JP2005309099A - Tunable filter and method of manufacturing the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a tunable filter which is operated at a lower voltage, and also to provide a method of manufacturing the tunable filter. <P>SOLUTION: The tunable filter comprises: a movable part 2 in which a movable body 21a, which has a movable reflection face, passes light having a predetermined wavelength and reflects light having a wavelength other than the predetermined wavelength by being displaced in the direction perpendicular to the movable reflection face, a connecting part 21b and a supporting part 21c which movably support the movable body 21a, and a movable comb-shaped electrode 21d which generates electrostatic force which moves the movable body 21a and moves together with the movable body 21a, are integrally formed; a fixed electrode part 1 which is provided having a first gap with the movable comb-shaped electrode 21d and formed a fixed comb-shaped electrode 12 which generates electrostatic force between the movable comb-shaped electrode 21d; and a fixed comb-shaped side substrate part 3 which has a fixed reflection face which is provided opposite to the movable reflection face with an optical gap OG and further reflects the light reflected on the movable reflection face. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a plurality of lights having different wavelengths that are transmitted through an optical fiber in a wavelength division multiplexing (WDM) optical communication network or the like (light here is not limited to visible light. The same applies hereinafter). ), A wavelength tunable filter that selectively transmits light to extract light having a desired wavelength, and a method of manufacturing the same.

  A conventional wavelength tunable filter uses the principle of a Fabry-Perot interferometer. The movable reflective surface of the movable body is disposed opposite to the fixed reflective surface on the substrate in parallel with a predetermined interval (hereinafter, the interval between the fixed reflective surface and the movable reflective surface is referred to as an optical gap). The movable body is suspended by a member having elasticity (flexibility). By applying a voltage between the movable electrode provided on the movable reflective surface and the fixed electrode provided on the fixed reflective surface (hereinafter, the interval between the fixed electrode and the movable electrode is referred to as an electrostatic gap), the movable body is Displacement (drive) causes the movable reflective surface to be displaced relative to the fixed reflective surface. This electrostatic gap is formed by providing a sacrificial layer having a predetermined shape and size between the fixed reflecting surface and the movable reflecting surface using a micromachining technique, and then etching all or one of the sacrificial layers by etching. It is formed by removing the portion (see, for example, Patent Document 1).

In addition, some conventional wavelength tunable filters use the silicon dioxide (SiO ₂ ) layer of a SOI (Silicon on Insulator) substrate (wafer) as a sacrificial layer to form the electrostatic gap (for example, Patent Document 2).
JP 2002-174721 A ([Claim 9], [0005], [0018], [0037], [0049] to [0056], FIG. 6) US Pat. No. 6341039 (columns 6-7, FIGS. 4A-4I)

  In the wavelength tunable filter, a drive voltage is applied to a parallel plate capacitor formed by a movable electrode provided on the movable reflective surface and a fixed electrode provided on the fixed reflective surface, thereby electrostatically connecting between the movable electrode and the fixed electrode. Attracting force is generated, the movable reflective surface is changed with respect to the fixed reflective surface, and light of different wavelengths is taken out at different timings (not limited to wavelength tunable filters, the method using a parallel plate is generally a parallel plate type Called).

  Here, in such a relationship between the movable electrode and the fixed electrode, when the electrostatic force becomes larger than the restoring force of the elastic member that tries to return the movable electrode side to the original position, the movable electrode rapidly changes to the fixed electrode. A phenomenon called pull-in that is attracted to the image occurs. When the movable reflecting surface (movable electrode) is displaced by the above-described method, it is displaced about 1/3 or more of the distance of the electrostatic gap at the time of non-displacement (the distance between the fixed electrode and the movable electrode is not displaced). Pull-in occurs when it is less than about 2/3 of the time).

  In many cases, it is important for the wavelength tunable filter to output light of a certain wavelength at a timing at which a signal (data) is included. For this purpose, the timing of displacement of the movable reflecting surface (movable electrode) is important. In order to stabilize the displacement of the movable reflecting surface, the displacement is performed before the pull-in occurs. However, in order to perform a necessary displacement at a stable portion before pull-in occurs, a distance about three times the displacement distance is required as an electrostatic gap. In such a case, since the electrostatic force is inversely proportional to the square of the distance, the applied voltage must be increased and the power consumption increases as the distance between the electrodes (electrostatic gap) increases.

  In order to solve the above-described problems, an object of the present invention is to obtain a wavelength tunable optical filter that can be driven at a lower voltage and a method for manufacturing the same.

  The wavelength tunable filter according to the present invention has a movable reflection surface, and is displaced in a direction perpendicular to the movable reflection surface, thereby transmitting light of a predetermined wavelength and reflecting light of a wavelength other than the light of the predetermined wavelength. Movable body to be movable, support body that is movable and displaceable, movable part that generates electrostatic force that displaces movable body, and movable comb electrode that is displaced together with movable body, and movable comb teeth A fixed electrode portion provided with a fixed comb-teeth electrode provided with an electrode and a first gap and generating an electrostatic force between the movable comb-teeth electrode, a movable reflective surface, and a second gap; And a fixed reflection portion having a fixed reflection surface that further reflects light reflected by the movable reflection surface. In the present invention, a movable comb electrode is provided on the movable part side, and a fixed comb electrode is provided on the fixed electrode part side to generate an electrostatic force with a comb structure. The ratio of the distance up to can be made larger than in the case of the parallel plate type. Therefore, since the distance between the electrodes can be shortened compared to the parallel plate type, when the same electrostatic force is obtained, the drive voltage can be lowered compared to the parallel plate type, and power saving can be achieved. A tunable filter can be obtained (when the voltage is the same, the electrostatic force can be increased).

