JPH0376407B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a double beam spectrophotometer, which is suitable for performing analysis in the infrared region, sometimes using a thermal infrared detector such as a vacuum thermocouple detector. It is related to.

As is well known, a double-beam spectrophotometer uses light to alternately enter the sample to be measured and the standard sample (or an empty cell), and the sample light that has passed through the sample to be measured and the reference light that has passed through the standard sample. The transmittance of the sample is determined by measuring the intensity of the light (also called standard light) and comparing the sample light intensity with the reference light intensity. The signal processing method in such a double-beam spectrophotometer, that is, the typical method for automatically outputting sample transmittance, is the optical method widely used in infrared region spectrophotometers. There are the target zero level method, the recently developed electrical direct ratio method, and the automatic gain adjustment method.

In the optical zeroing method, a switch alternately switches the sample light and the reference light attenuated by a mechanical attenuator, and the difference between the reference light intensity (I ₀ ) and the sample light intensity (I) is An alternating current signal with an amplitude proportional to (I ₀ - I) is extracted, this signal is treated as an error signal in a closed loop system _, and the machine in the reference light is By adjusting the dimmer in this way, the amount of light attenuation itself (in other words, the amount of movement of the dimmer) is proportional to the transmittance. By recording the movement of the optical device, changes in the transmittance of the sample will be recorded.

The optical zero level method described above is the best in terms of signal utilization and stability, but on the other hand,
There are various drawbacks as follows. In other words, first
In this case, it is difficult to reliably and stably obtain a highly accurate spectrophotometer because the accuracy of the transmittance is greatly influenced by the mechanical accuracy of the dimmer itself and the accuracy of its drive system. In addition, related to this, it is caused by uneven rotation of the drive servo motor for moving the wedge-shaped diaphragm normally used in dimmers, and errors in the linearity of the potentiometer that detects the position of the wedge-shaped diaphragm. However, there is a drawback that the amount of movement of the diaphragm and the amount of light attenuation are not necessarily proportional, and the accuracy of transmittance measurement often becomes low. Furthermore, since the optical system is included within the servo loop, the signal becomes more complex.
The configuration of the device is also complicated and expensive, and there are drawbacks such as low responsiveness. In addition, when absorption by the sample is large, the sample light intensity approaches zero, and the reference light intensity also approaches zero, resulting in a decrease in loop gain, which reduces reliability, and also reduces the dimmer itself. When the amount of light attenuation is extremely large (that is, when the absorption of the sample is extremely large), the accuracy decreases, and when these factors are combined, there is a problem that the measurement accuracy decreases significantly when the absorption is large. Furthermore, when the sample light intensity is zero, the reference light intensity is also zero, so the loop gain becomes zero and the device stops operating. Therefore, it is possible to perform equipment calibration, various adjustments, inspections, etc. using the sample light intensity as a reference of absolute zero. There is a flaw that cannot be done.

On the other hand, the automatic gain adjustment method does not use a dimmer, but uses a sector mirror or other optical path switching means to time-share the sample light, reference light, and dark light (dark light) in which both are cut off, so that the reference light intensity is always constant. This method automatically adjusts the gain of the signal processing system, such as the gain of the detector and the gain of the amplifier, so that the output of the amplifier corresponding to the sample light is directly the ratio of the sample light intensity to the reference light intensity, That is, since it corresponds to the transmittance, the transmittance of the sample light can be directly obtained by sample-holding and outputting the output of the amplifier corresponding to the sample light. In this method, when a detector whose gain can be controlled, such as a photomultiplier tube, is used and the gain of the detector is controlled by feedback to the detector, the entire electrical signal system is included in this loop. Therefore, the linearity and stability of detectors and amplifiers do not affect the readings, allowing for highly accurate measurements.In addition, it does not use a dimmer or mechanical servo system, and absolute zero can be taken. , almost all the drawbacks of the optical nulling method are eliminated. However, infrared spectrophotometers that perform analysis in the infrared region have no choice but to use a thermal detector such as a thermocouple, and since it is difficult to adjust the sensitivity of this thermal detector, an automatic gain control method is used. Even if a thermal detector were configured to control the gain of the amplifier, the response would be poor compared to the photomultiplier tube detectors used in visible UV spectrophotometers. Because the speed was extremely slow and the time constant was extremely large, the output waveform of the detector did not correspond to the waveform of the change in the light that passed through the optical path switching means, making it difficult to put it to practical use as an automatic gain adjustment method. .

