JP5454687B2 - Interferometer and Fourier transform spectrometer - Google Patents

Description

  The present invention relates to a Michelson type interferometer and a Fourier transform spectroscopic analysis apparatus including the interferometer.

  In the Michelson two-beam interferometer used for FTIR (Fourier Transform Infrared Spectroscopy), the infrared light emitted from the light source is split into two directions, a fixed mirror and a moving mirror, by a beam splitter. A configuration is adopted in which the light reflected and returned is combined into one optical path by the beam splitter. When the moving mirror is moved back and forth (in the direction of the optical axis of the incident light), the optical path difference between the two divided beams changes, so the intensity of the combined light changes according to the amount of movement of the moving mirror. Interference light (interferogram). By sampling this interferogram and performing AD conversion and Fourier transform, the spectral distribution of the incident light can be obtained, and the intensity of the interference light for each wave number (1 / wavelength) can be obtained from this spectral distribution. .

  In order to exhibit high performance in such FTIR, it is desirable to maintain the best interference efficiency in the interferometer. For this purpose, it is necessary to keep the angular relationship between the fixed mirror and the movable mirror and the beam splitter, respectively. In other words, the spectral accuracy (resolution) of FTIR depends on the amount of movement of the moving mirror. The larger the amount of movement, the higher the resolution. However, the larger the amount of movement of the moving mirror, the higher the translation of the moving mirror. Thus, a relative inclination is generated between the reflected light from the movable mirror and the reflected light from the fixed mirror (each reflected light is inclined from the optical axis), and the contrast of the interference light is lowered. For this reason, it is necessary to correct the inclination. For convenience, the relative inclination between the reflected light from the moving mirror and the reflected light from the fixed mirror, which is caused by the inclination during movement of the moving mirror, is also referred to as the light inclination between the two optical paths. Called.

  In addition, when high-sensitivity measurement is required, it is necessary to use light having a large light beam diameter (receiving light having a large diameter by a detector). However, if the beam diameter is large, the allowable amount of light tilt between the two optical paths is small, and the contrast of the interference light is lowered even if a slight light tilt occurs between the two optical paths. Therefore, especially when high-sensitivity measurement is required, it is necessary to reliably correct the tilt to realize a low tilt.

  Therefore, in Patent Document 1, for example, by adjusting the tilt angle of one reflection surface based on the output from the sensor that detects the interference light, the relative inclination of each light reflected by the two reflection surfaces is obtained. It is corrected.

US Pat. No. 4,053,231 (refer to FIG. 1 etc.)

  However, in the configuration of Patent Document 1, the reflective surface (moving mirror) that is translationally driven in the optical axis direction of the incident light among the two reflective surfaces is driven by a linear motion actuator. In such a configuration in which the movable mirror is moved by linear motion, in order to increase the amount of movement of the movable mirror and increase the resolution, it is necessary to increase the size of the actuator, which increases the size of the apparatus. In addition, when driving a large actuator, power consumption also increases.

  The present invention has been made to solve the above-mentioned problems, and its purpose is to make the movable mirror large with a small size and low power consumption configuration while being able to correct the inclination of light between two optical paths. Interferometer that can be moved, thereby realizing high-resolution measurement with a low tilt while realizing high resolution with a high stroke of the movable mirror in a compact, low-power configuration, and its interference An object of the present invention is to provide a Fourier transform spectroscopic analysis apparatus including a meter.

  The interferometer of the present invention includes a moving mirror and a fixed mirror, and a beam splitter that separates measurement light and guides it to the moving mirror and the fixed mirror, and combines the light reflected by the moving mirror and the fixed mirror. And an optical system that detects interference light obtained by synthesizing each light reflected by the movable mirror and the fixed mirror by the beam splitter, the interferometer comprising the movable mirror Between the reflected light from the movable mirror and the reflected light from the fixed mirror caused by the tilt of the movable mirror when driven by the movable mirror drive mechanism. An inclination correction unit that performs driving for correcting relative inclination with respect to at least one of the moving mirror and the fixed mirror, and the driving of the moving mirror by the moving mirror driving mechanism is resonance driving. Meanwhile, the tilt correction Drive to at least one of the moving mirror and the fixed mirror according is characterized in that a non-resonant drive.

  According to the present invention, the tilt correction unit performs driving for correcting the tilt of light between the two optical paths with respect to at least one of the movable mirror and the fixed mirror, and the driving is non-resonant driving. Therefore, it is possible to avoid a decrease in the contrast of the (first) interference light due to the inclination. Thereby, the (first) interference light can be measured (detected) with high sensitivity using light having a small allowable light inclination between the two optical paths and a large light beam diameter. Moreover, since the driving of the moving mirror by the moving mirror driving mechanism is resonance driving, it is possible to achieve a high resolution by securing a large amount of movement with a small configuration, and since the moving mirror driving mechanism is small, It is possible to reduce power consumption when the moving mirror is driven. That is, with a small size and low power consumption configuration, high sensitivity measurement with a low tilt can be performed while realizing high resolution with a high stroke of the movable mirror.

It is explanatory drawing which shows typically the structure of the outline of the Fourier-transform spectroscopy analyzer of one Embodiment of this invention. It is a top view which shows the structure of the outline of the 2nd photodetector of the interferometer applied to the said Fourier-transform spectroscopy analyzer. It is explanatory drawing which shows the phase signal output based on the detection result in the said 2nd photodetector. It is a perspective view which shows the structure of the outline of the movable mirror drive mechanism with which the interferometer of the said Fourier-transform spectroscopy analyzer is provided. It is sectional drawing of the said movable mirror drive mechanism. It is explanatory drawing which shows the schematic structure of the drive part of the said movable mirror drive mechanism, and the displacement method of a rigid body and a movable mirror. It is a graph which shows the relationship between the ratio of the length of the flat plate part of the leaf | plate spring part of the said movable mirror drive mechanism, and the length of a piezoelectric element, and the applied voltage to the said drive part. It is a perspective view which shows other structure of the said movable mirror drive mechanism. It is sectional drawing of the movable mirror drive mechanism of FIG. It is sectional drawing which shows other structure of the said movable mirror drive mechanism. It is sectional drawing which shows other structure of the said movable mirror drive mechanism. 12 is a flowchart showing a flow of an operation by voltage control in an initial operation in the movable mirror drive mechanism of FIG. 12 is a flowchart showing an operation flow by voltage control in a steady operation in the movable mirror driving mechanism of FIG. It is a flowchart which shows the rough flow at the time of manufacture of the movable mirror drive mechanism of FIG. (A)-(d) is sectional drawing which shows the manufacturing process of the said movable mirror drive mechanism. It is a perspective view of the board | substrate which spelled the several leaf | plate spring part in the sheet form. It is a perspective view of the support block pinched | interposed with two said board | substrates. It is a perspective view of the joined body which consists of the said board | substrate and the said support block before cut | disconnecting a movable mirror from a support piece. It is a perspective view of the above-mentioned joined body after separating the above-mentioned movable mirror from a support piece. (A)-(f) is sectional drawing which each shows the preparation process of the leaf | plate spring part of the movable mirror drive mechanism seen in the A-A 'arrow cross section of FIG. It is sectional drawing which shows other structure of the said movable mirror drive mechanism. It is sectional drawing which shows other structure of the said movable mirror drive mechanism. It is sectional drawing which shows other structure of the said movable mirror drive mechanism. FIG. 5 is an explanatory diagram showing tilt error amounts in the pitch and roll directions when the movable mirror is moved by the movable mirror drive mechanism of FIG. 4. (A) is explanatory drawing which shows Pitch direction, (b) is explanatory drawing which shows Roll direction. (A) is a side view which shows the schematic structure of the optical path correction apparatus of the said interferometer, (b) is a top view of the fixed mirror supported by the said optical path correction apparatus. It is a block diagram which shows the feedback control performed when carrying out non-resonance drive of the said fixed mirror. (A) is a top view which shows the other structure of the said optical path correction apparatus, (b) is a side view of the said optical path correction apparatus. It is a bottom view from the piezoelectric element side of the rotation member of the said optical path correction apparatus. It is a side view of the said optical path correction apparatus before and behind the rotation of the said rotation member.

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

[1. Configuration of Fourier transform spectroscopic analyzer]
FIG. 1 is an explanatory diagram schematically showing a schematic configuration of the Fourier transform spectroscopic analyzer of the present embodiment. This device is a device that splits measurement light using the principle of a Michelson interferometer, and includes an interferometer 1, a calculation unit 2, and an output unit 3. The interferometer 1 is a two-optical path branching Michelson interferometer, and details thereof will be described later. The computing unit 2 performs sampling, A / D conversion, and Fourier transform of the signal output from the interferometer 1, and indicates the spectrum of the wavelength included in the measurement light, that is, the light intensity for each wave number (1 / wavelength). Generate a spectrum. The output unit 3 outputs (for example, displays) the spectrum generated by the calculation unit 2. Hereinafter, details of the interferometer 1 will be described.

  The interferometer 1 includes a first optical system 10, a second optical system 20, and a tilt correction unit 100. Hereinafter, it demonstrates in order.

  The first optical system 10 includes a measurement light input unit 11, a reflection collimator 12, a BS (beam splitter) 13, a fixed mirror 14, a movable mirror 15, a reflection collimator 16, and a first photodetector 17. And a drive mechanism 18. Note that the positional relationship between the fixed mirror 14 and the movable mirror 15 with respect to the BS 13 may be reversed. Further, a correction plate for correcting the optical path length corresponding to the thickness of the BS 13 may be provided in the optical path.

  The measurement light input unit 11 is a portion where light (measurement light, near infrared light) emitted from a light source (not shown) and transmitted or reflected through the sample is incident. The reflection collimator 12 includes a reflection surface (collimating optical system) that reflects light from the measurement light input unit 11 and converts the light into parallel light and guides it to the BS 13. The BS 13 separates incident light, that is, light emitted from the measurement light input unit 11 into two lights, which are guided to the fixed mirror 14 and the movable mirror 15 and reflected by the fixed mirror 14 and the movable mirror 15. The combined lights are emitted as first interference light, and are composed of, for example, a half mirror.

  The reflection collimator 16 includes a reflection surface (condensing optical system) that reflects and collects the light synthesized and emitted by the BS 13 and condenses it to the first photodetector 17. The first photodetector 17 receives the first interference light incident from the BS 13 via the reflective collimator 16 and detects an interferogram (interference pattern). The above-described reflective collimators 12 and 16 may be collimator lenses.

  The drive mechanism 18 causes the movable mirror 15 to change the incident light so that the difference between the optical path of the light reflected by the fixed mirror 14 and the optical path of the light reflected by the movable mirror 15 (difference in optical path length) changes. This is a moving mirror drive mechanism that translates (translates) in the optical axis direction. In this embodiment, the drive mechanism is constituted by a parallel leaf spring type drive mechanism, the details of which will be described later.

  In the above configuration, the light (measurement light) emitted from the measurement light input unit 11 is converted into parallel light by the reflection collimator 12 and then separated into two light beams by transmission and reflection at the BS 13. One of the separated light beams is reflected by the movable mirror 15, and the other light beam is reflected by the fixed mirror 14. Each of the separated light beams returns to the original optical path and is superposed on the BS 13 to become the first interference light. At this time, the movable mirror 15 is continuously moved by the drive mechanism 18, but when the optical path length difference from the BS 13 to each mirror (movable mirror 15, fixed mirror 14) is an integral multiple of the wavelength, the superimposed light The strength of is the maximum. On the other hand, when there is a difference between the two optical path lengths due to the movement of the movable mirror 15, the intensity of the superimposed light changes. The first interference light is collected by the reflection collimator 16 and enters the first photodetector 17, where it is detected as an interferogram.

  The computing unit 2 samples a detection signal (interferogram) from the first photodetector 17 and performs A / D conversion and Fourier transform to generate a spectrum indicating the light intensity for each wave number. The above spectrum is output (for example, displayed) by the output unit 3, and based on this spectrum, the characteristics (material, structure, component amount, etc.) of the sample can be analyzed.

  Next, the second optical system 20 and the inclination correction unit 100 will be described. The second optical system 20 shares a part of the configuration with the first optical system 10 described above. The reflective collimator 12, the BS 13, the fixed mirror 14, the movable mirror 15, and the reflective collimator 16 described above. In addition, a reference light source 21, an optical path synthesis mirror 22, an optical path separation mirror 23, and a second photodetector 24 are included.