  The movable part of the wavelength tunable filter according to the present invention is formed by processing the active layer of the SOI substrate in which the active layer, the insulating layer, and the base layer are sequentially stacked, and the fixed electrode part is processed by the base layer of the SOI substrate. Is formed. Therefore, the fixed electrode portion and the movable portion can be integrally formed. In addition, since the active layer has a specularity, a movable part with good optical characteristics can be formed.

  In the wavelength tunable filter according to the present invention, the substrate constituting the fixed reflection portion is bonded to the fixed electrode portion. Accordingly, a space for reflecting light between the movable reflecting surface and the fixed reflecting surface can be formed using the fixed electrode portion as a spacer, and the process can be simplified and the degree of freedom can be increased.

  In the wavelength tunable filter according to the present invention, the substrate constituting the fixed reflection portion has a recess having the inner wall bottom surface portion as the fixed reflection surface, and is joined to the movable portion. Therefore, a space for reflecting light between the movable reflecting surface and the fixed reflecting surface can be formed by the side wall portion of the recess. Therefore, if the height (thickness) of the side wall portion of the recess is accurately adjusted, an optical wavelength filter with good optical characteristics can be obtained. If the substrate is a glass substrate, the processing conditions and process control can be made strict, so that the recess can be formed with higher accuracy.

  In the wavelength tunable filter according to the present invention, the substrate constituting the fixed reflection portion is a glass substrate that can be anodically bonded to the fixed electrode portion or the movable portion. If anodic bonding is performed, it is possible to perform strong bonding with high accuracy without separation later.

  In the wavelength tunable filter according to the present invention, the substrate constituting the fixed reflection portion is a substrate that can be surface-activated bonded to the fixed electrode portion or the movable portion. If surface activated bonding is performed, it is possible to perform high-precision bonding with high accuracy without being separated later.

In the wavelength tunable filter according to the present invention, a piezoresistive element whose resistivity changes based on stress is provided at a connection portion between the support and the movable body.
Therefore, by measuring the resistivity, it is possible to feedback control the displacement of the movable body so that the movable reflective surface and the fixed reflective surface are displaced in parallel, so that a wavelength tunable filter with good optical characteristics is obtained. be able to.

In the wavelength tunable filter according to the present invention, at least one of the movable comb electrode and the fixed comb electrode is covered with an insulating film.
Therefore, the movable comb electrode and the movable comb electrode can be prevented from being short-circuited and stuck (sticking), and the life can be extended.

  In the method for manufacturing a wavelength tunable filter according to the present invention, the movable body is displaced in the vertical direction with respect to the fixed reflection surface by electrostatic force, and the movable reflection surface is provided between the fixed reflection surface and the movable reflection surface provided on the movable body. A method of manufacturing a wavelength tunable filter that transmits light having a predetermined wavelength based on a distance between a fixed reflecting surface and a movable reflecting surface from reflected light, wherein an active layer, an insulating layer, and a base layer are sequentially stacked A first step of forming at least a fixed comb electrode on the base layer of the substrate, and a movable comb electrode and a movable electrode that are displaced together with the movable body by etching the active layer to generate an electrostatic force that displaces the movable body and the movable body A second step of forming a support for supporting the body and removing a predetermined portion of the insulating layer; and a third step of bonding the substrate having the fixed reflective surface and the laminated substrate. Since the wavelength tunable filter having the fixed comb electrode and the working comb electrode can be manufactured in this way, the ratio of the distance until the pull-in occurs is larger than the parallel plate type with respect to the distance between the electrodes. can do. Therefore, since the distance between the electrodes can be shortened compared to the parallel plate type, when the same electrostatic force is obtained, the drive voltage can be lowered compared to the parallel plate type, and power saving can be achieved.

  In the method for manufacturing a wavelength tunable filter according to the present invention, the substrate is etched or polished to a required thickness before the first step. Therefore, the process time can be shortened.

Embodiment 1 FIG.
FIG. 1 is a sectional view showing a wavelength tunable filter according to the first embodiment of the present invention. FIG. 1 is a cross-sectional view at a substantially central position of the wavelength tunable filter (see the AA ′ cross section in FIG. 2). The wavelength tunable filter according to the present embodiment includes a fixed electrode portion 1, a movable portion 2, and a fixed comb-side substrate portion 3. Here, in the present embodiment, the fixed electrode portion 1 and the movable portion 2 are integrally formed by processing, for example, an SOI substrate 100 described later. In the present embodiment, the upper side in FIG. 1 is referred to as the upper side, and the lower side is referred to as the lower side. Moreover, in each figure, in order to make each structural member easy to see, the thickness of each member and the ratio in relation to other members may be different from actual ones.

  The fixed electrode portion 1 includes a substrate portion 11, a fixed comb electrode 12, an electrode lead portion 13, and an insulating portion 14. The substrate unit 11 is formed by processing the base layer 101 of the SOI substrate 100 as will be described later. For example, light from an optical fiber (not shown) is incident on the opening 11a of the substrate portion 11, and a space (gap) for repeating reflection of light incident between the highly reflective films 23 and 32 described later. Is provided to form The substrate portion 11 also serves as a spacer for that purpose. Here, the highly reflective films 23 and 32 are provided with an interval of an optical gap OG (for example, about 30 μm). Since the insulating film 14 described later has a thickness of about 4 μm, for example, the substrate portion 11 has a thickness of about 26 μm.