Further, the electrical direct ratio method does not use a dimmer like the automatic gain adjustment method, and uses a signal obtained from a photodetector that includes a component corresponding to the sample light intensity and a component corresponding to the reference light intensity. This method involves amplifying the two components in a mixed state using the same amplifier, then electrically separating the two components, and electrically calculating the ratio of the two components. This method is roughly divided into two types, the frequency component detection method and the phase detection method, depending on the method of separating the sample light intensity component and the reference light intensity component from the output of the amplifier, but both are thermoelectric Signal component separation is possible even when using a pair of thermal detectors, and therefore it can be used in a spectrophotometer in the infrared region.

As a conventional frequency component detection method in the electrical direct ratio method as mentioned above, there is
There is a spectrophotometer described in Publication No. 10790.
This spectrophotometer uses a sector mirror that alternately switches the sample light and standard light on and off at a predetermined frequency f as an optical path switching means for switching and blocking the optical paths of the sample light and reference light, and The light flux is changed to a frequency 2f that is twice the intermittent frequency f of the sector mirror.
According to the publication, the frequency f component in the output signal of the detector is the difference (I ₀ - I) between the reference light I and the sample light intensity I ₀ . Correspondingly, since the component of frequency 2f corresponds to the sum (I ₀ + I) of the reference light intensity I ₀ and sample light intensity I, the output proportional to I ₀ is obtained by adding the frequency components of f and 2f. is obtained, and also f
It is said that by taking the difference between the frequency components of and 2f, an output proportional to I can be obtained, and therefore, by calculating the ratio of these, the value of I/I ₀ can be easily determined. However, in this method, even when the sample light is zero in the signal processing path, the frequency f
Both the component signal of There is a serious drawback that inspections etc. cannot be carried out easily and accurately.

On the other hand, as a conventional phase detection system in the electrical direct ratio system, there is, for example, an infrared spectrophotometer described in Japanese Patent Application Laid-Open No. 57-52832. This spectrophotometer extracts sample light and reference light with a phase difference of 90° using a sector mirror.
As a result, an output is obtained from the detector in which the signal of the sample light intensity component and the signal of the reference light intensity component are mixed with a 90° phase difference, and this output is periodically rectified to correspond to the sample light intensity. This method obtains a DC signal corresponding to the reference light intensity and a DC signal corresponding to the reference light intensity, and calculates the ratio between the two. However, in general, in spectroscopic analysis, when wavelength scanning is performed, the symmetry of the output waveform of the detector during one cycle may be lost in the region where the absorption of the sample is large, and the absorption due to water or CO ₂ in the air may be lost. This may affect both the light and the sample light, which may affect the output waveform of the detector during one cycle.In such cases, in the phase detection method, the collapse of the output waveform has a large effect on the phase, As a result, there is a drawback that the measurement error becomes large in the end. To explain this in more detail, when the absorption is large, the change in absorbance that accompanies wavelength scanning usually becomes rapid, which causes a slope in the sample light intensity during one cycle. Since the time constant of a detector, especially a thermal infrared detector such as a thermocouple, is large, the output signal is considered to be the effect of integrating the intensity waveform of the incident light. If the light intensity has a slope, the center of the output waveform will shift from the center of the input sample light intensity waveform, and as a result, the phase difference between the output sample light intensity component and the reference light intensity component will shift from 90°, causing an error. It happens. In addition, absorption by water and _CO2 in the air affects the intensity waveform of the light incident on the detector itself, disrupting the output signal of the detector and causing a phase shift as described above, resulting in measurement errors. .

Furthermore, in the case of the phase detection method, if part of the light beam is cut off due to dust adhering to the slit in the optical path of the reference beam or sample beam, or due to the influence of obstacles near the optical path, or if there is a deviation in the mechanical arrangement of the slit. In such cases, the rise or fall of the intensity waveform of the light incident on the detector deviates from the originally set phase, and as a result, the output waveform of the detector deviates, causing a measurement error as described above. Furthermore, in the phase detection method, when the sample light is zero, the corresponding DC signal of the sample light intensity separated by synchronous rectification is originally zero, so calibration and inspection can be performed using absolute zero as a reference. If the phase is shifted due to the above-mentioned causes, the DC signal corresponding to the sample light intensity will deviate slightly from zero, resulting in a problem that the accuracy of calibration etc. will be reduced.

As described above, each signal processing method in conventional spectrophotometers has its own advantages and disadvantages, and the reality is that none of them has been satisfactory, especially when performing spectroscopic analysis in the infrared region.