  The reference light source 21 is a light source for detecting the position of the movable mirror 15 and generating a timing signal for sampling in the calculation unit 2, and is a semiconductor laser that emits red light with a wavelength of 660 nm as reference light, for example. It is configured. That is, the semiconductor laser constituting the reference light source 21 emits laser light having a shorter wavelength than the shortest wavelength of light (near infrared light) emitted from the measurement light input unit 11. By using a semiconductor laser as the reference light source 21, the interferometer 1 can be reduced in size compared to a configuration using a large He—Ne laser.

  The optical path combining mirror 22 is an optical axis combining beam combiner that transmits the light from the measurement light input unit 11 and reflects the light from the reference light source 21, thereby combining the optical paths of these lights into the same optical path. The optical path separation mirror 23 transmits light that is emitted from the measurement light input unit 11 and incident via the BS 13 and the fixed mirror 14 (or the BS 13 and the movable mirror 15), and is emitted from the reference light source 21 and is transmitted to the BS 13 and the fixed mirror 14. It is a beam splitter that separates the optical path of these lights by reflecting the light incident through (or the BS 13 and the movable mirror 15).

  The second photodetector 24 is emitted from the reference light source 21, enters the optical path separation mirror 23 via the BS 13 and the fixed mirror 14 (or the BS 13 and the movable mirror 15), and is reflected there (second interference). For example, a quadrant sensor (SPD; Silicon Photo Diode).

  The tilt correction unit 100 performs driving for correcting the relative tilt between the reflected light from the movable mirror 15 and the reflected light from the fixed mirror 14 caused by the tilt of the movable mirror 15 during driving by the drive mechanism 18. This is an inclination correction unit for the fixed mirror 14. Hereinafter, the relative inclination between the reflected light from the movable mirror 15 and the reflected light from the fixed mirror 14 caused by the inclination during the movement of the movable mirror 15 is the inclination of the light between the two optical paths. (Or tilt error).

  By the way, in this embodiment, two light paths, that is, an optical path when one light separated by the BS 13 is reflected by the movable mirror 15 and enters the BS 13 again, and the other light separated by the BS 13 are fixed mirrors. The first optical system 10 and the second optical system 20 share (coaxial) the optical path when the light is reflected by 14 and incident on the BS 13 again. In this configuration, (1) the light traveling in the order of the measurement light input unit 11, BS13, the movable mirror 15, BS13, and the first photodetector 17, the measurement light input unit 11, BS13, the fixed mirror 14, BS13, (2) Reference light source 21, BS 13, movable mirror 15, BS 13, second light, relative inclination (hereinafter also referred to as first inclination) between the light traveling in the order of one light detector 17. The relative inclination between the light traveling in the order of the detector 24 and the light traveling in the order of the reference light source 21, BS 13, fixed mirror 14, BS 13, and second light detector 24 (hereinafter also referred to as the second inclination). The same). Therefore, the inclination correction unit 100 is equivalent to detecting the first inclination by detecting the second inclination based on the light reception signal of the second interference light from the second photodetector 24. Thus, the first inclination can be corrected based on the detection result.

  Specifically, such an inclination correction unit 100 includes a signal processing unit 101 and an optical path correction device 102. The signal processing unit 101 detects the inclination of the light between the two optical paths based on the intensity of the second interference light detected by the second photodetector 24. For example, as shown in FIG. 2, the four light receiving areas of the second photodetector 24 are E1 to E4 counterclockwise, and the light spot D of the second interference light is located at the center of the entire light receiving area. It shall be. When the sum of the light intensities detected in the light receiving areas E1 and E2 is A1, and the sum of the light intensities detected in the light receiving areas E3 and E4 is A2, the change in the intensity A1 and A2 over time is shown. Assuming that the signals shown in FIG. 3 are obtained as the phase signals, it is possible to detect the light inclination (relative inclination direction and amount of inclination) between the two optical paths based on these signals. In this example, the light is tilted between the two optical paths by an angle corresponding to the phase difference Δ in the direction in which the light receiving regions E1 and E2 and the light receiving regions E3 and E4 are arranged (vertical direction in FIG. 2). . In addition, the intensity | strength of the vertical axis | shaft of FIG. 3 is shown by the relative value. Note that when the frequency of the phase signal is slow (low), it is possible to detect the inclination of the light between the two optical paths not from the phase comparison but from the intensity ratio.

  Even if the light spot D of the second interference light is not located at the center of the entire light receiving region (even if the light spot D is deviated from the center of the light receiving surface), the light received by the location of the light receiving surface Since the intensities are different, the inclination of light between the two optical paths can be detected from the intensity ratio of each light receiving region. For example, the ratio (first ratio) between the sum of the light intensities detected in the light receiving areas E1 and E2 and the sum of the light intensities detected in the light receiving areas E3 and E4, and the detection in the light receiving areas E1 and E4 Determining the ratio (second ratio) between the sum of the intensities of the detected light and the sum of the intensities of the light detected in the light receiving regions E2 and E3, and comparing the first ratio with the second ratio Thus, the inclination of light between the two optical paths can be detected.

  As described above, since the second photodetector 24 is configured by the four-divided sensor, the inclination correcting unit 100 (signal processing unit 101) is configured so that the distance between the two optical paths is based on the signal from each region of the four-divided sensor. It is possible to reliably detect the inclination of the light.

  The signal processing unit 101 detects the position of the movable mirror 15 based on the intensity of the second interference light detected by the second photodetector 24 and generates a pulse signal indicating the sampling timing. To do. The arithmetic unit 2 samples the detection signal (interferogram) from the first photodetector 17 in synchronization with the generation timing of the pulse signal and converts it into digital data. In the second light detector 24, the intensity of the second interference light changes between light and dark as a whole in accordance with the position (optical path difference) of the movable mirror, so by detecting the intensity, The position of the movable mirror 15 can be detected.

  The optical path correction device 102 corrects the optical path of the light reflected by the fixed mirror 14 based on the inclination of the light between the two optical paths detected by the signal processing unit 101, and corrects the inclination of the light between the two optical paths. The details will be described later.

  In the above configuration, the light (reference light) emitted from the reference light source 21 is reflected by the optical path combining mirror 22, converted into parallel light by the reflection collimator 12, and then incident on the BS 13, where it is separated into two light beams. The One light beam separated by the BS 13 is reflected by the movable mirror 15, and the other light beam is reflected by the fixed mirror 14. Each of the light beams returns back to the original optical path and is superimposed by the BS 13 to become second interference light. Thereafter, the second interference light is reflected and collected by the reflection collimator 16, is reflected by the optical path separation mirror 23, and enters the second photodetector 24.

  As described above, the signal processing unit 101 of the inclination correction unit 100 detects the light inclination between the two optical paths based on the intensity of the second interference light detected by the second photodetector 24. The optical path correction device 102 adjusts the attitude of the fixed mirror 14 (angle with respect to the BS 13) based on the detection result of the signal processing unit 101, and corrects the optical path of the reflected light from the fixed mirror 14. By performing feedback control (see FIG. 27) that repeatedly corrects the optical path of the reflected light and detects the inclination of the light between the two optical paths, the light inclination between the two optical paths is finally reduced to zero. You can get closer. Thereby, it can be avoided that the contrast of the first interference light detected by the first photodetector 17 is lowered.

[2. (Moving mirror drive mechanism)
(2-1. Overall configuration)
Next, the details of the drive mechanism 18 described above will be described. FIG. 4 is a perspective view illustrating a schematic configuration of the drive mechanism 18, and FIG. 5 is a cross-sectional view of the drive mechanism 18. In addition to the movable mirror 15, the drive mechanism 18 includes two leaf spring portions 31, 32, two rigid bodies 33, 34, a drive portion 35, a voltage application portion 36, and a holding portion 37. It is comprised by the parallel leaf spring which was made. 5 and the subsequent sectional views, illustration of an extraction electrode 53 and a fixed electrode 54, which will be described later, is omitted for convenience.

  As shown in FIG. 4, the drive mechanism 18 has different widths in the X direction on the rigid body 33 side and the rigid body 34 side, which is the region where the extraction electrode 53 and the fixed electrode 54 are formed, and the holding portion 37. This is to secure the formation area of the movable mirror 15, and this does not affect the parallel movement of the movable mirror 15.

  The leaf spring portions 31 and 32 are leaf springs disposed opposite to each other (in parallel) via the rigid bodies 33 and 34. These leaf spring portions 31 and 32 are formed using, for example, an SOI (Silicon on Insulator) substrate. The SOI substrate for forming the leaf spring portion 31 is configured by laminating a support layer 31a made of silicon, an insulating oxide film layer (BOX layer) 31b made of silicon oxide, and an active layer 31c made of silicon. Yes. Similarly, the SOI substrate for forming the leaf spring portion 32 is also configured by laminating a support layer 32a made of silicon, an insulating oxide film layer (BOX layer) 32b, and an active layer 32c made of silicon. . The support layers 31a and 32a are on the inside and the active layers 31c and 32c are on the outside, that is, the support layers 31a and 32a are closer to the rigid bodies 33 and 34 than the active layers 31c and 32c. The spring portions 31 and 32 are arranged to face each other. In addition, the direction where the leaf | plate spring parts 31 and 32 are facing is also called the Z direction below. This Z direction is the same as the moving direction of the movable mirror 15.

Support layer 31a, insulating oxide film layer 31b, support layer 32a, and insulating oxide film layer 32b are partially removed. More specifically, in the support layer 31a and the insulating oxide film layer 31b, a region facing the rigid body 33 and a region facing the rigid body 34 remain, and the other portions are removed. Note that the region of the support layer 31a that faces the rigid body 33 and the region that faces the rigid body 34 are the support layer 31a ₁ that directly faces the rigid body 33 in the support layer 31a and the support layer 31a ₂ that directly faces the rigid body 34, respectively. Point to. Further, the facing region and a region opposed to the rigid 34 and rigid body 33 in the insulating oxide film layer 31b, the insulating oxide film layer 31b, the insulating oxide film layer _31b 1 facing the rigid 33 through the support layer 31a _1, And the insulating oxide film layer 31b ₂ facing the rigid body 34 with the support layer 31a ₂ interposed therebetween.

Similarly, in the support layer 32a and the insulating oxide film layer 32b, a region facing the rigid body 33 and a region facing the rigid body 34 remain, and the other portions are removed. It should be noted that the region facing the rigid body 33 and the region facing the rigid body 34 in the support layer 32a are the support layer 32a ₁ directly facing the rigid body 33 in the support layer 32a and the support layer 32a ₂ directly facing the rigid body 34, respectively. Point to. Further, the facing region and a region opposed to the rigid 34 and rigid body 33 in the insulating oxide film layer 32b, the insulating oxide film layer 32b, the insulating oxide film layer _{32 b} 1 facing the rigid 33 through the support layer 32a _1, And the insulating oxide film layer 32b ₂ facing the rigid body 34 through the support layer 32a ₂ .

As a result of the partial removal of the support layers 31a and 32a and the insulating oxide film layers 31b and 32b in this way, portions of the active layer 31c excluding the region facing the rigid body 33 and the region facing the rigid body 34, In the active layer 32 c, the region excluding the region facing the rigid body 33 and the region facing the rigid body 34 are directly opposed via the space between the rigid body 33 and the rigid body 34. In the active layer 31c, the region facing the rigid body 33 and the region facing the rigid body 34 are the active layer 31c ₁ facing the rigid body 33 through the support layer 31a ₁ and the insulating oxide film layer 31b ₁ in the active layer 31c. And the active layer 31c ₂ facing the rigid body 34 through the support layer 31a ₂ and the insulating oxide film layer 31b ₂ respectively. Further, the facing region and a region opposed to the rigid 34 and rigid body 33 in the active layer 32c, the active layer 32c, an active layer 32c ₁ facing the rigid 33 through the support layer 32a ₁ and the insulating oxide film layer 32 b ₁ And the active layer 32c ₂ facing the rigid body 34 through the support layer 32a ₂ and the insulating oxide film layer 32b ₂ .