  The fixed comb electrode 12 is provided with a predetermined interval (first gap) from the movable comb electrode 21d formed on the movable portion 2, and is paired with the movable comb electrode 21d to form a comb. A tooth electrode structure is adopted. A potential difference is generated by the electric charge supplied via the electrode lead-out part 13 and the substrate part 11 to generate an electrostatic force (Coulomb force, electrostatic attraction) between the movable comb-tooth electrode 21d and restore the connecting part 21b and the like. Along with the force, the movable comb electrode 21d is driven in the vertical (vertical, vertical) direction to displace the position. Here, in the present embodiment, the fixed comb electrode 12 is integrally formed from the SOI substrate 100 similarly to the substrate portion 11. The electrode lead-out part 13 is an interface part composed of terminals and the like for receiving external charge supply (AC or pulse voltage application) in order to supply charges to the fixed comb electrode 12. Further, in order to insulate between the fixed electrode portion 1 and the movable portion 2, for example, an insulating portion 14 of about 4 μm is provided. In this embodiment, the insulating portion 14 is formed by processing an insulating layer 102 of the SOI substrate 100 described later. In the present embodiment, the thickness of the insulating portion 14 is substantially equal to the distance in the vertical (vertical, vertical) direction between the fixed comb electrode 12 and the movable comb electrode 21d.

  FIG. 2 is a view of the movable part 2 as viewed from above. The movable part 2 includes a movable part substrate 21, an antireflection film 22, a high reflection film 23, and an electrode terminal part 24. In the present embodiment, the movable part substrate 21 is formed of an active layer 103 of the SOI substrate 100 described later.

  The movable part substrate 21 includes a movable body part 21a, four connecting parts 21b serving as suspension means for holding the movable body 21a in a space, a support part 21c, and movable comb teeth provided on each connecting part 21b. The electrode 21d is integrally formed. Here, the thickness of the movable body 21a, the connecting portion 21b, the support portion 21c, and the movable comb electrode 21d is about 10 μm, for example, and is divided into each part by being partitioned by an open space. The movable body 21 a is formed in a substantially circular (disc) shape at the approximate center of the movable part substrate 21.

  A support body that supports the movable body 21a is configured by the four connecting portions 21b and the support portion 21c formed on the peripheral portion of the movable body 21a. Each connection part 21b has flexibility (elasticity), and it is provided in the peripheral part of the movable body 21a in the position where each adjacent thing made an angle of about 90 degree | times. The movable body 21a is supported by this support body, and in a direction perpendicular to the surface on which the high reflection film 23 is formed (the surface on which the high reflection film 18 is formed) due to the restoring force of the electrostatic force, the connecting portion 21b, and the like. The position is displaced. The movable comb electrode 21d generates the electrostatic force with the fixed comb electrode 12 of the fixed electrode portion 1. The movable comb electrode 21d is integrally formed with the connecting portion 21b. The support at this time also serves as a conductive portion for supplying the movable comb electrode 21d with a charge having a polarity opposite to that of the charge supplied to the fixed comb electrode 12 from, for example, an external electrode (not shown).

  Here, in the case of the comb-tooth type, the ratio of the stable distance until the occurrence of the pull-in described above with respect to the electrostatic gap is generally larger than that of the parallel plate type. For example, a pull-in does not occur unless the electrostatic gap is displaced by more than about 1/2 of the electrostatic gap at the time of non-displacement (the distance between the fixed side and the movable side is about ½ or less of the non-displacement). . The same applies to the case of a comb tooth type of vertical (vertical) drive in which the movable comb teeth are displaced in a direction (vertical direction) perpendicular to the tooth tip direction with respect to the fixed comb teeth as in this embodiment. is there. Therefore, the distance between the electrodes can be shortened, and compared with the parallel plate type, when the electrostatic force is the same, the applied voltage can be lowered and power saving can be achieved (even if the same voltage is used). Can increase the electrostatic force).

  Here, in this Embodiment, although the movable body 21a is formed in circular shape, a shape is not restricted to this. For example, the shape may be any shape that can be displaced vertically, such as a regular polygon shape, while the high reflection film 23 is kept parallel to the high reflection film 32. Further, regarding the shape and the number of the connecting portions 21b, a movable comb electrode 21d can be provided, the weight of the movable body 21a can be supported, and the desired movable body 21a can be displaced with a predetermined driving voltage. For example, the shape and number are not limited to those shown in the present embodiment.