This invention was made in view of the above circumstances, and it eliminates the drawbacks of the conventional methods, and is particularly suitable for performing spectroscopic analysis in the infrared region, has high precision, has little error, and is based on absolute zero. The purpose of this invention is to provide a spectrophotometer that can be calibrated, inspected, etc.

In other words, the spectrophotometer of the present invention was developed as a result of various studies to improve the frequency component detection method in particular of the conventional electrical direct ratio method and to make it possible to use absolute zero as a reference. The extraction order and combination of light and dark are described in the above-mentioned Japanese Patent Application Laid-Open No.
We have discovered that absolute zero can be used as a reference by using a unique order and combination different from the order of the conventional method described in Publication No. 10790, and by adopting a corresponding frequency component detection method, and have developed the present invention. I arrived at the eggplant.

Specifically, the spectrophotometer of the present invention includes an optical path switching means provided in an optical path to periodically interrupt and switch sample light and reference light, and light that passes through the optical path switching means and is spectrally separated. a photodetector for detecting;
A double beam spectrophotometer comprising a signal processing means for processing the output signal of the photodetector to obtain a ratio between the sample light intensity and the reference light intensity,
The photometer switching means is in a dark light state in which both the sample light and the reference light are cut off during one cycle;
It is configured to switch in order into four states: a first reference light selection state in which only the reference light passes through, a sample light selection state in which only the sample light passes, and a second reference light selection state in which only the reference light passes again. and a signal component separation circuit in which the signal processing means separates and extracts a fundamental frequency component corresponding to one period of the optical path switching means and a frequency component twice the fundamental frequency from the output signal of the photodetector. and an arithmetic circuit that performs arithmetic processing on the signal corresponding to the fundamental frequency component of the signal output from the signal separation/extraction circuit and the signal corresponding to the component of twice the frequency. It is something to do.

The spectrophotometer of this invention will be explained in more detail below.

FIG. 1 shows a sector mirror 1 as an example of the optical path switching means used in the spectrophotometer of the present invention, and FIG. 2 shows a sector mirror 1 of the present invention using the sector mirror 1 shown in FIG. This figure shows an example of an optical system of a spectrophotometer.

In Fig. 2, the light from the light source 2 is in the reference side optical path 3.
The light that is guided to the sample side optical path 4, passes through the standard sample 5 or an empty cell in the reference side optical path 3, passes through the sample 6 to be measured in the sample side path 4, and further passes through the standard sample 5, that is, the reference After the light and the light passing through the measurement sample 6, that is, the sample light, are periodically interrupted and switched by the sector mirror 1 as an optical path switching means as described later,
The monochromator 7 separates the light into monochromatic light, which is incident on a photodetector, for example, a thermocouple 8 as a thermal infrared detector.

As shown in FIG. 1, the sector mirror 1 has four regions P arranged at 90° intervals with respect to the rotation center axis O.
It is divided into 1, P2, P3, and P4. The first region P1 is a region for creating a dark light state by cutting out both the reference light and the sample light, and in the illustrated example, it is configured with a non-reflective wall that does not transmit the reference light and reflect the sample light. ing. The second region P2 is a region for setting the first reference light selection state in which only the reference light is extracted, and in the illustrated example, it is a spatial region that transmits the reference light and does not reflect the sample light. The third region P3 is a region for setting a sample light selection state in which only the sample light is extracted, and in the illustrated example, it is a mirror region that does not transmit the reference light and reflects the sample light. The fourth region P4 is a region for setting the second reference light selection state from which only the reference light is extracted again, and in the illustrated example, it is a spatial region like the second region P2.

By rotating the sector mirror 1 as described above, the optical path detector 8 is equipped with a monochromator 7.
The reference light R and the sample light S are incident at the timing shown in FIG. In other words, if the dark light (D) starts at 0°, after the dark light state continues for 90°, the first reference light selection state is entered and the reference light R reaches 90°.
Then, the sample light selection state is entered and the sample light S enters for 90°, and then the second reference light selection state is entered and the reference light R enters again for 90°. In this way, the dark light D, the reference light R, the sample light S and the reference light R are repeated as one cycle (T). In addition, if the reference side optical path 3 and sample side optical path 4 are reversed in FIG. 2, each reference beam area P2 and fourth area P4 are used as mirrors in the sector mirror 1 in FIG. 1, and correspond to the sample beam area. The third region P3 may be a spatial region, and in this case as well, (D, R, S, R) can be obtained at the timing shown in FIG.

