ, t) * Шτ (t) ｛Σ = ｛Σδ (t-Δτ-nτ)｝ [∫T ₁ (σ, t-t ') ｛Σδ (t '- mτ)} dt'] = {Σδ (t- Δτ-nτ)} {ΣT 1 (σ, a t-mτ)} ··· (A2 ) m also an integer. Since the stimulus period τ is longer than the lifetime τ ′ of the transient phenomenon of the sample, T ₁ (σ, t) = 0 when t <0 and t ≧ τ. Therefore, a signal can be obtained only when n = m in the equation (A2). Therefore, equation (A2) can be modified as follows.

｛Σδ (t-Δτ-nτ)｝ ｛ΣT ₁ (σ, t-mτ)｝ = Σ ｛δ (t-Δτ-nτ) T ₁ (σ, t-nτ) Σ ｛= Σ ｛δ (t- Δτ-nτ) T ₁ (σ, Δτ)｝ ... (A3) Since T ₁ (σ, Δτ) is a constant and can be rounded down, = T ₁ (σ, Δτ) Σδ (t- Δτ −nτ) = T ₁ (σ, Δτ) Шτ (t−Δτ) (A4) Therefore, T (σ, t) = T ₀ (σ) + T ₁ (σ, t) ((1)
Using equation (4a), F ′ (x, t) in equation (4a) is obtained (4b)
The formula, F ′ (x, t) = Шτ (t−Δτ) ∫T (σ, Δτ) B (σ) cos2 πxσ dσ (4b)

The above method can be modified in various ways as in the case of Japanese Patent Application No. 1-335748. For example, the gate circuit may be replaced by turning on / off the supply voltage to the detector in synchronization with the trigger signal from the stimulus generator. Further, the present invention can be similarly applied to fluorescence spectroscopy and Raman spectroscopy in which a sample is irradiated with a pulsed laser. In the case of Raman spectroscopy, as shown in FIG. 6, a continuous wave laser 27 for Raman excitation and a sample excitation (for stimulation)
Pulse laser 28 is used. Then, the pulse laser 28 for sample excitation is operated by a 1/2 frequency divided signal of the clock signal of the timer 24, and the clock signal 1/2 of the timer 24 is supplied.
The gate circuit 32 is operated at the timing when τ is delayed by Δτ by the variable delay circuit 26. Thereafter, the same processing as in the embodiment shown in FIG. 1 is performed. In the case of fluorescence spectroscopy, as shown in FIG. 7, the sample is stimulated using a pulse laser 33 in the same manner as the configuration of FIG. The signal taken in by the detector 23 performs the same processing as in the embodiment shown in FIG.

As in the case of Japanese Patent Application No. 2-259357, a plurality of channels including a gate circuit, a low-pass filter, etc. are provided in parallel in the apparatus shown in FIGS. By inputting the same detection signal to the channel and making the delay time Δτ from applying the stimulus to the sample to opening each gate circuit different for each channel, the stimulus can be obtained according to the principle described with reference to FIGS. , The interferograms of the differences at different delay times are simultaneously obtained, and the Fourier transform is performed on the interferograms. The spectra can be measured simultaneously. FIG. 8 shows an example of the configuration of an apparatus for performing such time-resolved Fourier transform spectrometry. The sample 3 is stimulated by a repetitive pulse having a period τ from a stimulus generator 5 that generates a repetitive pulse stimulus as shown in FIG. 3D, and the light from the stimulated sample 3 is converted into a measurement signal by a detector 4. 1, 6 and 7 (the continuous wave laser 27 is not shown). A signal from the detector 4 is supplied to a plurality of gate circuits 101, 10 arranged in parallel.
2, 103... On the other hand, the period τ from the timer 7
/ 2 are input to a plurality of delay circuits 81, 82, 83,... Delay circuits 81, 82,
83 are different delay times Δτ1, Δτ2, Δτ
3... And the delay time Δτ1, Δτ2
, Δτ3... After each gate circuit 101, 102, 103
A trigger signal is sent to ... .. Are output from the gate circuits 101, 102, 103,..., Respectively, by replacing Δτ in FIG. 3G with Δτ1, Δτ2, Δτ3,. These are the delay times Δτ1, Δτ2,
Are digital interferograms obtained by sampling the interferogram of Δτ3 and the interferogram in the normal state. The low-pass filters 111, 112, 113,... Connected to the gate circuits 101, 102, 103,.
As a result, an interferogram of the difference spectrum modulated at the frequency 1 / τ as shown in FIG. 5B is obtained.
Then, the interferogram of each difference is calculated by the main amplifiers 121 and 12 provided for each channel in the same manner as in FIG.
Are amplified by the A / D converters 131 and 13
2, 133..., Sampled at the period τ ₀ of the reference signal of the interferometer, taken in the CPU 14, processed in the same manner as in FIG. 1, and subjected to Fourier transform to obtain the delay time Δτ 1, Δτ 2, Δτ 3 A difference spectrum representing the state of the sample 3 is obtained at the same time. Therefore, FIG.
The measurement time can be reduced as compared with the methods of FIGS.