The antireflection film 22 is formed on the upper surface of the movable body 21a so that the light transmitted through the movable body 21a is not reflected. The antireflection film 22 is formed of a multilayer film in which, for example, a thin film of silicon dioxide (SiO ₂ ) and a thin film of tantalum pentoxide (Ta ₂ O ₅ ) are alternately stacked using an evaporation method or the like. Alternatively, a silicon nitride (SiN _x ), silicon oxynitride (SiON) thin film, or the like can be stacked. On the other hand, the high reflection film 23 serving as a movable reflective surface is formed on the lower surface (the fixed electrode portion 1 and the fixed comb-tooth side substrate portion 3 side) of the movable body 21a. The highly reflective film 23 is formed of a multilayer film (about 2 to 40 layers) in which thin films of silicon dioxide (SiO ₂ ) and thin films of tantalum pentoxide (Ta ₂ O ₅ ) are alternately stacked using a vapor deposition method or the like. In some cases, silicon nitride (SiN) can also be used. The high reflection film 23 further reflects light incident from below (the light reflected by the high reflection film 32 included in the fixed comb-tooth side substrate unit 3), so that the reflection with the high reflection film 32 is performed a plurality of times. The interference is repeated, and only the light having a predetermined wavelength that satisfies the interference condition is transmitted. The wavelength of the light to be transmitted varies depending on the interval of the optical gap OG that changes due to the displacement of the movable body 21a. Therefore, a plurality of lights (signals) having different wavelengths are multiplexed and transmitted by including data at a predetermined timing, and the wavelength variable filter changes the interval of the optical gap OG in accordance with the timing. Can be divided and taken out. Here, by forming the highly reflective film 23 on the entire lower surface of the movable body 21a, the function can be exhibited more. Here, the antireflection film 22 and the high reflection film 23 are made of the same material, but the thickness of the thin film of each layer is different. By adjusting the film thickness, it can be a reflection film or an antireflection film for light of a predetermined wavelength. Here, in the present embodiment, the antireflection film 22 and the high reflection film 23 are formed of a multilayer film, but may be a single layer film as long as the function can be achieved. The same applies to the high reflection film 32 and the antireflection film 33 described later. Note that the high reflection film means, for example, a reflection film having a reflectance of 95% or more for light having a desired wavelength. In the multilayer film of the present embodiment, the reflectance is 98% or more.

  In the present embodiment, the fixed comb-tooth side substrate portion 3 which is a fixed reflection portion is composed of a glass substrate 31, a high reflection film 32 and an antireflection film 33. The glass substrate 31 is made of glass containing an alkali metal such as sodium (Na) or potassium (K), for example. This is because it is convenient when the fixed electrode portion 1 and the fixed comb-side substrate portion 3 are joined by anodic bonding as will be described later. Examples of this type of glass include borosilicate glass, soda glass, and potassium glass containing an alkali metal. Further, in the case of bonding by anodic bonding, the glass substrate 31 is heated, so that the glass substrate 31 is preferably substantially equal in thermal expansion coefficient to silicon that is a constituent material of the fixed electrode portion 1. As a glass that meets the above requirements, for example, there is # 7740 (trade name) manufactured by Corning, and it is preferable to use such glass for the glass substrate 31.

  The high reflection film 32 constituting the fixed reflection surface is the same film as the high reflection film 23 and is provided on the surface of the glass substrate 31 facing the movable body 21a. The antireflection film 33 is the same film as the antireflection film 22. The glass substrate 31 is provided on the surface opposite to the surface on which the highly reflective film 32 is provided.

  3 to 5 are diagrams showing manufacturing steps of the wavelength tunable filter according to the first embodiment. FIG. 3 shows a manufacturing process of the fixed electrode portion 1 including the first process. Moreover, FIG. 4 represents the manufacturing process of the movable part 2 including a 2nd process. Further, FIG. 5 shows a manufacturing process of the fixed comb-tooth side substrate portion 3 including the third process and a process up to manufacturing a wavelength tunable filter by bonding.

First, a manufacturing process of the fixed electrode portion 1 will be described with reference to FIG. In the present embodiment, the SOI substrate 100 is used to manufacture the substrate portion 11 (opening portion 11a) and the fixed comb electrode 12 of the fixed electrode portion 1 and the movable portion substrate 21 of the movable portion 2 (FIG. 3A). ). The SOI substrate 100 is composed of, for example, a base layer 101 having a thickness of about 500 μm and an active layer 103 having a thickness of about 10 μm, for example, sandwiching an insulating layer 102 having a thickness of about 4 μm. The base layer 101 and the active layer 103 are made of silicon (Si), and the insulating layer 102 is made of silicon dioxide (SiO ₂ ). Since the insulating layer 102 serves as an etching stopper, the movable part substrate 21 with particularly high thickness accuracy can be obtained for the parts that become the movable body 21a, the coupling part 21b, the support part 21c, and the movable comb electrode 21d. Moreover, since the active layer 103 has a mirror surface property, the optical characteristics of the movable body 21a can be improved.

  Here, although the thickness of the base layer 101 is about 500 μm, it is too thick for the substrate portion 11, and may be polished or etched in advance to have a predetermined thickness. In particular, when polishing or wet etching is performed, for example, a plurality of members formed on the wafer can be collectively processed, which is convenient. Thereby, the etching time of the base layer 101 can be shortened when forming the openings 11a and the like which will be described later.

  Next, a silicon oxide film 110 serving as a resist film is formed on the SOI substrate 100. As a formation method, there are various methods such as a chemical vapor deposition (CVD) method and thermal oxidation of the surface by heating. Then, using a photolithography method, a photoresist pattern for patterning the silicon oxide film 110 is formed from the applied photoresist. Then, the portion of the silicon oxide film 110 that becomes the opening 11a is removed, and the photoresist is removed. The etching pattern by the silicon oxide film 110 is left (FIG. 3B).