FIG. 4 shows an example of the electrical signal processing means 9 of the spectrophotometer of the present invention, that is, an example of the detector 8 and subsequent parts of FIG. 2.

In FIG. 4, the output of a detector 8 such as a thermocouple is passed through a preamplifier 10 to a signal component separation circuit 11.
added to. This signal component separation circuit 11 has a fundamental frequency corresponding to one cycle of the optical path switching means, that is, a fundamental frequency f given by f=1/T, where the repetition cycle of the four states by the sector mirror 1 is T seconds. component (hereinafter referred to as f component)
This is to separate and extract the 2f frequency component (hereinafter referred to as the 2f component) and the 2f frequency component (hereinafter referred to as the 2f component).
Specifically, the signal separation circuit 11 includes an f-component passing filter 12, a 2f-component passing filter 13, a synchronizing signal generator 14 that generates a synchronizing signal synchronized with the operation of the sector mirror 1, and a synchronizing signal generator 14 that generates a synchronizing signal synchronized with the operation of the sector mirror 1. an f-component synchronous rectifier circuit 15 for synchronously rectifying the signals that have passed through each of the filters 12 and 13 to generate DC signals corresponding to the f-component and the 2f-component;
and a 2f component synchronous rectifier circuit 16. The f component signal and the 2f component output signal extracted from such a signal component separation circuit 11, that is, f
The output signals of the component synchronous rectifier circuit 15 and the 2f component synchronous rectifier circuit 16 are applied to an arithmetic circuit 17 for arithmetic processing of these signals. This arithmetic circuit 17
Specifically, it is comprised of an adder circuit 18 that adds the f component signal and the 2f component signal, and a ratio calculation circuit 19 that calculates the ratio between the output signal of the adder circuit 18 and the f component signal.

The operation of the signal processing means 9 as described above will be explained below in correspondence with the operation of the mirror sector 1.

The output waveform f(t) of the preamplifier 7 is schematically shown in FIG. 5. However, it goes without saying that the actual waveform is slower than the waveform shown in FIG. 5 due to the time constant of the detector, etc. It is. In FIG. 5, I ₀ represents the reference light intensity, I represents the sample light intensity, and ω represents the angular velocity.

The Fourier expansion of the waveform f(t) in Fig. 5 is expressed as (0,
2π/ω) are periodically connected and find the fundamental wave A(1) and second harmonic A(2), we get
The fundamental wave A(1), that is, the component A(1) about the fundamental frequency f is given by A(1)=√2I/π (1). Also, the component A(2) for the second frequency A(2), that is, the frequency 2f which is twice the fundamental frequency.
is given by A(2)=(2I ₀ −I)/π (2).

From equation (1), it is clear that the component regarding the fundamental frequency f (the output of the f component synchronous rectifier circuit 15) is proportional to the sample light intensity I. Also, from equation (2), the component for frequency 2f (2f component synchronous rectifier circuit 16
) is twice the reference light intensity I ₀ and the sample light intensity I
It is clear that it is proportional to the difference between Therefore, by adding the output of the f-component synchronous rectifier circuit 15 and the output of the 2f-component synchronous rectifier circuit 16 in FIG.
A signal proportional to I ₀ is obtained. However, in reality, it goes without saying that the difference in coefficients in equations (1) and (2) is adjusted by midway gain adjustment.

As described above, a signal proportional to the reference light intensity I ₀ outputted from the adding circuit 18 and a signal proportional to the sample light intensity I outputted from the f-component synchronous rectification circuit 15 are added to the ratio calculation circuit 19, The ratio calculation circuit 19 outputs a signal corresponding to the ratio I/I ₀ between the sample light intensity I and the reference light intensity I ₀ . That is,
An output corresponding to the transmittance (reciprocal of absorbance) of the sample is obtained from the ratio calculation circuit 19, and the output is recorded by the recorder 20.

In the above explanation, the component of the fundamental frequency f is (1)
As is clear from the equation, it is a function of only the sample light I, and is unrelated to the reference light intensity _I0 . That is, when the sample light intensity I becomes zero, the component A(1) of the fundamental frequency f becomes zero. Therefore, it is possible to calibrate and inspect the apparatus using this as a reference while the sample light intensity I is set to absolute zero by blocking the sample light path. This is significantly different from the method described in Japanese Patent Laid-Open No. 10790/1983, which is a conventional example of a frequency component detection method. In other words, as mentioned above, in the method described in the above publication, even when the sample light intensity is zero, the component of the fundamental frequency f does not become zero, so it is difficult to calibrate or check using the sample light intensity of absolute zero as described above. However, in the case of this invention, the gain adjustment, inspection, etc. of each circuit in the device is performed using absolute zero sample light intensity as a reference. This not only simplifies calibration and inspection, but also significantly improves accuracy.