Moreover, the leaf | plate spring part 31 * 32 has the flat plate part 31p * 32p, respectively. The flat plate portions 31p and 32p are flat plate portions of the plate spring portions 31 and 32 that face each other with an air layer between the rigid body 33 and the rigid body 34 interposed therebetween. Here, the flat plate portions 31p and 32p are formed from the respective SOI substrates, the regions facing the rigid body 33 (support layers 31a ₁ and 32a ₁ , insulating oxide film layers 31b ₁ and 32b ₁ ) and the regions facing the rigid body 34 (supports). When the support layers 31a and 32a and the insulating oxide film layers 31b and 32b are removed except for the layers 31a ₂ and 32a ₂ and the insulating oxide film layers 31b ₂ and 32b ₂ ), a space between the rigid body 33 and the rigid body 34 is formed. The active layers 31c and 32c are opposed to each other.

The regions (support layers 31a ₁ and 31a ₂ ) facing the rigid bodies 33 and 34 in the support layer 31a of the leaf spring portion 31 are connected to the rigid bodies 33 and 34, respectively. Similarly, a region opposed to the rigid 33, 34 in the support layer 32a of the leaf spring portion 32 (supporting layer _32a 1 - 32a ₂₎ are respectively connected with the rigid 33, 34.

  The rigid bodies 33 and 34 are arranged apart from each other in the direction perpendicular to the direction in which the plate spring portions 31 and 32 face each other (Z direction). The direction in which the rigid bodies 33 and 34 are arranged apart from each other, that is, the direction in which the rigid bodies 33 and 34 are arranged side by side through the air layer is also referred to as the Y direction below. Here, the XYZ directions described above are orthogonal to each other.

The rigid body 33 is connected to the leaf spring portion 31 (particularly the support layer 31a ₁ ) and is connected to the leaf spring portion 32 (particularly the support layer 32a ₁ ). Similarly, the rigid body 34 is connected to the leaf spring portion 31 (particularly the support layer 31a ₂ ) and to the leaf spring portion 32 (particularly the support layer 32a ₂ ).

The rigid bodies 33 and 34 are both made of glass that is thicker than the flat plate portions 31p and 32p of the leaf spring portions 31 and 32. In the present embodiment, alkali glass containing, for example, sodium oxide (Na ₂ O) or potassium oxide (K ₂ O) is used as the glass.

In the present embodiment, the rigid bodies 33 and 34 are made of glass, and the support layers 31a ₁ and 31a ₂ of the leaf spring portion 31 and the support layers 32a ₁ and 32a _{2 of the} leaf spring portion 32 are both made of silicon. The rigid bodies 33 and 34 and the leaf spring portions 31 and 32 are connected by, for example, anodic bonding. Note that anodic bonding is a technique in which a direct voltage of several hundred volts is applied to silicon and glass at a temperature of several hundred degrees Celsius to form Si—O covalent bonds, thereby directly bonding the two.

  The holding portion 37 is a portion that is held by a fixing member or the like when the drive mechanism 18 is fixed to the interferometer 1, and is positioned above and below the rigid body 34 so that the drive mechanism 18 can be clamped and held up and down. The leaf springs 31 and 32 are provided at the edges of the outer surfaces (the surfaces opposite to the rigid bodies 33 and 34).

  The drive unit 35 translates the rigid body 33 and the movable mirror 15 (in the Z direction) relative to the rigid body 34 by bending and deforming one of the leaf springs 31 and 32. In this embodiment, the drive part 35 is provided on the surface of the leaf spring part 31, and the details of the arrangement position will be described later. On the other hand, the movable mirror 15 is provided above the rigid body 33 in the leaf spring portion 31 and on the surface opposite to the rigid body 33. The drive unit 35 and the movable mirror 15 may be provided on the leaf spring unit 32.

  Here, the drive unit 35 includes a piezoelectric element (PZT element) 35a that expands and contracts in accordance with an applied voltage from a voltage application unit 36 described later. As shown in FIG. 6, the piezoelectric element 35 a has a structure in which a piezoelectric material PZT (lead zirconate titanate) 41 is sandwiched between electrodes 42 and 43. By applying a positive or negative voltage to the electrodes 42 and 43 and expanding and contracting the PZT 41 in the horizontal direction, the leaf spring portion 31 can be bent and deformed, and the movable mirror 15 can be displaced together with the rigid body 33. For example, when the PZT 41 extends in the horizontal direction by applying a voltage to the electrodes 42 and 43, the leaf spring 31 is deformed so as to be convex upward, so that the movable mirror 15 is displaced downward together with the rigid body 33. On the other hand, when the PZT 41 is contracted in the horizontal direction by applying a voltage having a reverse polarity to the electrodes 42 and 43, the leaf spring 31 is deformed so as to protrude downward, so that the movable mirror 15 is moved upward along with the rigid body 33. It is displaced to.

  In this way, by applying a positive or negative voltage to the electrodes 42 and 43 and expanding and contracting the PZT 41 in the horizontal direction, the leaf spring portion 31 can be bent and deformed, whereby the rigid body 33 and the rigid body 34 can be deformed. The movable mirror 15 can be displaced.

  The voltage application unit 36 shown in FIG. 4 applies a voltage to the piezoelectric element 35a. Such application of voltage to the piezoelectric element 35a can be realized by the following configuration. A lead electrode 53 and a fixed electrode 54 are formed on the same surface as the surface on which the piezoelectric element 35 a is provided in the leaf spring portion 31. Prior to the formation of the piezoelectric element 35a, a metal film as the lead electrode 53 is vapor-deposited on the leaf spring portion 31, and the electrode 43 on the lower surface of the piezoelectric element 35a is brought into contact with the metal film, whereby the electrode 43 on the lower surface is formed. It can be pulled out. The extraction electrode 53 is wire bonded to the voltage application unit 36.

  The fixed electrode 54 is wire-bonded to the electrode 42 on the upper surface of the piezoelectric element 35a, and is also wire-bonded to the voltage applying unit 36. With this configuration, the voltage application unit 36 can apply a voltage to the piezoelectric element 35 a via the extraction electrode 53 and the fixed electrode 54. Note that the extraction electrode 53 and the fixed electrode 54 may be formed anywhere on the surface of the leaf spring portion 31 as long as wire bonding is easily performed above the rigid body 34.

(2-2. Resonant primary mode)
By the way, in the parallel leaf spring, even if a voltage is applied to the piezoelectric element, the moving body (for example, corresponding to the rigid body 33 and the moving mirror 15) may move in an inclined manner instead of a parallel movement. This is considered to be caused by the fact that only one leaf spring portion expands or contracts (deforms) due to expansion and contraction of the piezoelectric element, resulting in a difference in length between the two leaf spring portions.

Therefore, in the present embodiment, focusing on the fact that the movable body moves in parallel in the resonance primary mode of the parallel leaf spring, the voltage application unit 36 is determined by a system including a part of the leaf spring portions 31 and 32 and the rigid body 33. A voltage is applied to the piezoelectric element 35a at the same frequency f (= f ₀ ) as the primary resonance frequency f ₀ . As shown in FIG. 6, the resonance primary mode is a point A ₀ that does not displace at all in the Z direction due to expansion and contraction of the piezoelectric element 35 a in the leaf spring portion 31, for example, and the first antinode ( This refers to a vibration mode in which the point A ₁ ) has a maximum displacement at the position of the free end. The resonant frequency f ₀ (Hz) at this time is expressed as follows.
f ₀ = (1 / 2π) · √ (k / m)
However,
k: Spring constant of the spring part m: Mass of the translation part (g)
It is.

The above-mentioned “spring part” refers to a part that functions as a spring substantially by deformation in the leaf spring parts 31, 32. Specifically, the flat plate part 31 p of the leaf spring part 31 and the leaf spring part 32. The flat plate portion 32p. The “translation part” is a part that translates due to the deformation of the spring part. Specifically, the support layer 31 a ₁ , the insulating oxide film layer 31 b _1, and the active layer 31 c of the leaf spring part 31. _1, the support layer 32a ₁ of the plate spring portion 32, an insulating oxide film layer 32 b ₁ and the active layer 32c _1, refers to a rigid 33. Note that the mass of the movable mirror 15 is not considered in the above m. This is because the movable mirror 15 is a thin film and its mass can be almost ignored.

Thus, the voltage applying unit 36, since the plate spring portion 31, 32 and the rigid 33 applies a voltage to the piezoelectric element 35a at the same frequency f and the resonance frequency f ₀ when resonating together, rigid 33 and It can suppress that the movable mirror 15 inclines and moves. In addition, since the rigid body 33 and the movable mirror 15 are displaced by resonance, the amount of displacement is reliably increased as compared with the case where the rigid body 33 and the movable mirror 15 are displaced by applying a voltage to the piezoelectric element 35a at another frequency. Can be made.

(2-3. Arrangement Position of Drive Unit)
Next, details of the arrangement position of the drive unit 35 (piezoelectric element 35a) will be described. As shown in FIGS. 4 and 5, the piezoelectric element 35 a is provided on the surface of the leaf spring portion 31 opposite to the leaf spring portion 32. Moreover, the piezoelectric element 35a is closer to the rigid body 34 side than the center C of the flat plate portion 31p of the leaf spring portion 31 in the direction (Y direction) in which the rigid bodies 33 and 34 are arranged side by side on the surface of the leaf spring portion 31. Is provided. With the arrangement of the piezoelectric element 35a, the following effects can be obtained.

  First, since the piezoelectric element 35a has a structure in which the PZT 41 is sandwiched between the thin electrodes 42 and 43 as described above, it is much smaller and thinner than an electromagnetic drive source using a magnet and a coil such as a VCM. It is. Also, when using an electromagnetic drive source, the installation position (wide space) must also be secured, and the drive mechanism itself is enlarged, but when using a small and thin piezoelectric element 35a as a drive source, As in the present invention, the piezoelectric element 35a may be formed directly on the portion to be bent and deformed (the surface of the leaf spring portion 31), and it is not necessary to secure a wide installation space. Therefore, the small drive mechanism 18 can be reliably realized by configuring the drive unit 35 with the piezoelectric element 35 a and providing it on the surface of the leaf spring unit 31. As a result, the interferometer 1 to which the drive mechanism 18 is applied, and thus the spectroscopic analyzer can be reliably reduced in size.

  Further, since the piezoelectric element 35a is provided not on the entire surface of the flat plate portion 31p of the leaf spring portion 31, but on a part of the surface, the load on the piezoelectric element 35a during bending deformation of the leaf spring portion 31 is reduced. The piezoelectric element 35a can be driven even with a low driving voltage. Moreover, since the piezoelectric element 35a is provided on the rigid body 34 side with respect to the center C of the flat plate portion 31p, even if the piezoelectric element 35a is driven with a low driving voltage, the leaf spring portion 31 is reliably secured as shown in FIG. Accordingly, the rigid body 33 and the movable mirror 15 can be greatly displaced. Therefore, in a spectroscopic analyzer to which such a drive mechanism 18 is applied, high resolution can be reliably realized.

  Further, as shown in FIG. 5, a surface on the rigid body 34 side of the piezoelectric element 35a is a surface S1, and a surface on the rigid body 33 side is a surface S2. A surface on the rigid body 33 side of the rigid body 34 is defined as a surface S3. In the present embodiment, in the Y direction, the surface S1 is located on the opposite side of the rigid body 33 from the surface S3, and the surface S2 is on the rigid body 34 side with respect to the center C of the flat plate portion 31p, and the surface It is located closer to the rigid body 33 than S3.

  As described above, the piezoelectric element 35a is provided on the surface of the leaf spring portion 31 so as to straddle the surface S3, that is, a part of the piezoelectric element 35a is located above the rigid body 34. If the lead electrode 53 and the fixed electrode 54 are formed on the surface of the leaf spring portion 31 and above the rigid body 34, the lead electrode 53 and the fixed electrode 54 and other parts (for example, the electrode 42 on the upper surface of the piezoelectric element 35a, the voltage) It is possible to avoid breakage of the leaf spring portion 31 when the application portion 36) is connected by wire bonding. That is, since the bonding operation at this time is performed above the rigid body 34, no external stress acts on the flat plate portion 31 p, and damage to the plate spring portion 31 due to this can be avoided. Further, since the bonded wire is located above the rigid body 34 rather than above the flat plate portion 31p, the wire may suppress the flat plate portion 31p and inhibit deformation (resonance) of the leaf spring portion 31 even after bonding. Therefore, it is possible to avoid adversely affecting the deformation of the leaf spring portion 31.

At this time, in the Y direction, when the length of the flat plate portion 31p of the leaf spring portion 31 is L1 (mm) and the length of the piezoelectric element 35a on the rigid body 33 side from the surface including the surface S3 is L2 (mm),
L2 / L1 ≦ 0.3
It is desirable to satisfy The reason is as follows.