  Etching is then performed to form the opening 11a (FIG. 3 (3)). Here, since the insulating layer 102 serves as a stopper for stopping the progress of etching into the active layer 103, the active layer 103 is not damaged, and a wavelength tunable filter with a high yield can be manufactured. Hereinafter, the wet etching removal method and the dry etching removal method will be described. The wet etching removal method is appropriate in that removal can be performed collectively. Then, by further etching, the fixed comb electrode 12 is formed (FIG. 3 (4)).

(1) Wet etching removal method For example, a predetermined portion of the base layer 101 is immersed in a potassium hydroxide (KOH) aqueous solution having a concentration of 1 to 40% by weight (preferably around 10% by weight). Etch. Etching solutions used in this case include tetramethylammonium hydroxide (TMAH) aqueous solution, ethylenediamine-pyrocatechol-diazine (EPD) aqueous solution, and hydrazine (Hydrazine) aqueous solution. Here, when a plurality of wafers are formed, batch processing (batch processing) can be performed while making the production conditions and the like substantially equal, and productivity can be improved.

(2) Dry etching removal method For example, after placing the bonding structure in a chamber of a dry etching apparatus and making it in a vacuum state, for example, xenon difluoride (XeF ₂ ) having a pressure of 390 Pa is introduced into the chamber for 60 seconds. Thus, the base layer 101 is etched. Note that a plasma etching method using carbon tetrafluoride (CF ₄ ) or sulfur hexafluoride (SF ₆ ) can also be used.

FIG. 4 is a diagram illustrating a process of forming the movable part 2. Next, formation of the movable part 2 will be described. After removing the silicon oxide film 110, a photoresist pattern 111 is formed again using a photolithography method or the like (FIG. 4A). The substrate portion 11 (opening portion 11a) and the fixed comb electrode 12 are registered. And the hole used as the movable body 21a, the connection part 21b, the movable comb-tooth electrode 21d, and the electrode extraction part 13 is formed by anisotropic dry etching (FIG. 4 (2)). As an anisotropic dry etching method, for example, after placing the SOI substrate 100 in a chamber (container) of a dry etching apparatus, for example, sulfur hexafluoride (SF ₆ ) as an etching gas is supplied at a flow rate of 130 cm ³ / min ( (sccm) for 6 seconds, and cyclobutane octafluoride (C ₄ F ₈ ) as a deposition (deposition) gas is alternately introduced into the chamber at a flow rate of 50 cm ³ / min (sccm) for 7 seconds. Because of anisotropic dry etching, it is possible to prevent wraparound (side etching) to a portion resisted by a resist pattern. In particular, the strength of the connecting portion 21b and the movable comb electrode 21d is not impaired. After the anisotropic dry etching, the photoresist pattern is removed using, for example, oxygen plasma (FIG. 4 (3)). Then, resist or the like is performed again, and a metal film (not shown) or the like is deposited, thereby forming the electrode terminal portion 24 and the extraction electrode portion 13. The movable part 2 is produced by the process described above. Then, a predetermined portion of the insulating layer 26 is removed by, for example, a wet etching method using hydrofluoric acid (HF) to form the insulating portion 14 (FIG. 4 (4)).

And after applying a resist etc. to parts other than the part of the movable body 21a, the antireflection film 22 and the high reflection film 23 are formed on the upper surface and the lower surface of the movable body 21a, respectively. The antireflection film 22 and the high reflection film 23 are formed by using a thin film of silicon dioxide (SiO ₂ ) and tantalum pentoxide (Ta ₂ O ₅ ) using a CVD method, a PVD method (Physical Vapor Deposition), or the like. ) Are alternately stacked using vapor deposition or the like (here, for example, about 10 to 20 layers). Then, the fixed electrode portion 1 and the movable portion 2 are formed by removing the resist (FIG. 4 (5)).

  The manufacturing process and joining process of the fixed comb side substrate part 3 will be described with reference to FIG. In order to produce the fixed comb-tooth side substrate part 3, for example, a glass substrate 31 such as Corning # 7740 (trade name) is used (FIG. 5 (1)). Then, a highly reflective film 32 is formed on the upper surface (the outer surface) of the glass substrate 31. Further, an antireflection film 33 is formed on the lower surface (movable part 2 side) (FIG. 5 (2)). The formation method of the high reflection film 32 and the antireflection film 33 is the same as that of the high reflection film 23 and the antireflection film 22 described above, and therefore description thereof is omitted. The fixed comb-tooth side substrate portion 3 is manufactured by the steps described above.

  And the fixed electrode part 1 and the fixed comb-tooth side board | substrate part 3 are joined so that the highly reflective film 23 and the highly reflective film 32 may oppose in a parallel state (FIG. 5 (3)). For this bonding, for example, at least one bonding method of anodic bonding, surface activation bonding, bonding with an adhesive, or low-melting glass bonding is used at least once. At the time of this joining, joining may be performed in a vacuum chamber so that the inside becomes a vacuum (vacuum sealing) or joining under an optimum pressure such as a reduced pressure state. In the present embodiment, silicon and glass are bonded. For example, when silicon is bonded to each other, surface activated bonding can be performed when the surface is mirror-finished.