It should be noted that a method for simultaneously obtaining spectra at a plurality of different delay times is disclosed in Japanese Patent Application No. Hei.
As in the case of 259357, various other modifications and applications are possible, and refer to the specification.

In the above description, a light source that continuously emits light is used as a light source to observe the transient state at the delay time Δτ after the stimulus is applied, and the output of the detector at this delay time and the output of the normal state are output by the gate circuit. Has been described in the form of a comb, and a digital interferogram is output. However, the method of the present invention uses a pulse light source as a light source, and sets a delay time Δτ from a stimulus at a half cycle of a stimulus repetition cycle. The present invention is also applicable to a device that emits light from a pulsed light source and outputs a signal in which a digital interferogram in a transient state and a digital interferogram in a normal state are superimposed with a phase difference of a half cycle from a detector. FIG. 9 shows the configuration of the apparatus in such a case. The pulse light source 16, the interferometer 2, the sample 3,
Detector 4, preamplifier 10, low-pass filter 11, timer 7, variable delay circuit 8, light source power supply 17, frequency divider 6, stimulus generator 5, main amplifier 12, AD converter 1
3. The CPU 14 is connected as shown in the figure, and instead of using the gate circuit to cut out the signal in the transient state after Δτ and the normal state after τ / 2 + Δτ after applying the stimulus to the sample 3, the CPU 14 is configured as shown in FIG. And after applying a stimulus to sample 3, Δ
By exciting the pulse light source 16 after τ and after τ / 2 + Δτ, the same operation as in FIG. 1 is performed.
In this example, a pulse light source is provided to stimulate the sample and emit pulse light delayed by the time Δτ and τ / 2 + Δτ. However, without providing such a pulse light source, as shown in FIG. Next, the pulse laser 28 for sample excitation is operated by a 1/2 frequency-divided signal of the clock signal of the timer 24 using the pulse laser 27 'for Raman excitation and the pulse laser 28 for sample excitation (stimulation). 24
Pulse signal 27 ′ for Raman excitation at the timing of delaying the clock signal τ / 2 by Δτ by the variable delay circuit 26.
Can be operated to perform time-resolved FT-Raman spectroscopy.