  In the configuration in which the piezoelectric element 35a is provided on one of the leaf spring portions 31 and 32 (here, the leaf spring portion 31), when the length of the piezoelectric element 35a on the flat plate portion 31p is increased, the leaf spring portions 31 and 32 are interposed. The balance at the time of deformation is lost, the rigid body 33 and the movable mirror 15 are tilted and moved, and the tilt is increased. In other words, the shorter the piezoelectric element 35a is on the flat plate portion 31p, the smaller the inclination during movement of the rigid body 33 and the movable mirror 15 may be.

  FIG. 7 is a graph showing the relationship between L2 / L1 and the applied voltage (electric field strength) to the piezoelectric element 35a when the displacement amount of the movable mirror 15 is constant. Note that the electric field strength on the vertical axis in FIG. 7 is normalized by the length of the piezoelectric element 35a. From the figure, when L2 / L1> 0.3, the applied voltage itself can be reduced, but the change amount of the applied voltage with respect to the change amount of L2 / L1 is small, and the effect of greatly reducing the applied voltage is small.

  Therefore, by satisfying L2 / L1 ≦ 0.3, L2 is sufficiently small with respect to L1, so that the piezoelectric element can be efficiently obtained while sufficiently obtaining the effect of suppressing the inclination of the rigid body 33 and the movable mirror 15 during movement. 35a can be driven.

  In addition, said conditional expression can be obtained regardless of the constituent material of the leaf | plate spring parts 31 * 32 and the rigid body 33, and the thickness of each flat plate part 31p * 32p. The reason is as follows.

  When L2 is smaller than L1, stress is most applied to the end of the piezoelectric element 35a on the rigid body 33 side, and force is applied in the direction in which the end separates from the leaf spring portion 31, so that the rigid body 33 and the movable mirror 15 In order to obtain a desired displacement, it is necessary to apply a large voltage to the piezoelectric element 35a (see FIG. 7). On the other hand, when L2 is larger than L1, the piezoelectric element 35a follows along with the application of voltage to the piezoelectric element 35a, so that the end of the piezoelectric element 35a on the rigid body 34 side is most stressed. Even if the constituent materials of the leaf springs 31 and 32 and the rigid body 33 are changed and the thicknesses of the flat plate portions 31p and 32p are changed, the tendency is that the displacement amount of the rigid body 33 and the movable mirror 15 is constant. Is the same. Therefore, depending on the constituent materials of the leaf springs 31 and 32 and the rigid body 33 and the thickness of the flat plate parts 31p and 32p, the applied voltage to the piezoelectric element 35a to obtain a desired displacement amount of the rigid body 33 and the movable mirror 15 is obtained. Although the value (absolute value) of fluctuates, the ratio of L2 / L1 is 0.3 or less regardless of the constituent materials of the leaf spring portions 31 and 32 and the rigid body 33 and the thickness of the flat plate portions 31p and 32p. I can say that.

(2-4. Configuration in which two piezoelectric elements are provided)
FIG. 8 is a perspective view showing another configuration of the drive mechanism 18, and FIG. 9 is a cross-sectional view of the drive mechanism 18 of FIG. This drive mechanism 18 is different from the drive mechanism 18 of FIGS. 4 and 5 in that it includes a piezoelectric element 38 different from the piezoelectric element 35a. The piezoelectric element 38 is a surface of the leaf spring portion 32 opposite to the leaf spring portion 31 and is perpendicular to the surface R perpendicular to the opposing direction (Z direction) of the leaf spring portions 31 and 32. It is provided in the position which becomes symmetrical.

  By providing the piezoelectric element 38 in this way, the upper and lower balance of the drive mechanism 18 constituting the parallel leaf spring can be kept good, that is, the leaf spring portions 31 and 32 can be deformed in a balanced manner. It is possible to improve parallelism when the movable mirror 15 is translated. That is, the rigid body 33 and the movable mirror 15 can be moved (displaced) in a state that is almost as parallel as possible. In addition, the piezoelectric element 38 only needs to be provided as described above, and wiring for applying a voltage to the piezoelectric element 38 is not necessary. Therefore, the parallelism during parallel movement can be easily improved with a simple configuration.

  FIG. 10 is a cross-sectional view showing still another configuration of the drive mechanism 18. The driving mechanism 18 has a configuration in which the two piezoelectric elements 35a and 38 are provided as shown in FIGS. 8 and 9, and the voltage applying unit 36 applies a voltage to both the piezoelectric elements 35a and 38. That is, on the leaf spring portion 32 side, electrodes (not shown) corresponding to the lead electrode 53 and the fixed electrode 54 provided on the leaf spring portion 31 side are provided, and these electrodes and the upper and lower electrodes of the piezoelectric element 38 and The voltage application unit 36 is electrically connected by wire bonding. In this configuration, the voltage application unit 36 applies the piezoelectric elements 35a and 38 so that the deformation of the leaf spring part 31 due to the expansion and contraction of the piezoelectric element 35a is the same as the deformation of the leaf spring part 32 due to the expansion and contraction of the piezoelectric element 38. It is desirable to apply a voltage.

  That is, in FIG. 10, in order to move the rigid body 33 and the movable mirror 15 upward (Z direction) by deformation of the leaf spring portions 31 and 32, the piezoelectric element 35a is contracted in the horizontal direction, while the piezoelectric element 38 is in the horizontal direction. It is necessary to apply a voltage extending to the piezoelectric elements 35a and 38. On the contrary, in order to move the rigid body 33 and the movable mirror 15 downward by deformation of the leaf spring portions 31 and 32, a voltage is applied so that the piezoelectric element 35a extends in the horizontal direction while the piezoelectric element 38 contracts in the horizontal direction. It is necessary to apply to each piezoelectric element 35a and 38.

  Therefore, in FIG. 10, if the polarization directions of PZT in the piezoelectric elements 35a and 38 are opposite to each other (one is upward and the other is downward), among the two electrodes sandwiching the PZT of the piezoelectric elements 35a and 38, PZT The voltage applying unit 36 may apply a voltage to each of the piezoelectric elements 35a and 38 so that voltages having the same polarity are applied to the electrodes located on the same side with respect to the same frequency. For example, the voltage application unit 36 includes an upper electrode 42 of the piezoelectric element 35a (an electrode on the opposite side of the rigid body 34 with respect to the PZT 41) and an upper electrode of the piezoelectric element 38 (an electrode on the rigid body 34 side of the PZT). The voltage may be applied to each of the piezoelectric elements 35a and 38 so that the same polarity voltage is applied to the piezoelectric elements 35a and 38.

  Further, if the polarization directions of PZT in the piezoelectric elements 35a and 38 are, for example, the same direction (for example, both are upward), the two electrodes sandwiching the PZT of the piezoelectric elements 35a and 38 are located on the same side with respect to the PZT. The voltage applying unit 36 may apply a voltage to each of the piezoelectric elements 35a and 38 so that voltages having opposite polarities are applied to the electrodes at the same frequency. For example, when a positive voltage is applied to the upper electrode 42 of the piezoelectric element 35a, the voltage application unit 36 applies a negative voltage to the upper electrode of the piezoelectric element 38 (an electrode on the rigid body 34 side with respect to PZT). As described above, a voltage may be applied to each of the piezoelectric elements 35a and 38.

  When the voltage application unit 36 applies a voltage to each of the piezoelectric elements 35a and 38 as described above, the leaf spring units 31 and 32 can be deformed (resonated) in the same manner, and one deformation changes the other. It is easy to resonate without being disturbed. Therefore, the parallelism when the rigid body 33 and the movable mirror 15 are translated can be reliably improved.

(2-5. Configuration capable of handling fluctuations in resonance frequency)
By the way, the piezoelectric element expands and contracts when a voltage is applied. Conversely, when the piezoelectric element is deformed by applying a force, it outputs a voltage corresponding to the distortion. When the displacement of the rigid body 33 and the movable mirror 15 becomes maximum at the time of resonance, the piezoelectric element is also greatly distorted, so that the voltage (for example, absolute value) output from the piezoelectric element becomes maximum. Therefore, by utilizing this fact and monitoring the voltage (particularly the maximum voltage) output from the piezoelectric element, it is possible to detect whether or not the rigid body 33 and the movable mirror 15 are displaced due to resonance. It is also possible to cope with fluctuations. Note that the fluctuation of the resonance frequency may occur, for example, when the shape of the spring portion changes due to thermal expansion or contraction due to a change in environmental temperature due to heat generation of the reference light source 21 and the spring constant changes. Hereinafter, a specific configuration that can cope with fluctuations in the resonance frequency will be described.

  FIG. 11 is a cross-sectional view showing still another configuration of the drive mechanism 18. The drive mechanism 18 includes a detection unit 39 and a control unit 40 in addition to the configuration in which the two piezoelectric elements 35a and 38 are provided as shown in FIGS. Since the configurations of FIGS. 8 and 9 are basic, the voltage applying unit 36 applies a voltage only to one piezoelectric element 35a, and does not apply a voltage to the other piezoelectric element 38 for the sake of safety. Keep it.

  The detection unit 39 is a sensor that detects the maximum displacement of the rigid body 33 and the movable mirror 15 from the voltage corresponding to the distortion of the piezoelectric element 38 that is output from the piezoelectric element 38 when the leaf spring portion 32 is deformed. As the voltage (for example, absolute value) output from the piezoelectric element 38 is larger, the rigid body 33 and the movable mirror 15 are displaced more greatly. Therefore, the detection unit 39 is the maximum voltage of the voltage output from the piezoelectric element 38 ( For example, the maximum displacement of the rigid body 33 and the movable mirror 15 can be detected by detecting the absolute value. The detection unit 39 detects the displacement amount and direction (displaced position) of the rigid body 33 and the movable mirror 15 based on the magnitude and direction (positive / negative sign) of the voltage output from the piezoelectric element 38. You can also.

  The control unit 40 controls voltage application to the piezoelectric element 35 a by the voltage application unit 36. More specifically, the control unit 40 determines the voltage so that the frequency of the applied voltage to the piezoelectric element 35a varies according to the variation of the maximum displacement of the rigid body 33 and the movable mirror 15 detected by the detection unit 39. The application unit 36 is controlled. The control unit 40 that performs such control includes a CPU (Central Processing Unit).

  The voltage application unit 36 is configured to include a VCO (voltage controlled oscillator) circuit. The VCO circuit is a circuit that changes the frequency of the output voltage in accordance with the input voltage. Therefore, the voltage application unit 36 can change the frequency of the output voltage (voltage applied to the piezoelectric element 35 a) under the control of the control unit 40.

  Next, the flow of operation by the voltage control of the control unit 40 will be described based on FIGS. 12 and 13. 12 and 13 are flowcharts showing the flow of operation by voltage control of the control unit 40. FIG. 12 shows the initial operation for searching for the resonance frequency, and FIG. It shows the one in the steady operation to follow.

  In the initial operation, first, the control unit 40 starts increasing the input voltage of the VCO circuit of the voltage applying unit 36 and increases the frequency of the output voltage (S1). Then, the detection unit 39 detects the maximum value (for example, absolute value) of the voltage output from the piezoelectric element 38, and detects the maximum displacement of the rigid body 33 and the movable mirror 15 (S2).

  Subsequently, the control unit 40 compares the maximum displacement detected in S2 with the maximum displacement detected before that (S3). Here, since the maximum displacement of the rigid body 33 and the movable mirror 15 is zero at the start of operation, the maximum displacement detected in S2 is necessarily larger than the maximum displacement detected before that (No in S3). ). Therefore, the control unit 40 further increases the input voltage of the VCO circuit and further increases the frequency of the output voltage (S4).

  Thereafter, the steps S2 to S4 are repeated, and the control unit 40, when the maximum displacement detected in S2 becomes smaller than the maximum displacement detected before (Yes in S3), the rigid body 33 and the movable mirror It is determined that 15 has reached the maximum displacement, and the increase in the input voltage of the VCO circuit is stopped (S5). When the rigid body 33 and the movable mirror 15 reach the maximum displacement, it can be determined that the rigid body 33 and the movable mirror 15 vibrate due to resonance. Therefore, the frequency of the output voltage of the VCO circuit at this time can be considered as the resonance frequency. . That is, by such an initial operation, the resonance frequency can be obtained without using the theoretical calculation described above.