Here, for example, anodic bonding is performed through the following steps. First, the surface of the fixed electrode portion 1 on which the high reflection film 23 is formed and the surface of the fixed comb tooth side substrate portion 3 on which the high reflection film 32 is formed are placed facing each other. Then, the minus side of the DC power source and the glass substrate 31 are connected, and the plus side of the DC power source and the substrate unit 11 (base layer 101) are connected. Next, for example, a DC voltage of about several hundred volts is applied between the glass substrate 31 and the substrate portion 11 while heating the glass substrate 31 to about several hundred degrees Celsius. By heating the glass substrate 31, alkali metal positive ions, for example, sodium ions (Na ⁺ ) contained in the glass substrate 31 can easily move. As the alkali metal positive ions move in the glass substrate 31, the bonding surface of the glass substrate 31 with the substrate portion 11 is relatively negatively charged. On the other hand, the bonding surface of the substrate unit 11 with the glass substrate 31 is positively charged. As a result, silicon (Si) and oxygen (O) share an electron pair and are covalently bonded, thereby joining the glass substrate 31 and the substrate unit 11. In addition, when so-called low-melting glass is used as the glass substrate 31, low-melting-point glass bonding can be performed by fusing and bonding the glass interface.

  In surface activated bonding (SAB), an oxide layer formed by a reaction with oxygen in the atmosphere and a layer of adsorbed gas molecules that are usually present on the surface of a substance are mixed with argon (Ar) in a vacuum. The surface of the material is removed by etching with a beam of inert gas such as, and the surfaces are activated and bonded together. Since the activated surface has a strong binding force with other molecules, it can be firmly bonded. Further, surface activated bonding can be performed at room temperature. Therefore, it is possible to increase the degree of freedom such as simplification of the process. In this case, in the present embodiment, for example, a silicon substrate is used instead of the glass substrate 31, and the surfaces of the silicon substrate and the silicon serving as the substrate portion 11 are activated in a mirror state and bonded together.

  Here, for example, in order to prevent the fixed comb electrode 12 and the movable comb electrode 21d from contacting and short-circuiting and sticking (sticking) from occurring, the fixed comb electrode 12 and the movable comb electrode 12 are prevented. An insulating film (not shown) may be formed on at least one of 21d. The insulating film can be realized, for example, by vapor deposition by a CVD method, a PVD method, or the like.

  Next, the operation of the wavelength tunable filter having the above configuration will be described with reference to FIG. A voltage (hereinafter referred to as drive voltage) is applied to the fixed comb electrode 12 and the movable comb electrode 21d (movable body 21a). This drive voltage is, for example, an AC sine wave voltage of 60 Hz or a pulsed voltage. The fixed comb electrode 12 is externally applied via the extraction electrode 13, while the movable comb electrode 21 d is applied via the electrode terminal portion 24, the support portion 21 c and the connecting portion 21 b, thereby A potential difference is given to the tooth electrode 12. Due to the potential difference due to the driving voltage, an electrostatic force is generated between the fixed comb electrode 12 and the movable comb electrode 21d, and the movable comb electrode 21d is displaced (drawn) toward the fixed comb electrode 12 side. Thereby, the movable body 21a is displaced to the fixed comb electrode 12 side, that is, the optical gap OG is changed. At this time, since the connecting portion 21b has elasticity, the movable body 21a changes elastically.

  Light having a plurality of wavelength bands (for example, 60 to 100) (for example, light in the infrared region. Each wavelength band includes, for example, a data signal) is below the fixed electrode portion 1 (see the arrow in FIG. 1). Then, the light enters the wavelength tunable filter and passes through the glass substrate 31 and the highly reflective film 32. The high reflection film 23 is provided above and the high reflection film 32 is provided below, and enters a space (reflection space) formed by the opening 11a and the like.

  The incident light is repeatedly reflected between the high reflection film 23 and the high reflection film 32. In this process, only light having a wavelength satisfying the interference condition corresponding to the optical gap OG is reflected in the high reflection film 23 and the movable body. 21a passes through the antireflection film 22 and exits from above the wavelength tunable filter. On the other hand, light having a wavelength that does not satisfy the interference condition is rapidly attenuated. Therefore, the wavelength of light to be transmitted can be selected by changing the optical gap OG by displacing the movable body 21a (the highly reflective film 23 serving as a movable reflective surface).

  As described above, according to the first embodiment, the comb-like electrode (capacitor) is formed by the fixed comb electrode 12 formed on the fixed electrode portion 1 and the movable comb electrode 21d formed on the movable portion 2. Since it comprised, the ratio of the distance which can perform the stable displacement until pull-in generate | occur | produces with respect to the distance between electrodes can be made larger than the case of a parallel plate type. Therefore, since the distance between the electrodes can be shortened compared to the parallel plate type, when the same electrostatic force is obtained, the driving voltage can be lowered compared to the parallel plate type wavelength tunable filter, thereby saving power. (When the drive voltage is the same, the electrostatic force is increased).

  Further, the fixed electrode portion 1 and the movable portion 2 of the wavelength tunable filter can be manufactured by using the SOI substrate 100 which is one substrate. In particular, in the case of the SOI substrate 100, since the active layer 103 has specularity, high optical characteristics can be obtained by the movable body 21a formed from the active layer 103. Further, although the fixed comb electrode 12 is formed on the base layer 101 of the SOI substrate 100, it can also be used as a spacer, so that it is not necessary to form a spacer again, and the process for packaging is simplified. The time can be shortened. Also, by polishing or etching the base layer 101 to a predetermined thickness (thickness necessary for providing the optical gap OG) before joining at least the base layer 101 and the fixed comb-side substrate portion 3. Further, the time for forming the opening 11a and forming the fixed comb electrode 12 can be shortened. In particular, by polishing, a desired thickness can be formed with high accuracy, and a desired optical gap OG can be obtained, so that optical characteristics can be improved.