As is clear from the above explanation, the spectrophotometer of the present invention employs a frequency component detection method, which is one of the electrical direct ratio methods, as its signal processing method, but it is different from the conventional one using the same method. Unlike conventional methods, each circuit within the device can be calibrated, adjusted, and inspected using the absolute zero sample light intensity as a reference, making calibration, adjustment, and inspection easy, as well as highly accurate. The frequency component detection method requires two types of members, a chopper and a sector mirror, as optical path switching means, but the spectrophotometer of this invention does not require these, and the configuration of the optical system and its drive system is simple and inexpensive. You can get the following effect. Also, compared to phase detection type spectrophotometers, spectrophotometers with this output are more effective when absorbing large amounts of water or water in the air.
Even if the detected waveform is distorted due to the influence of CO ₂ , the effect of the waveform distortion on the frequency component is much smaller than the effect on the phase, so there is little chance that errors will become large due to these effects. In addition, the effects of misalignment of the slits in the optical path, dust, nearby obstacles, etc. are much smaller than in the phase detection method, and therefore the measurement accuracy can always be maintained at a high level of accuracy. moreover,
Compared to the optical zeroing method, the accuracy of the entire device is limited by the mechanical precision of the dimmer and its drive system, as it does not control the mechanical light intensity using a dimmer. It has the advantage that it can be reliably obtained with high accuracy without any problems, the device is simple and inexpensive, and the measurement accuracy is high even when the absorption is large.

This invention is superior to the automatic gain control method in that it can use a thermal infrared detector such as a thermocouple, that is, it can be suitably applied to measurements in the infrared region. Of course, it is also possible to use a photoelectron amplifier tube or the like as a device for measurement in the visible and ultraviolet region.

[Brief explanation of drawings]

FIG. 1 is a schematic diagram showing the principle of an example of a sector mirror as an optical path switching means used in the spectrophotometer of the present invention, and FIG. 2 is a spectral diagram of the present invention using the sector mirror shown in FIG. A schematic diagram showing an example of an optical system in a photometer, FIG. 3 is a waveform diagram showing the timing of optical path switching and blocking in the optical system of FIG. 2, and FIG. 4 is an example of a signal processing means in the spectrophotometer of the present invention. FIG. 5 is a schematic waveform diagram for Fourier expansion of the output waveform of the preamplifier in the circuit of FIG. 1...Sector mirror (as optical path switching means), 8...Photodetector, 9...Signal processing means, 11
...Signal component separation circuit, 17...Arithmetic circuit.

Claims (1)

[Scope of Claims] 1. An optical path switching means provided in an optical path to periodically interrupt and switch the sample light and reference light, and a photodetector that detects the light that passes through the optical path switching means and is separated. , and a signal processing means for processing the output signal of the photodetector to obtain a ratio between the sample light intensity and the reference light intensity, the optical path switching means comprising: a dark light state that blocks both the sample light and the reference light during the cycle;
It is configured to switch in order into four states: a first reference light selection state in which only the reference light passes through, a sample light selection state in which only the sample light passes, and a second reference light selection state in which only the reference light passes again. and a signal component separation circuit in which the signal processing means separates and extracts a fundamental frequency component corresponding to one period of the optical path switching means and a frequency component twice the fundamental frequency from the output signal of the photodetector. and an arithmetic circuit that performs arithmetic processing on a signal corresponding to the fundamental frequency component and a signal corresponding to the double frequency component output from the signal component separation circuit. 2. The spectrophotometer according to claim 1, wherein a thermal infrared detector is used as the photodetector. 3. A circuit in which the arithmetic circuit calculates the sum of a signal corresponding to the fundamental frequency component from the signal component separation circuit and a signal corresponding to a frequency component twice that frequency, and a signal of the sum and a signal corresponding to the fundamental frequency component. 2. The spectrophotometer according to claim 1, wherein the spectrophotometer is configured to include a circuit for determining the ratio of . 4. The optical path switching means is constituted by a sector mirror, and the sector mirror is arranged in four states at 90° intervals: the dark light state, the first reference light selection state, the sample light selection state, and the second reference light selection state. The spectrophotometer according to claim 1, wherein the spectrophotometer is divided into four regions corresponding to the four regions in that order.