  In the steady operation, the detection unit 39 detects the maximum value (for example, absolute value) of the voltage output from the piezoelectric element 38, and detects the maximum displacement of the rigid body 33 and the movable mirror 15 (S11). The control unit 40 compares the maximum displacement detected in S11 with the maximum displacement detected before that (for example, the maximum displacement detected immediately before stopping the increase of the input voltage of the VCO circuit in S5 of FIG. 12). (S12). Here, when the resonance frequency does not fluctuate, the maximum displacement detected in S11 is the same as the previous maximum displacement (No in S12). In this case, this flow ends. When there are a plurality of resonance frequencies, the maximum displacements of the rigid body 33 and the movable mirror 15 are stored in memory, and the frequency at which the maximum displacement is obtained is defined as the resonance frequency.

  On the other hand, when the resonance frequency varies, the maximum displacement detected in S11 becomes smaller than the previous maximum displacement (Yes in S12). In this case, the control unit 40 increases the input voltage of the VCO circuit, The frequency is increased (S13). Then, the detection unit 39 detects the maximum value of the voltage output from the piezoelectric element 38, and detects the maximum displacement of the rigid body 33 and the movable mirror 15 (S14).

  Subsequently, the control unit 40 compares the maximum displacement detected in S14 with the maximum displacement detected before that (for example, the maximum displacement detected in S11) (S15). When the maximum displacement detected in S14 is larger than the maximum displacement detected before (No in S15), it can be seen that the maximum displacement increases in the direction in which the frequency increases. The input voltage is further increased to increase the frequency of the output voltage (S16). Then, the detection unit 39 detects the maximum value of the voltage output from the piezoelectric element 38, and detects the maximum displacement of the rigid body 33 and the movable mirror 15 (S17). Then, the controller 40 compares the maximum displacement detected in S17 with the maximum displacement detected before that (for example, the maximum displacement detected in S14) (S18).

  If the maximum displacement detected in S17 is larger than the maximum displacement detected before that (No in S18), it is considered that there is room for the maximum displacement to increase, so the steps S16 to S18 are repeated. Then, when the maximum displacement detected in S17 becomes smaller than the maximum displacement detected before (Yes in S18), the control unit 40 reaches the maximum displacement of the rigid body 33 and the movable mirror 15, and the piezoelectric element. It is determined that the frequency of the voltage applied to 35a matches the resonance frequency, and this flow is finished.

  In S15, when the maximum displacement detected in S14 is smaller than the maximum displacement detected before (Yes in S15), it can be seen that the maximum displacement increases in the direction in which the frequency decreases. Decreases the input voltage of the VCO circuit and decreases the frequency of the output voltage (S19). Then, the detection unit 39 detects the maximum value of the voltage output from the piezoelectric element 38, and detects the maximum displacement of the rigid body 33 and the movable mirror 15 (S20). Then, the controller 40 compares the maximum displacement detected in S20 with the maximum displacement detected before that (for example, the maximum displacement detected in S14) (S21).

  If the maximum displacement detected in S20 is larger than the maximum displacement detected before that (No in S21), it is considered that there is room for the maximum displacement to increase, so the steps S19 to S21 are repeated. When the maximum displacement detected in S20 becomes smaller than the maximum displacement detected before that (Yes in S21), the control unit 40 reaches the maximum displacement of the rigid body 33 and the movable mirror 15, and the piezoelectric element. It is determined that the frequency of the voltage applied to 35a matches the resonance frequency, and this flow is finished.

  As described above, the control unit 40 determines the frequency of the output voltage of the VCO circuit (the frequency of the voltage applied to the piezoelectric element 35a) in accordance with the change in the maximum displacement of the rigid body 33 and the movable mirror 15 detected by the detection unit 39. The voltage application unit 36 is controlled so as to fluctuate, so that the frequency of the output voltage of the VCO circuit is changed to a fluctuating resonance frequency even if the resonance frequency fluctuates due to a change in environmental temperature or the like during the operation of the drive mechanism 18. Can be followed.

  In particular, the control unit 40 controls the voltage application unit 36 (VCO circuit) so that the frequency of the voltage applied to the piezoelectric element 35a matches the resonance frequency based on the voltage detected by the detection unit 39. Even when the drive mechanism 18 of the present invention is used in an environment where the resonance frequency fluctuates or is likely to fluctuate, a stable resonance state can always be maintained.

  In the above description, the displacement of the rigid body 33 and the movable mirror 15 is monitored based on the voltage detected by the detection unit 39, and the voltage application unit 36 is controlled based on the displacement. The vibration frequency is detected from the number of inversions of the positive and negative signs of the detected voltage within a predetermined time, and the voltage application unit 36 is controlled based on the vibration frequency (the frequency of the voltage applied to the piezoelectric element 35a is varied). It is also possible to follow the resonance frequency).

(2-6. Manufacturing method of drive mechanism)
Next, as a method for manufacturing the drive mechanism 18, a method for manufacturing the drive mechanism 18 shown in FIG. 4 will be described as an example. Note that the same manufacturing method can be applied to the other drive mechanisms 18 described above.

  FIG. 14 is a flowchart showing a rough flow at the time of manufacturing the drive mechanism 18 of FIG. 15A to 15D are cross-sectional views showing the manufacturing process of the drive mechanism 18. First, as shown to Fig.15 (a), the two leaf | plate spring parts 31 * 32 are produced (S31). In addition, the detail of the production methods of the leaf | plate spring parts 31 * 32 is mentioned later.

  Subsequently, as shown in FIG. 15 (b), the rigid bodies 33 and 34 are arranged apart from each other, and the flat plate portions 31p and 32p are opposed to each other through the space between the rigid body 33 and the rigid body 34. The leaf spring portions 31 and 32 are arranged via the rigid bodies 33 and 34 (S32).

  Next, as shown in FIG. 15C, the movable mirror 15 is formed on the leaf spring portion 31 (S33), and the drive portion 35 is formed at a predetermined position of the leaf spring portion 31 (S34). The movable mirror 15 is formed in S33 by, for example, sputtering Au to the leaf spring portion 31. Alternatively, the movable mirror 15 may be formed by forming a metal material such as Al or Pt on the leaf spring portion 31 by vapor deposition or adhesion. The drive unit 35 is formed in S34 by, for example, bonding the above-described piezoelectric element 35a to the leaf spring unit 31 using an adhesive. At this time, the lead electrode 53 and the fixed electrode 54 (see FIG. 4) are simultaneously formed on the leaf spring portion 31 by sputtering of a metal material or the like.

  In addition, you may change the order of S32-S34 suitably. For example, S34 may be performed before S33, or the step S32 may be performed after S33 and S34.

  Then, as shown in FIG.15 (d), the rigid bodies 33 * 34 and the leaf | plate spring parts 31 * 32 are connected (S35). However, the connection at this time is performed by anodic bonding under a high temperature and high electric field. Then, the electrode 42 on the upper surface of the piezoelectric element 35a and the fixed electrode 54, the fixed electrode 54 and the voltage applying unit 36, and the lead electrode 53 and the voltage applying unit 36 are connected by wire bonding (S36). Thereby, the drive mechanism 18 is completed.

  Although the case where one drive mechanism 18 is manufactured has been described above, a plurality of (for example, four) drive mechanisms 18 can be manufactured at the same time. In that case, the following may be performed.

  FIG. 16 is a perspective view of a substrate 51 in which four leaf spring portions 31 (or four leaf spring portions 32) are bound in a sheet shape, and is a perspective view seen from a side facing a support block 52 described later. It is. When four drive mechanisms 18 are manufactured simultaneously, two such substrates 51 are prepared (corresponding to S31).

  Then, the two substrates 51 and 51 are arranged to face each other via the alkali glass support block 52 shown in FIG. 17 (corresponding to S32). The support block 52 is provided with four rigid bodies 33 and 34 in a state where a space is provided between the rigid body 33 and the rigid body 34 constituting one drive mechanism 18, and these are continuously formed. Is.

Subsequently, as shown in FIG. 18, the movable mirror 15 and the drive unit 35 are respectively formed on predetermined portions of the substrate 51 (corresponding to S33 and S34). At this time, the extraction electrode 53 is formed in common with the electrode 42 on the lower surface of the piezoelectric element 35a of the adjacent drive unit 35, and the fixed electrode 54 is formed corresponding to each piezoelectric element 35a. The substrates 51 and 51 and the support block 52 are joined by anodic bonding while positioning with the three positioning pins 55 (corresponding to S35). Thereafter, the conjugate (the substrates 51, 51, the support block 52) a dicer cuts along the bold line _D 1, _{D 2,} disconnecting the movable mirror 15 from the support piece 56.

Further, as shown in FIG. 19, the joined body is cut by a dicer along the thick lines D ₃ and D ₄ and divided into four drive mechanisms 18. Finally, after unnecessary portions are further dicer cut, the electrode 42 and the fixed electrode 54 on the upper surface of each piezoelectric element 35a, the fixed electrode 54 and the voltage applying unit 36, and the lead electrode 53 and the voltage applying unit 36 are connected by wire bonding. Connect (corresponding to S36). Thereby, the four drive mechanisms 18 are completed.

(2-7. About the manufacturing method of a leaf | plate spring part)
Next, the detail of the manufacturing method of the leaf | plate spring part 31 * 32 mentioned above is demonstrated. Here, in order to facilitate understanding of the description, details of a method for manufacturing the leaf spring portion 31 using the substrate 51 of FIG. 16 will be described. A similar method can be adopted for the method for producing the leaf spring portion 32.

  20A to 20F are cross-sectional views when the manufacturing process of the leaf spring portion 31 is viewed in the cross-section taken along the line A-A ′ in FIG. 16. For convenience of explanation, the portion corresponding to the space around the leaf spring portion 31 is defined as the penetration portions 71 and 72 through the substrate 51 on the A-A ′ line in FIG. Further, a portion of the substrate 51 corresponding to the flat plate portion 31 p of the leaf spring portion 31 is defined as a region 73.

  First, as shown in FIG. 20A, thermal oxide films 62 and 63 serving as masks are sequentially formed on the SOI substrate 61 by a photolithography process (not shown). The SOI substrate 61 is configured by laminating a support layer 31a made of silicon, an insulating oxide film layer 31b made of silicon oxide, and an active layer 31c made of silicon. The thermal oxide films 62 and 63 are formed on the support layer 31 a side in the SOI substrate 61.

  Subsequently, as shown in FIG. 20B, removal of the support layer 31a located in the through portions 71 and 72 in the SOI substrate 61 is started by dry etching using the thermal oxide film 62 as a mask, and the thermal oxide film The removal of the thermal oxide film 62 located in the region 73 is started using 63 as a mask. Then, after the thermal oxide film 62 in the region 73 is completely removed, as shown in FIG. 20C, the support located in the through portions 71 and 72 is formed by dry etching using the remaining thermal oxide film 62 as a mask. The layer 31a and the support layer 31a located in the region 73 are simultaneously removed. When the support layer 31a located in the through portions 71 and 72 is completely removed by such stepwise removal of the support layer 31a, the support layer 31a in the region 73 remains slightly.

  Next, as shown in FIG. 20D, the insulating oxide film layer 31b located in the through portions 71 and 72 is removed by dry etching using the support layer 31a as a mask. Thereafter, as shown in FIG. 20E, the active layer 31c located in the through portions 71 and 72 and the support layer 31a located in the region 73 are simultaneously removed by dry etching using the thermal oxide film 62 as a mask. Finally, as shown in FIG. 20F, the insulating oxide film layer 31b in the region 73 is removed by dry etching using the support layer 31a as a mask. By removing the remaining thermal oxide film 62, the leaf spring portion 31 in the substrate 51 of FIG. 16 is completed.

  As described above, by forming the two leaf spring portions 31 and 32 of the drive mechanism 18 using the SOI substrate 61, as described above, so-called MEMS (Micro Electro Mechanical Systems) technology, that is, photolithography and The drive mechanism 18 can be manufactured using a technique that combines a semiconductor manufacturing technique such as etching and a bonding technique such as anodic bonding. Further, by using the MEMS technology, it is possible to avoid variations in the lengths of the two flat plate portions 31p and 32p in one driving mechanism 18 as long as the lithography mask accuracy is ensured with high accuracy. . As a result, it is possible to suppress the inclination of the movable part (the rigid body 33 and the movable mirror 15) when the drive mechanism 18 is assembled or translated, and to translate the movable part with considerable accuracy (about ± 0.3 minutes). be able to. In addition, it is possible to eliminate individual differences, that is, to avoid variations in the length of the flat plate portions 31p and 32p for each individual of the plurality of drive mechanisms 18, so that the plurality of drive mechanisms 18 can be stably manufactured.