  Further, since the substrate portion 11 of the fixed electrode portion and the glass substrate 31 are joined together with anodic joining, they can be joined firmly without being separated later. Further, even when the substrate of the fixed comb-tooth side substrate portion 3 is formed of a silicon substrate, it can be bonded by performing surface activation bonding, and thus can be firmly bonded in the same manner as anodic bonding.

Embodiment 2. FIG.
FIG. 6 is a sectional view showing a wavelength tunable filter according to the second embodiment of the present invention. In FIG. 6, those denoted by the same reference numerals as those in FIG. 1 perform the same functions as those described in the first embodiment, and thus description thereof is omitted. In FIG. 6, in the movable part 2A, the high reflection film 23 is provided on the upper surface side of the movable body 21a. On the other hand, it differs from the movable part 2 in that the antireflection film 22 is provided on the lower surface side of the movable body 21a. The fixed comb-tooth side substrate portion 3A is different from the fixed comb-tooth side substrate portion 3 in that it has an antireflection film 34 instead of the high reflection film 32. In the present embodiment, the fixed comb-tooth side substrate portion 3A is not provided with a highly reflective film, and is not a fixed reflection portion like the fixed comb-tooth side substrate portion 3.

  In the present embodiment, the movable portion side substrate portion 4 serving as a fixed reflection portion is constituted by a glass substrate 41 having a recess 41a, a high reflection film 42, and an antireflection film 43. The high reflection film 42 is provided on the bottom surface of the recess 41a (the surface facing the movable portion 2A), which is a fixed reflection surface. An antireflection film 43 is provided on the opposite surface (outer surface).

  The formation of the recess 41a by etching or the like of the glass substrate 41 can be strictly controlled (process management) as compared to the formation by wet etching or the like of silicon. Therefore, the distance (optical gap OG) between the high reflection film 23 and the high reflection film 42 on the movable body 21a can be formed with high accuracy, and more optical characteristics as designed can be obtained.

  In the present embodiment, in addition to the steps of the first embodiment, a step of joining the movable part side substrate part 4 and the movable part 2 is also included. Regarding the bonding, the anodic bonding, low melting point glass bonding, bonding with an adhesive, or the like described in the first embodiment can be used, and thus description thereof is omitted.

  Here, in the present embodiment, the opening 11a is formed for the purpose of forming the fixed comb electrode 12 and the like, and then packaged by the fixed comb side substrate 3 to protect the movable body 21a and the like. Yes. For example, the fixed comb electrode 12 can be formed without providing the opening 11a, or if the opening 11a is not required, such as using light in a wavelength region that transmits silicon, the silicon substrate can be packaged as it is. You can also plan.

  As described above, according to the second embodiment, since the optical gap OG is provided between the highly reflective film 23 of the movable part 2A, the bottom surface of the inner wall of the concave portion 41a of the glass substrate 41 is used as a fixed reflective surface. 42 was formed. The formation of the recess 41a of the glass substrate 41 can make the condition management (process management) stricter than the etching of silicon, and the surface formed by etching can be smoothed by adding alcohol. Thus, it is possible to obtain a more accurate optical gap OG and improve optical characteristics. Moreover, strong bonding can be performed by performing anodic bonding as described above. Moreover, the spacer does not need to be formed by forming the recess 41a.

Embodiment 3 FIG.
In the present embodiment, a piezoresistive element is provided in at least one connecting portion 21b. For example, when the four connecting portions 21b are provided every 90 ° as in the above-described embodiment, it is preferable to provide a piezoresistive element in the pair of opposing connecting portions 21b. Further, if possible, it is more preferable to provide all the connecting portions 21b. In this embodiment, by providing a piezoresistive element, feedback control using the piezoresistive effect is performed in which the resistivity changes in proportion to the stress applied to the piezoresistive element. Thereby, the stress of the connecting portion 21b is adjusted, and the movable body 21a (high reflection film 23) is displaced in the vertical (up and down) direction in a parallel state with respect to the high reflection films 32 and 42 serving as fixed reflection surfaces. And the optical characteristics can be improved.

Embodiment 4 FIG.
In the above-described embodiment, the fixed electrode portion 1 and the movable portion 2 are manufactured from the same SOI substrate 100. However, the present invention is not limited to this. For example, the fixed electrode portion 1 and the movable portion 2 can be formed on different substrates, and then joined or the like via the insulating portion 14. In addition, after the fixed comb electrode 12 is formed in the fixed electrode portion 1, the movable portion 2 and the like can be manufactured after the substrate is bonded via the insulating portion 14. Thus, by forming each separately, especially the fixed comb-tooth electrode 12 can be formed with high precision. The fixed comb electrode can be formed of a metal film or the like.

Embodiment 5 FIG.
In the above-described embodiment, the high reflection film 18 is provided on the drive electrode portion 1 and the high reflection film 23 is provided on the movable portion 2 in relation to the wavelength band of incident light, and the fixed reflection surface or the movable reflection surface is configured. The light was reflected and transmitted. The present invention is not limited to this. For example, the bottom surface of the movable body 21a and the second recess 11b of the drive electrode substrate 11 is mirror-finished, and each can serve as a reflective surface (as a reflective surface). In particular, the highly reflective films 18 and 22 need not be provided. As a result, the process can be simplified. The same applies to an antireflection film having a function as an antireflection surface.