When such a pulsed light source is used, as in the case of Japanese Patent Application No. 2-259358, FIG.
In the device as shown in FIG. 10, a plurality of channels including a low-pass filter and the like are provided in parallel, and a signal having a different delay time is distributed to each channel using a distributor, so that the same response to a stimulus is repeatedly measured. It is also possible to simultaneously measure difference spectra at different times in the reaction process of the subject. FIG. 11 shows an example of the configuration of an apparatus for that purpose. In this case, the clock signal generated by the timer 7 is input to a plurality of delay circuits 81, 82, 83,... Arranged in parallel, and delay times Δτ1, Δτ2, Δτ3..., Τ / 2 + Δτ1, τ / 2+
.DELTA..tau.2, .tau. / 2 + .DELTA..tau.3... The trigger signals from the delay circuits 81, 82, 83,... Are synthesized by the trigger signal synthesizer 31, and after the stimulus is applied to the sample 3, the delay times Δτ1, Δτ2, Δτ3,.
1, .tau. / 2 + .DELTA..tau.2, .tau. / 2 + .DELTA..tau.3... A trigger is applied to the light source power source 17 later, and the pulse light source 16
To drive. Therefore, for a repeated stimulus, Δτ1, Δτ2, Δτ3 ..., τ / 2 +
This is the same as sampling only the detection signal when the gate is applied with a delay time of Δτ1, τ / 2 + Δτ2, τ / 2 + Δτ3,.
2, a comb-shaped signal obtained by adding the interferogram of Δτ3... And the digital interferogram obtained by sampling the interferogram in the normal state is obtained. The comb-like signal is input to the distributor 32 via the preamplifier 10, and the delay times Δτ1 and τ / 2 + Δτ1 and Δ
τ2 and τ / 2 + Δτ2, Δτ3 and τ / 2 + Δτ3 ...
Each time, the comb-like signal from the detector 4 is distributed to the low-pass filters 111, 112, 113,. Thereafter, in the same manner as in the case of FIG.
Difference spectra representing the state of the sample 3 at 1, Δτ2, Δτ3,... Are simultaneously obtained. So, in this case,
The measurement time can be reduced as compared with the methods of FIGS. The method of simultaneously obtaining the spectra at a plurality of different delay times is described in Japanese Patent Application No. 2-2593.
As in the case of No. 58, various other modifications and applications are possible, and refer to the specification.

Although several embodiments of the time-resolved Fourier transform spectroscopy of the present invention have been described above, particularly when m = 1, the present invention is not limited to these embodiments, and various modifications may be made. It is possible. For example, a band-pass filter may be used instead of the device configuration described above with the low-pass filter.

[0037]

As described above, in both the first method using the gate circuit and the second method using the pulse light source, the modulation frequency by the interferometer is f and the repetition frequency of the stimulus is 1 / τ. Then, m / 2τ <f <(m + 1)
/ 2τ (m is a positive integer), so that it can be applied when the detector output signal exists only at the maximum frequency f _max of the interferogram signal.
In the case of a measurement longer than the reciprocal of double, that is, for a measurement object whose response cycle to a stimulus is longer than a sample that was conventionally applicable, the time-resolved Fourier spectrometry can be applied. In addition, for example, when m = 1, it becomes possible only by replacing the band-pass filter with a low-pass filter as compared with the conventional method.

As in the case of the conventional method, the stimulus can be given asynchronously with the reference signal of the interferometer, so that the restriction on the stimulus can be greatly eased and the apparatus which is an advantage of the difference measurement method. And the advantage of being unaffected by changes in the measurement environment.

[Brief description of the drawings]

FIG. 1 is a diagram showing a basic configuration of a time-resolved Fourier transform spectrometer for implementing a first time-resolved Fourier transform spectrometer according to the present invention.

FIG. 2 is a diagram for explaining analog interferograms and spectra included therein.

FIG. 3 is a waveform chart for explaining the operation of the time-resolved Fourier transform spectrometry according to the present invention.

FIG. 4 is a diagram for explaining a spectrum of a digital interferogram in an excited state and a normal state and a spectrum included in an output of a gate circuit.

5 is a diagram for explaining a frequency response characteristic of a low-pass filter used in the device of FIG. 1 and an output spectrum thereof.

FIG. 6 is a diagram showing a main part of a configuration of an embodiment in which the first time-resolved Fourier transform spectrometry of the present invention is applied to Raman spectroscopy.