(2-8. Other configurations of drive mechanism)
FIG. 21 is a cross-sectional view showing still another configuration of the drive mechanism 18. As shown in the figure, the flat plate portion 31p of the plate spring portion 31 of the drive mechanism 18 may be composed of two layers of an insulating oxide film layer 31b and an active layer 31c, and the flat plate portion 32p of the plate spring portion 32. May be composed of two layers of an insulating oxide film layer 32b and an active layer 32c.

  FIG. 22 is a cross-sectional view showing still another configuration of the drive mechanism 18. As shown in the figure, the leaf spring portions 31 and 32 of the drive mechanism 18 may be constituted by flat silicon substrates 81 and 81, respectively. In addition, anodic bonding can be used for the connection between the leaf spring portions 31 and 32 (silicon substrates 81 and 81) and the rigid bodies 33 and 34 made of glass. In this configuration, the drive mechanism 18 can be easily realized with a simple configuration in which the rigid bodies 33 and 34 are sandwiched between the flat silicon substrates 81 and 81. Moreover, compared with the case where the SOI substrate 61 is used, the manufacturing process of the leaf spring portions 31 and 32 (the process of S31) can be greatly simplified.

  FIG. 23 is a cross-sectional view showing still another configuration of the drive mechanism 18. As shown in the figure, the plate spring portions 31 and 32 of the drive mechanism 18 may be configured by flat glass substrates 91 and 91, respectively. In this case, for example, glass substrates 91 and 91 can be obtained by performing laser processing or dicing processing on glass (for example, alkali glass) having a thickness of 100 μm or less. Optical contact or diffusion bonding can be used for connection between the leaf spring portions 31 and 32 (glass substrates 91 and 91) and the rigid bodies 33 and 34 made of glass. The optical contact is a method in which smooth surfaces are brought into close contact with each other and two members are connected by molecular attraction. Diffusion bonding is a method of obtaining a bonded portion by heating and pressurizing and holding a base material without melting it and diffusing atoms at the bonded interface across the bonded surface.

  Thus, by configuring the leaf spring portions 31 and 32 with the glass substrates 91 and 91, respectively, the drive mechanism 18 can be easily realized with a simple configuration in which the rigid bodies 33 and 34 are sandwiched between the flat glass substrates 91 and 91. can do. Moreover, compared with the case where the SOI substrate 61 is used, the manufacturing process of the leaf spring portions 31 and 32 (the process of S31) can be greatly simplified. Further, since the constituent materials of the rigid bodies 33 and 34 and the leaf spring portions 31 and 32 are both glass, the deformation of the drive mechanism 18 due to a temperature change can be reliably prevented, and the movable part (the rigid body) is caused by the temperature change. 33 and the movable mirror 15) can be reliably prevented from tilting.

  In addition, the connection between the rigid bodies 33 and 34 and the leaf spring portions 31 and 32 employs an adhesive-free method such as the anodic bonding described above, optical contact, diffusion bonding, or the like. Manufacturing errors (influence of shrinkage of adhesive during manufacturing) can be eliminated, and a large corner cube should be installed when the drive mechanism 18 is applied to the interferometer 1 or the spectroscopic analyzer. In addition, high resolution by high-precision interference can be realized. That is, it is possible to achieve high resolution by high-precision interference while miniaturizing the interferometer 1 and thus the spectroscopic analyzer.

  In addition, the leaf | plate spring parts 31 * 32 may be comprised with the flat plate which consists of metals (iron, aluminum, an alloy, etc.) instead of said silicon substrate 81 or glass substrate 91. FIG.

  The rigid bodies 33 and 34 described above may be made of silicon instead of glass. At this time, when the connecting portion between the leaf spring portions 31 and 32 and the rigid bodies 33 and 34 is a connection between silicon and glass, anodic bonding can be used as a bonding method, and the connecting portion is formed between silicon and silicon. In this case, optical contact or diffusion bonding can be used as a bonding method.

[3. About the performance of the movable mirror during resonance driving)
Next, the performance when the movable mirror 15 is resonantly driven by the drive mechanism 18 having the above-described configuration will be described. FIG. 24 shows tilt error amounts (light tilt amounts between two optical paths) in the pitch and roll directions when the movable mirror is moved ± 1.5 mm in the Z direction by the drive mechanism 18. Here, FIGS. 25A and 25B show the Pitch direction and the Roll direction, which are the tilt directions of the movable mirror 15. Each direction of Pitch and Roll is defined as follows. That is, in the configuration in which the extending direction of the leaf spring portion 31 of the drive mechanism 18 (the direction in which the rigid bodies 33 and 34 are arranged) is the Y direction, the movable mirror 15 is translated in the Z direction by resonance of the leaf spring portion 31. When the direction perpendicular to X is the X direction, the Pitch direction is the tilt direction when the movable mirror 15 tilts in the YZ plane, and the Roll direction is the tilt direction when the movable mirror 15 tilts in the ZX plane. It is.

  When the movable mirror 15 is translated ± 1.5 mm in the Z direction, from FIG. 24, the tilt error in the pitch direction remains about ± 0.3 minutes, and the tilt error in the roll direction remains about ± 0.7 minutes. I understand that. Factors that cause tilt errors in the pitch direction include, for example, the thickness error in the Z direction of the parallel leaf spring, the tilt toward the tip of the parallel moving part, the thickness error in the Z direction of the parallel moving part and the support part, the parallel moving part and the support. The flatness of the part may be reduced. On the other hand, the cause of the tilt error in the Roll direction is, for example, the thickness error in the X direction of the parallel moving part, the shift error in the X direction of the parallel leaf spring and the parallel moving part, and the sticking error in the X direction of the piezoelectric element 35a. And twisting of parallel leaf springs in the Roll direction.

In addition, said parallel leaf | plate spring points out the part which functions as a spring substantially by deformation | transformation in leaf | plate spring part 31 * 32, specifically, the flat plate part 31p of the leaf | plate spring part 31 and the leaf | plate spring part 32 It refers to the flat plate portion 32p. In addition, the translation unit is a part that translates by deformation of the parallel plate spring. Specifically, for example, in the configuration of FIG. 5, the support layer 31 a ₁ of the plate spring unit 31, the insulating oxide film The layer 31b ₁ and the active layer 31c ₁ , the support layer 32a ₁ of the leaf spring portion 32, the insulating oxide film layer 32b ₁ and the active layer 32c _1, and the rigid body 33 are indicated. In addition, the above-mentioned support portion refers to the support layer 31 a ₂ , the insulating oxide film layer 31 b ₂ , the active layer 31 c ₂ , the support layer 32 a ₂ , the insulating oxide film layer 32 b ₂ , the active layer 32 c ₂ , and the rigid body 34.

  In the present embodiment, the tilt error generated by the resonance driving of the movable mirror 15 is corrected by non-resonant driving of the fixed mirror 14 by the optical path correction device 102 of the correction unit 100 described above. Details of the optical path correction apparatus 102 will be described below.

[4. Details of the optical path correction device)
(4-1. Configuration of optical path correction device)
FIG. 26A is a side view showing a schematic configuration of the optical path correction device 102, and FIG. 26B is a plan view of the fixed mirror 14 supported by the optical path correction device 102. The optical path correction device 102 has a plurality of (four in FIG. 26B) piezoelectric elements 103 that expand and contract by voltage application. One end surface of each piezoelectric element 103 in the expansion / contraction direction is connected to the fixed mirror 14, while the other end surface is connected to the fixed base 104, and each piezoelectric element 103 is supported. The piezoelectric elements 103 are arranged at regular intervals (every 90 degrees) in the circumferential direction with respect to an axis passing through the center of the bottom surface of the fixed mirror 14. As a result, the two piezoelectric elements 103 and 103 are disposed to face each other with the shaft interposed therebetween, and there are two sets of piezoelectric elements 103 and 103 that are disposed to face each other in this way.

  Here, when the movable mirror 15 is resonantly driven at a resonance frequency of 50 Hz, for example, if the tilt error of the movable mirror 15 includes up to the fourth-order high frequency component, the optical path correction device 102 has a non-frequency at 200 Hz. A performance capable of resonance driving (servo driving described later) is required. Therefore, in this embodiment, a high-speed response is realized by using a plurality of stacked piezoelectric elements 103. By using four piezoelectric elements 103, tilt error correction (also referred to as tilt correction) in each direction of pitch and roll can be performed with each pair of piezoelectric elements. This will be specifically described below. Note that tilt correction is possible even when the number of piezoelectric elements 103 is three.

  Assuming that the voltage that can be applied to the piezoelectric element 103 is 150 V, the optical path correction device 102 is designed so that the voltage application up to 100 V can be adjusted with a slight margin with respect to 150 V. When the mechanical tilt error (design error) between the fixed mirror 14 side and the movable mirror 15 side is zero, the applied voltage to each piezoelectric element 103 is set to 50 V, for example. On the other hand, when there is a tilt error in each direction of Pitch and Roll, a voltage larger than 50 V is applied to one of the piezoelectric elements 103 and 103 arranged opposite to each other, and a voltage smaller than 50 V is applied to the other. By applying, one piezoelectric element 103 expands and the other piezoelectric element 103 contracts, so that the fixed mirror 14 can be tilted. If the piezoelectric element 103 having a length of about 5 mm is used and the opposing piezoelectric elements 103 and 103 are arranged at intervals of 2 mm, tilt adjustment of about ± 4 minutes can be performed.

  The spectroscopic analyzer measures the tilt amount (light tilt amount between two optical paths) in the initial state before the measurement by the above-described method, and adjusts each piezoelectric so that the tilt amount becomes a level of about ± 0.2 minutes. The tilt of the fixed mirror 14 is adjusted by expanding and contracting the element 103. Thereafter, the movable mirror 15 is driven to translate (resonate), and the tilt error is monitored in real time, and the fixed mirror 14 is driven non-resonantly by the expansion and contraction of each piezoelectric element 103, and servo-driven so that the tilt error is always zero. .

  FIG. 27 is a block diagram showing feedback control (servo drive) performed when the fixed mirror 14 is non-resonantly driven. As shown in the figure, the optical path correction device 102 includes a PID controller 105 in addition to the piezoelectric element 103 described above. The PID controller 105 is a controller that performs control combining a proportional operation, an integral operation, and a differential operation. The PID controller 105 sets the tilt amount (tilt error) detected by the signal processing unit 101 and the servo target value (tilt amount zero). Based on this, the application of voltage to each piezoelectric element 103 is controlled so that the tilt amount approaches the target value (always becomes zero), and the tilt of the fixed mirror 14 is caused by the expansion and contraction of each piezoelectric element 103 by non-resonant driving. Adjust. By such feedback control, even when the tilt error fluctuates according to the movement of the movable mirror 15, the tilt error can be corrected appropriately. Note that such PID control is performed in each of the pitch and roll directions in this embodiment.

  It should be noted that the Pitch and Roll axes in the second photodetector 24 (four-divided sensor) and the Pitch and Roll axes in the tilt correction of the fixed mirror 14 are mechanically adjusted. Since it is not easy, in reality, the former axis is converted to the latter axis by using axis conversion calculation means such as a microcomputer.

  By the way, the optical path correction device 102 described above further includes an attachment frame 106 as shown in FIG. The attachment frame 106 is an attachment member for attaching the optical path correction device 102 to the interferometer 1, and is provided on the fixed base 104 so as to surround the fixed mirror 14 from the side. Each of the four surfaces constituting the attachment frame 106 is provided with a plurality of holes 106a for position adjustment when attached to the interferometer 1. The diameter of the hole 106a is slightly larger than the diameter of the screw inserted into the hole 106a. Therefore, when the mounting frame 106 is screwed to the interferometer 1, the position of the mounting frame 106 can be finely adjusted.

  Thereby, the attachment position of the optical path correction device 102 to the interferometer 1, that is, the position of the fixed mirror 14 can be finely adjusted, and initial tilt correction can be performed before product shipment. For example, if a tilt error of ± 15 minutes occurs due to accumulation of assembly errors in each block at the time of product shipment, the tilt error of the movable mirror 15 is ± 1 due to the mechanical adjustment of the initial mounting position described above. It can be adjusted to a minute level.