Embodiment 6 FIG.
Although the above embodiment has been described in detail with reference to the drawings, the specific configuration is not limited to this embodiment, and even if there is a design change or the like without departing from the gist of the present invention. It is included in the present invention.

  For example, in the above-described embodiment, the example in which the fixed comb-tooth side substrate portion 3 and the movable portion side substrate portion 4 are configured by the glass substrates 31 and 41 has been described. A material that transmits light in the transmission wavelength band, for example, silicon, sapphire, germanium, or the like may be used as the substrate material.

Further, since both the silicon dioxide (SiO ₂ ) film and the tantalum pentoxide (Ta ₂ O ₅ ) film constituting the highly reflective film 23 formed on the movable body 21a are insulators, the highly reflective film 23 is formed as an insulating film. You may also use as. In this case, it is not necessary to have the same number of layers for the portion functioning as the highly reflective film and the other portions. Thus, if the highly reflective film 23 is also used as an insulating film, the same effects as those of the above-described embodiment can be obtained with a small number of manufacturing steps, and a wavelength tunable filter can be configured at low cost.

1 is a cross-sectional view of a wavelength tunable filter according to a first embodiment. The top view of the movable part board | substrate 21 which comprises a wavelength variable filter. The figure showing the preparation process of the fixed electrode part 1. FIG. The figure showing the manufacturing process of the movable part 2. FIG. The figure showing the process until preparation of the fixed comb-tooth side board | substrate part 3, and wavelength variable filter manufacture. Sectional drawing of the wavelength tunable filter which concerns on 2nd Embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Fixed electrode part, 11 Substrate part, 11a Opening part, 12 Fixed comb electrode, 13 Electrode extraction part, 14 Insulation part, 2 Movable part, 21 Movable part board | substrate, 21a Movable body, 21b Connection part, 21c Support part, 21d Movable comb electrode, 22, 33, 34, 43 Antireflection film, 23, 32, 42 High reflection film, 24 electrode terminal part, 3 fixed comb side substrate part, 31 glass substrate, 4 movable part side substrate part, 41 Glass substrate, 41a recess, 100 SOI substrate, 101 base layer, 102 insulating layer, 103 active layer, 110 silicon oxide film, 111 photoresist pattern, OG optical gap

Claims (10)

A movable body having a movable reflecting surface, and transmitting light of a predetermined wavelength by reflecting in a direction perpendicular to the movable reflecting surface and reflecting light of a wavelength other than the light of the predetermined wavelength, the movable body A movable part integrally formed with a support that supports the movable body and an electrostatic force that displaces the movable body, and a movable comb electrode that is displaced together with the movable body;
A fixed electrode portion provided with a first gap and the movable comb electrode, and a fixed comb electrode formed with the fixed comb electrode for generating the electrostatic force between the movable comb electrode;
A wavelength comprising: a fixed reflection portion provided opposite to the movable reflection surface with a second gap, and having a fixed reflection surface that further reflects light reflected by the movable reflection surface Variable filter.   The movable part is formed by processing the active layer of the SOI substrate in which an active layer, an insulating layer, and a base layer are sequentially stacked, and the fixed electrode part is formed by processing the base layer of the SOI substrate. The wavelength tunable filter according to claim 1, wherein:   The wavelength tunable filter according to claim 1, wherein a substrate constituting the fixed reflection portion is bonded to the fixed electrode portion.   3. The wavelength tunable filter according to claim 1, wherein the substrate constituting the fixed reflection portion has a recess having an inner wall bottom surface portion as the fixed reflection surface, and is joined to the movable portion.   The wavelength tunable filter according to any one of claims 1 to 4, wherein the substrate constituting the fixed reflection portion is a glass substrate capable of anodic bonding with the fixed electrode portion or the movable portion.   The wavelength tunable filter according to any one of claims 1 to 4, wherein the substrate constituting the fixed reflecting portion is a substrate that can be surface activated bonded to the fixed electrode portion or the movable portion.   The tunable filter according to any one of claims 1 to 6, wherein a piezoresistive element whose resistivity changes based on stress is provided at a connecting portion of the support body with the movable body.   The wavelength tunable filter according to claim 1, wherein at least one of the movable comb electrode and the fixed comb electrode is covered with an insulating film. The movable body is displaced in the vertical direction with respect to the fixed reflecting surface by electrostatic force, and the fixed reflecting surface and the movable are reflected from the light reflected between the fixed reflecting surface and the movable reflecting surface provided on the movable body. A method of manufacturing a wavelength tunable filter that transmits light of a predetermined wavelength based on a distance from a reflecting surface,
A first step of forming at least a fixed comb electrode on a base layer of a laminated substrate in which an active layer, an insulating layer, and a base layer are sequentially laminated;
The active layer is etched to generate a movable body, an electrostatic force that displaces the movable body, a movable comb electrode that is displaced together with the movable body, and a support that supports the movable body, and a predetermined insulating layer is formed. A second step of removing a portion of
A method of manufacturing a wavelength tunable filter, comprising: a third step of bonding the substrate having the fixed reflection surface and the laminated substrate. 10. The method of manufacturing a wavelength tunable filter according to claim 9, wherein the substrate is etched or polished to a necessary thickness before the first step.

Images