FIG. 7 is a diagram showing a main part of a configuration of an embodiment in which the first time-resolved Fourier transform spectrometry of the present invention is applied to fluorescence spectroscopy.

FIG. 8 is a diagram showing a configuration of an embodiment in a case where the first time-resolved Fourier transform spectrometry of the present invention is performed in multiple channels.

FIG. 9 is a diagram showing a basic configuration of a time-resolved Fourier transform spectrometer for performing a second time-resolved Fourier transform spectrometer according to the present invention.

FIG. 10 is a diagram showing a main part of the configuration of an embodiment in which the second time-resolved Fourier transform spectrometry of the present invention is applied to Raman spectroscopy.

FIG. 11 is a diagram showing a configuration of one embodiment in a case where the second time-resolved Fourier transform spectrometry of the present invention is performed in multiple channels.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Light source 2 ... Interferometer 3 ... Sample 4 ... Detector 5 ... Stimulus generator 6 ... 1/2 frequency divider 7 ... Timer 8 ... Variable delay circuit 9 ... Preamplifier 10 ... Gate circuit 11 ... Low-pass filter 12 ... Main Amplifier 13 AD converter 14 CPU 16 Light source 17 Power supply for light source

Claims (2)

(57) [Claims] 1. A stimulus is periodically given to a measurement object that repeatedly shows the same response to a stimulus, and an output of a detector using a rapid scan interferometer is sampled to obtain an interferogram for a delay time from the stimulus. Is a time-resolved Fourier spectrometry for measuring the reaction state of a measurement object that exhibits the same response to a stimulus by Fourier transform and a spectrum, wherein the modulation frequency by an interferometer is f and the repetition frequency of the stimulus is 1 / τ M / 2
In a time-resolved Fourier spectroscopy method applied when a detector output signal exists only at τ <f <(m + 1) / 2τ (m is a positive integer), the measurement target is repeated at a period longer than twice the reaction period The stimulus is applied, and the detector output is sampled by controlling the delay time from the stimulus at a half cycle of the stimulus repetition cycle, and n / 2τ <f <(n + 1) / 2τ (n Is a zero or positive integer represented by n = m + 2k or n = − (m + 1) + 2k, where k is an odd number), and the signal is subjected to phase correction to obtain a signal from the excited state sample. The interferogram of the difference between the obtained interferogram and the interferogram obtained from the sample in the normal state is taken out, and the interferogram of the difference is subjected to Fourier transform, whereby the control is performed. Time-resolved Fourier transform spectrometry, characterized in that to obtain a difference spectrum of the spectrum to be measured at the spectrum and the normal state of the measurement object in a delay time. 2. A method in which a stimulus is periodically given to a measurement object which repeatedly shows the same response to the stimulus, an output of a detector using a rapid scan interferometer is sampled, and an interferogram for a certain delay time from the stimulus is obtained. Is a time-resolved Fourier spectrometry for measuring the reaction state of a measurement object that exhibits the same response to a stimulus by Fourier transform and a spectrum, wherein the modulation frequency by an interferometer is f and the repetition frequency of the stimulus is 1 / τ M / 2
In a time-resolved Fourier spectroscopy method applied when a detector output signal exists only at τ <f <(m + 1) / 2τ (m is a positive integer), a half of the stimulus repetition period is determined by a pulse light source. And emits light by controlling the delay time from the stimulus, and n / 2τ <f <of the obtained detector output signal.
(N + 1) / 2τ (n is n = m + 2k or n = − (m +
1) Zero or a positive integer represented by + 2k. k is an odd number) by taking out frequency components and subjecting the signal to phase correction to take out an interferogram of a difference between an interferogram obtained from a sample in an excited state and an interferogram obtained from a sample in a normal state. A time-resolved Fourier transform spectrometry method, wherein a difference spectrum between a spectrum of a measurement target at the controlled delay time and a spectrum of a measurement target in a normal state is obtained by Fourier-transforming the difference interferogram.