(4-2. Other configurations of optical path correction device)
FIG. 28A is a plan view showing another configuration of the optical path correction apparatus 102, and FIG. 28B is a side view of the optical path correction apparatus 102. In FIG. 28A, the fixed mirror 14 is not shown for convenience. The optical path correction device 102 shown in FIGS. 28A and 28B can also be applied to the interferometer 1 and the spectroscopic analysis device.

  The optical path correction device 102 includes a rotating member 111 that rotates while supporting the fixed mirror 14. The rotating member 111 is composed of a mirror support having a shape in which two cylindrical portions 111a and 111b made of metal such as stainless steel and having different diameters are coaxial and connected by an axis passing through the rotation center P. Yes. The diameter of the large-diameter column portion 111a is substantially the same as the diameter of the fixed mirror 14 (here, a disc shape), and the fixed mirror 14 is supported on the end surface of the column portion 111a opposite to the column portion 111b. ing. On the other hand, four quadrangular columnar piezoelectric elements 112 (112a to 112d) arranged in a small gap on the end surface of the small diameter cylindrical portion 111b opposite to the large diameter cylindrical portion 111a are arranged. It is fixed. The radius of the small-diameter cylindrical portion 111 b is set to approximately half the length of one side of the cross section of the piezoelectric element 112.

The piezoelectric element 112 is a displacement member that expands and contracts in a direction corresponding to the direction in which the rotating member 111 rotates by voltage application. The piezoelectric element 112, stretch direction only in some areas 112S ₁ vertical end surface 112S, which is connected to the pivot member 111 through an adhesive 113 (see FIG. 30). This division, the region of the rotating member 111 to the end face 112S and the counter of the piezoelectric element 112, first a region 111R ₁ closer to the rotation center P, on the farther second region 111R ₂ from the rotation center P when the piezoelectric element 112, during extension by voltage application in some areas 112S ₁ of the end face 112S, to press the first region 111R ₁ of the rotating member 111 via the adhesive 113, the rotating member 111 can be rotated.

Here, FIG. 29 is a bottom view of the rotating member 111 from the piezoelectric element 112 side. It is described in more detail first region 111R ₁ and the second region 111R _2. At the pivot member 111 and the end face 112S and the region opposed to the piezoelectric element 112 and 111R, the first region 111R ₁ is of the areas 111R, located at the end face of the small-diameter cylindrical portion 111b of the rotating member 111 This is an area that is closer to the rotation center P of the rotation member 111 (see FIG. 28B) than the center axis L along the expansion / contraction direction passing through the center of the end face 112S of the piezoelectric element 112. On the other hand, the second region 111R _2, the remaining areas of the region 111R, that is, a region located on the end surface of the large diameter of the cylindrical portion 111a. In this way, the area 111R is first a region 111R ₁ closer to the rotation center P, it is divided into a farther second region 111R ₂ from the rotation center P.

  In addition, a plurality of (four in this embodiment) piezoelectric elements 112 are provided as described above in order to press different areas of the rotating member 111. The expansion / contraction direction of each of the piezoelectric elements 112a to 112d corresponds to the rotation direction of the rotation member 111, but in FIG. 28B, for example, the expansion / contraction direction (BB ′) of the piezoelectric elements 112a and 112c. Direction) and the rotation direction (AA ′ direction) of the rotation member 111 correspond to each other. The piezoelectric element 112 a and the piezoelectric element 112 c, and the piezoelectric element 112 b and the piezoelectric element 112 d are disposed so as to face an axis passing through the rotation center P of the rotation member 111. Each piezoelectric element 112 is fixed to a fixed base 114 made of a metal such as stainless steel via an epoxy adhesive on the entire end surface opposite to the rotating member 111.

The manner of pressing the rotating member 111 by the piezoelectric element 112 described above can be said to be common to all the piezoelectric elements 112a to 112d. Therefore, the first region 111R ₁ and the second region 111R _second rotating member 111 described above is provided corresponding to the piezoelectric elements 112 a to 112 d, the piezoelectric elements 112 a to 112 d, at the time of elongation, adhesive 113 through, thus rotating the rotating member 111 by pressing the first region 111R ₁ corresponding.

  The adhesive 113 is a connecting member that bonds and connects each piezoelectric element 112 and the rotating member 111. A connecting portion (connecting portion) by the adhesive 113 is an important portion for determining the characteristics of the optical path correction device 102. In this embodiment, an epoxy adhesive having a relatively high Young's modulus or an appropriate elasticity is used as the adhesive. Epoxy / modified silicone adhesives having the above are appropriately selected according to specifications. Further, since the thickness of the adhesive 113 is also important, it is desirable that spherical plastic beads having a uniform diameter are mixed with the adhesive 113. In this embodiment, as the adhesive 113, an epoxy adhesive mixed with plastic beads having a diameter of 30 μm was used.

  Each piezoelectric element 112 described above is connected to a voltage applying unit (not shown), and the piezoelectric elements 112a to 112d are individually controlled by individually controlling the voltage application to the piezoelectric elements 112a to 112d. It is possible to extend and contract.

  FIG. 30 is a side view of the optical path correction device 102 and shows the posture of the rotating member 111 before and after rotation. In FIG. 30, the piezoelectric element 112b is not shown for convenience. A voltage of + v (V) is applied to the piezoelectric element 112a, and the piezoelectric element 112a extends by d (mm) (displacement of d (+)). On the other hand, a voltage of −v (V) is applied to the piezoelectric element 112c, and the piezoelectric element 112c is contracted by d (mm) (displacement of d (−)). As a result, the piezoelectric element 112a presses the cylindrical portion 111b of the rotating member 111 via the adhesive 113, and the entire rotating member 111 supports the fixed mirror 14, and the center of rotation P is the center of FIG. It is rotated by a rotation angle θ (°) around the y-axis shown in (a). Note that the rotation center P when the rotation member 111 rotates is determined by the mechanical relationship of the constituent members.

  On the other hand, when the fixed mirror 14 is rotated about the x axis shown in FIG. 28A, a voltage whose phase is shifted by 180 ° may be applied to the piezoelectric element 112b and the piezoelectric element 112d. Thereby, the rotation member 111 can be rotated around the rotation axis P around the rotation center P with the fixed mirror 14 supported by the same principle as that around the y axis.

Here, the case where the fixed mirror 14 and the rotating member 111 are rotated about the y-axis and the x-axis is shown. However, by combining these rotations about the y-axis and the x-axis, any swing motion can be performed. It is possible to realize. Thereby, tilt correction can be performed appropriately. For example, using four piezoelectric elements 112 having a cross section of 1.65 mm × 1.65 mm, the radius of the small diameter cylindrical portion 111b, that is, the distance from the axis passing through the rotation center P to the circumference of the _first region 111R1 When the fixed mirror 14 is driven in a non-resonant manner with 0.7 mm, tilt correction of ± 7 minutes can be performed with a small configuration.

[5. Supplement)
In the above, an example in which the drive mechanism 18 for driving the movable mirror 15 is configured by a parallel leaf spring mechanism using a piezoelectric element has been described. However, for example, an electromagnetic drive mechanism using, for example, an MM (moving magnet) is used. It is also possible to configure and move the movable mirror 18 by resonance.

  In the above, the tilt correction by the optical path correction device 102 is performed on the fixed mirror 14, but it can also be performed on the movable mirror 15, and both the fixed mirror 14 and the movable mirror 15 are performed. It is also possible to do this. For example, if one end face of each piezoelectric element 103 of the optical path correction device 102 is connected to the base of the drive mechanism 18, the tilt correction is performed by driving the movable mirror 15 in a non-resonant manner by the expansion and contraction of each piezoelectric element 103. Is possible.

[6. (Summary)
As described above, in the interferometer 1 of the present embodiment, the driving of the movable mirror 15 by the driving mechanism 18 is resonance driving, while the driving of the fixed mirror 14 (or the moving mirror 15) by the tilt correction unit 100 is non-driven. Resonant drive. By the non-resonant driving of the fixed mirror 14 (or the movable mirror 15) by the tilt correction unit 100, it is possible to avoid the decrease in the contrast of the first interference light due to the tilt of the light (tilt error) between the two optical paths. The first interference light can be measured (detected) with high sensitivity using light having a small allowable tilt error and a large beam diameter.

  Further, since the drive of the movable mirror 15 by the drive mechanism 18 is resonance drive, a large amount of movement of the movable mirror 15 can be secured and high resolution can be realized with a small configuration. Moreover, since the drive mechanism 18 is small, the power consumption when driving the movable mirror 15 can be reduced.

  Therefore, according to the configuration of the present embodiment, it is possible to perform high-sensitivity measurement with a low tilt while realizing high resolution with a high stroke of the movable mirror 15 with a small size and low power consumption configuration. In addition, the resonance driving of the movable mirror 15 enables the movable mirror 15 to be driven at a high speed, shortening the measurement time, and being strong against disturbances (external vibrations, etc.).

  Further, the drive for correcting the tilt error by the tilt correction unit 100 is a servo drive that adjusts the tilt correction amount by feedback control while detecting the tilt error, so that the tilt error corresponds to the movement of the movable mirror 15. Even when it fluctuates, the tilt error can be corrected appropriately.

  In addition, the tilt correcting unit 100 detects the position of the movable mirror 15 and detects a tilt error based on the detection result of the second interference light by the second photodetector 24. In this way, by using the second optical system 20 for detecting the position of the movable mirror 15 to detect the tilt error, it is not necessary to separately use a dedicated optical system for detecting the tilt error. The interferometer 1 can be reduced in size.

  In addition, since the tilt correction unit 100 performs non-resonant driving for correcting the tilt error on the fixed mirror 14, the configuration is complicated compared to the case where the non-resonant driving is performed on the movable mirror 15. The tilt error can be corrected without any problem.

  Further, since the tilt correction unit 100 adjusts the tilt of the fixed mirror 14 by the expansion and contraction of the piezoelectric element 103 (or the piezoelectric element 112), it is possible to easily and reliably correct the tilt error.

The slope correction unit 100 has the configuration having a piezoelectric element 112, when extension of the piezoelectric element 112, by pressing the pivot member 111 which is connected to the partial region 112S ₁ of the end face 112S of the piezoelectric element 112 rotates By doing so, the inclination of the fixed mirror 14 supported by the rotating member 111 is adjusted.

In this configuration, since a part of the region 112S ₁ of the end face 112S of the piezoelectric element 112 and the rotating member 111 is connected, as compared to the structure sip connecting the whole and the pivot member 111 of the end face 112S, the apparent The position Q of the upper force point can be made closer to the rotation center P side than the center of the end face 112S of the piezoelectric element 112 (see FIG. 30). The apparent power point position Q is the position of the power point when the piezoelectric element 112 is considered to be equivalent to pressing the rotating member 111 at one point via the adhesive 113 when the piezoelectric element 112 is extended. Strictly speaking, the adhesive 113 deviates from the center of the contact surface with the rotating member 111, but for convenience, it can be considered as the center of the contact surface. Thereby, even if the extension amount of the piezoelectric element 112 is the same, the rotation member 111 can be pressed and rotated greatly, and the tilt amount of the fixed mirror 14 can be adjusted greatly with a small configuration.

Further, the rotating member 111 is a part of the region 112S ₁ of the end surface 112S of the piezoelectric element 112 on the rotation center P side with respect to the central axis L along the expansion / contraction direction passing through the center of the end surface 112S of the piezoelectric element 112. Therefore, the effect that the tilt amount of the fixed mirror 14 can be largely adjusted with the above-described small configuration can be surely obtained.

  In addition, since the inclination correction unit 100 includes a plurality of piezoelectric elements 103 (or piezoelectric elements 112), the inclination of the fixed mirror 14 can be adjusted in an arbitrary direction by selecting the piezoelectric element to be expanded. And tilt error can be corrected reliably.

  Further, in the drive mechanism 18 described above, the two leaf spring portions 31 and 32 are arranged in parallel via the rigid bodies 33 and 34, and the leaf spring portion 31 is resonated by the expansion and contraction of the piezoelectric element 35a so that the movable mirror 15 is made incident light. By adopting a configuration in which the movable mirror 15 is moved in the optical axis direction, the resonance drive of the movable mirror 15 can be reliably realized.

  In this embodiment, a case has been described in which light is applied to a sample outside the interferometer and light obtained through the sample is incident on the interferometer to perform spectroscopic analysis. For example, from the outside of the interferometer, Even if interference light is generated by the interferometer using the introduced light and the interference light is applied to the sample for spectroscopic analysis, or light incident from the outside of the interferometer itself is analyzed, this It is possible to apply the interferometer of the invention. If a measurement light source is arranged instead of the measurement light input unit 11 shown in the present embodiment, the measurement light source can be analyzed. Further, the interferometer may include a measurement light source and irradiate the sample with measurement light emitted from the measurement light source. Therefore, the measurement light for obtaining the first interference light may be light emitted from a light source built in the interferometer, or may be light incident from the outside of the interferometer. It can be said.

  As described above, the interferometer described in the present embodiment separates the measurement light by the beam splitter and guides it to the movable mirror and the fixed mirror, and each reflected light from the movable mirror and the fixed mirror is transmitted to the beam splitter. An interferometer including an optical system that guides the combined interference light to the photodetector, and a movable mirror drive mechanism that moves the movable mirror in the optical axis direction of incident light, and driving by the movable mirror drive mechanism At least one of the movable mirror and the fixed mirror is driven to correct the relative tilt between the reflected light from the movable mirror and the reflected light from the fixed mirror, which is caused by the tilt of the movable mirror at the time. The movable mirror driving mechanism is driven by a resonance drive, while the tilt correction unit is driven non-resonant by at least one of the movable mirror and the fixed mirror. Drive That.

  According to the above configuration, the measurement light is separated by the beam splitter and guided to the movable mirror and the fixed mirror. The reflected lights from the movable mirror and the fixed mirror are combined by a beam splitter and guided to a photodetector as interference light. At this time, the moving mirror is moved (translated) in the optical axis direction of the incident light by the moving mirror driving mechanism, so that the optical path difference is changed between the reflected light from the moving mirror and the reflected light from the fixed mirror. The intensity of the interference light changes. Therefore, when the interferometer of the present embodiment is applied to, for example, a Fourier transform spectroscopic analyzer, the wave number is calculated from this spectrum distribution by Fourier transforming the interference light (interferogram) to obtain the spectrum distribution of the incident light. The intensity of the interference light for each (1 / wavelength) can be obtained.

  Here, if the amount of movement of the movable mirror is large, it becomes difficult to translate the movable mirror, and the movable mirror is tilted, and the relative tilt between the reflected light from the movable mirror and the reflected light from the fixed mirror ( The inclination of light between the two optical paths occurs. However, since the tilt correction unit performs driving for correcting the tilt of light between the two optical paths with respect to at least one of the movable mirror and the fixed mirror, and the driving is non-resonant driving, the tilt correction unit Therefore, it is possible to avoid a decrease in the contrast of the interference light. Accordingly, interference light can be measured (detected) with high sensitivity using light having a small allowable light inclination between the two optical paths and a large light beam diameter.

  In addition, since the movable mirror is driven by the movable mirror drive mechanism by resonance, a large amount of movement can be secured and high resolution can be achieved with a small configuration. In addition, since the moving mirror driving mechanism is small, power consumption during driving of the moving mirror can be reduced.

  That is, according to the above configuration, it is possible to perform high-sensitivity measurement with low tilt while realizing high resolution with a high stroke of the movable mirror with a small size and low power consumption configuration.

  The light source that emits the measurement light may be inside or outside the interferometer. That is, the measurement light for obtaining the interference light by the interferometer may be light emitted from a light source built in the interferometer, or may be light incident from the outside of the interferometer. Good.

  In the interferometer of this embodiment, the tilt correction unit is driven by feedback control while detecting the relative tilt between the reflected light from the movable mirror and the reflected light from the fixed mirror. It is desirable that the servo drive adjusts the angle.

  By such servo drive, the tilt can be appropriately corrected even when the tilt of the light between the two optical paths varies according to the movement of the movable mirror.

  The interferometer of this embodiment includes a reference light source when the interference light, the photodetector, and the optical system are a first interference light, a first photodetector, and a first optical system, respectively. The light from the reference light source is separated by the beam splitter and guided to the movable mirror and the fixed mirror, and the reflected light from the movable mirror and the fixed mirror is synthesized by the beam splitter. A second optical system that guides the interference light to a second photodetector, and the tilt correction unit moves the movement based on a detection result of the second interference light by the second photodetector. While detecting the position of a mirror, the structure which detects the relative inclination of the reflected light in the said movable mirror and the reflected light in the said fixed mirror may be sufficient.

  In this configuration, the tilt correction unit also uses the second optical system for detecting the position of the movable mirror (for detecting the optical path difference between the movable mirror side and the fixed mirror side), and also calculates the tilt of light between the two optical paths. Therefore, it is not necessary to separately use a dedicated optical system for detecting the inclination of light between the two optical paths, and the interferometer can be downsized.

  In the interferometer of the present embodiment, it is desirable that the second photodetector is constituted by a four-divided sensor.

  In this configuration, the tilt correction unit is configured to tilt the other light with respect to one light (tilt direction and tilt amount (angle difference)) based on a signal from each region of the quadrant sensor as the second photodetector. Can be reliably detected.

  In the interferometer of this embodiment, it is preferable that non-resonant driving by the tilt correction unit is performed on the fixed mirror.

  The tilt correction unit performs non-resonant driving on a fixed mirror that is different from the resonance-driven movable mirror, thereby correcting the light tilt between the two optical paths without complicating the configuration. .

  In the interferometer of the present embodiment, the tilt correction unit may be configured to adjust the tilt of the fixed mirror by expansion and contraction of a piezoelectric element.

  In this configuration, the inclination of the fixed mirror can be easily adjusted by the expansion and contraction of the piezoelectric element, and the light inclination between the two optical paths can be corrected.

  In the interferometer of this embodiment, the tilt correction unit is supported by the support base by pressing and rotating a support base connected to a part of the end face of the piezoelectric element when the piezoelectric element is extended. It is also possible to adjust the tilt of the fixed mirror.

  In this configuration, since a part of the end face of the piezoelectric element is connected to the support base, the support base is pressed even with the same extension amount of the piezoelectric element as compared to the configuration in which the entire end face is connected to the support base. Thus, the tilt amount of the fixed mirror can be greatly adjusted with a small configuration.

  In the interferometer according to the present embodiment, the support base is connected to a part of the end face of the piezoelectric element on the rotation center side with respect to the central axis along the expansion / contraction direction passing through the center of the end face of the piezoelectric element. It is desirable.

  With this configuration, the support base can be greatly rotated with a small extension amount of the piezoelectric element, and the effect that the tilt amount of the fixed mirror can be largely adjusted with a small configuration can be reliably obtained.

  In the interferometer of this embodiment, it is preferable that a plurality of the piezoelectric elements are provided.

  In this configuration, the tilt of the fixed mirror can be adjusted in an arbitrary direction by selecting the piezoelectric element to be extended.

  In the interferometer of this embodiment, the movable mirror drive mechanism has two leaf springs arranged in parallel via a rigid body, and a piezoelectric element arranged on one of the two leaf spring portions. The movable mirror may be disposed on one of the two leaf springs and may move in the optical axis direction of incident light by resonance of the leaf spring due to expansion and contraction of the piezoelectric element.

  In this way, the two leaf springs are arranged in parallel via a rigid body, and the movable mirror is moved by resonating the leaf springs by expansion and contraction of the piezoelectric element, so that the movable mirror is driven by the movable mirror drive mechanism. It can be realized reliably.

  The Fourier transform spectroscopic analyzer of the present embodiment may be configured to include the above-described interferometer of the present embodiment and an arithmetic unit that Fourier transforms an interferogram obtained by the interferometer.

  According to the interferometer of the present embodiment, even if the translational property of the moving mirror is lost when the moving amount of the moving mirror is increased, the inclination correction unit corrects the inclination of light between the two optical paths (first (1) The contrast of the (first) interference light detected by the photodetector can be suppressed from decreasing. Therefore, in the Fourier transform spectroscopic analysis apparatus provided with the interferometer of this embodiment, spectroscopic analysis based on the spectrum obtained by the Fourier transform in the calculation unit can be performed with high accuracy. That is, a high-performance Fourier transform spectroscopic analyzer can be realized.

  The present invention can be used in a Michelson interferometer and a Fourier transform spectroscopic analyzer that performs spectroscopic analysis using the interferometer.

DESCRIPTION OF SYMBOLS 1 Interferometer 2 Calculation part 10 1st optical system 11 Measurement light input part 13 BS (beam splitter)
14 Fixed Mirror 15 Moving Mirror 17 First Photodetector 18 Drive Mechanism (Moving Mirror Drive Mechanism)
20 2nd optical system 21 Reference light source 24 2nd photodetector (4-part dividing sensor)
Reference Signs List 31 leaf spring part 32 leaf spring part 33 rigid body 34 rigid body 35a piezoelectric element 100 tilt correction unit 101 signal processing unit (tilt correction unit)
102 Optical path correction device (tilt correction unit)
103 Piezoelectric element 111 Rotating member (support)
112 Piezoelectric element

Claims (12)

Moving and fixed mirrors,
A beam splitter that separates measurement light and guides it to the movable mirror and the fixed mirror, and combines the light reflected by the movable mirror and the fixed mirror;
An interferometer including an optical system including a photodetector that detects interference light obtained by combining the light reflected by the movable mirror and the fixed mirror by the beam splitter;
A movable mirror drive mechanism for moving the movable mirror in the optical axis direction of incident light;
Drive for correcting the relative tilt between the reflected light from the movable mirror and the reflected light from the fixed mirror, which is caused by the tilt of the movable mirror during driving by the movable mirror drive mechanism, An inclination correction unit that performs at least one of the fixed mirrors,
The drive of the movable mirror by the movable mirror drive mechanism is resonant drive, while the drive to at least one of the movable mirror and the fixed mirror by the tilt correction unit is non-resonant drive ,
The moving mirror drive mechanism has two leaf springs arranged in parallel via a rigid body,
The interferometer is characterized in that the movable mirror is disposed on one of the two leaf spring portions and moves in the optical axis direction of incident light by resonance of the leaf spring portion. The movable mirror driving mechanism has a piezoelectric element disposed on one of the two leaf springs, and resonates the movable mirror by causing the leaf spring to resonate by expansion and contraction of the piezoelectric element. The interferometer according to claim 1. The interferometer according to claim 1, wherein the movable mirror driving mechanism resonates and drives the movable mirror by causing the leaf spring portion to resonate with an electromagnetic driving mechanism. The drive by the tilt correction unit is a servo drive that adjusts the correction amount of the tilt by feedback control while detecting the relative tilt between the reflected light from the movable mirror and the reflected light from the fixed mirror. The interferometer according to any one of claims 1 to 3. When the interference light, the photodetector, and the optical system are respectively a first interference light, a first photodetector, and a first optical system,
A reference light source is provided, the light from the reference light source is separated by the beam splitter and guided to the movable mirror and the fixed mirror, and each reflected light from the movable mirror and the fixed mirror is synthesized by the beam splitter. A second optical system for guiding the second interference light thus formed to the second photodetector;
The tilt correction unit detects the position of the movable mirror based on the detection result of the second interference light by the second photodetector, and reflects the reflected light from the movable mirror and the fixed mirror. The interferometer according to claim 4, wherein a relative inclination with respect to the reflected light is detected. The interferometer according to claim 5, wherein the second photodetector is constituted by a quadrant sensor. The interferometer according to any one of claims 1 to 6, wherein the non-resonant drive by the tilt correction unit is performed on the fixed mirror. The interferometer according to claim 7, wherein the tilt correction unit adjusts the tilt of the fixed mirror by expansion and contraction of a piezoelectric element. The tilt correction unit presses and rotates a support base connected to a part of the end face of the piezoelectric element when the piezoelectric element is extended, thereby tilting the fixed mirror supported by the support base. The interferometer according to claim 8, wherein the interferometer is adjusted. 10. The support base is connected to a part of the end face of the piezoelectric element at a rotation center side with respect to a center axis along a stretching direction passing through the center of the end face of the piezoelectric element. The interferometer described in 1. The interferometer according to claim 8, wherein a plurality of the piezoelectric elements are provided. An interferometer according to any of claims 1 to 11,
A Fourier transform spectroscopic analyzer, comprising: an arithmetic unit that Fourier transforms an interferogram obtained by the interferometer.

Images