JP6041389B2 - Light modulator - Google Patents

Description

  The present invention relates to an optical modulator that performs optical modulation by applying stress.

  With the increase in capacity of optical communication, speeding up of optical communication is required not only in the backbone network but also in the subscriber network. To increase the speed of optical communication, it is important to increase the speed of an optical active element that performs optical modulation. At present, optically active elements using compound semiconductors are mainly used because of demands for miniaturization and low power consumption. However, in order to apply to subscriber network devices, it is important that the cost is low, and development of an optical active element using a silicon-based material is progressing instead of an optical active element using an expensive compound semiconductor. It is out.

  As an optical active element using a silicon-based material, for example, there is a light absorbing element using free electron absorption by carrier injection (see Patent Document 1). There is also an element using a Mach-Zehnder interferometer (see Patent Document 2). In addition, there is an optical modulator using the electric field absorption effect of germanium (see Non-Patent Document 1). Among these, the technique using carrier injection and a Mach-Zehnder interferometer consumes a large amount of power, and it is difficult to cope with low power consumption. The technology using the electro-absorption effect can reduce the power consumption, but the light absorption edge of germanium has a long wavelength region of 1650 nm or more and is out of the communication wavelength band, so it is considered difficult to use for optical communication. It was done.

  On the other hand, in recent years, a technique for extending the light absorption edge of germanium to the optical communication wavelength band by using the quantum effect has been proposed. According to this technology, there is a possibility that a low-cost optical modulator that can be used even in a subscriber network can be realized with low power consumption. However, the optical band that can be modulated is not as wide as a few nm, and in order to cover the C band and the L band in the optical communication wavelength band, 10 or more types of quantum well structures are separately formed.

  On the other hand, a technique for obtaining a wider light modulation band with one element has been proposed (see Patent Document 3). In this technique, an optical active element having the above-described configuration is formed on a cantilever beam that can be displaced, and stress is applied to the optical active element to obtain a wider light modulation region. In addition, a technique has been proposed in which the beam can be deformed with a smaller driving voltage and the power consumption can be reduced by adjusting the structure of the cantilever. In this technique, the element is arranged in a region where stress is concentrated between the movable electrode forming portion for driving the beam and the support portion for supporting the beam. By making this region easier to deform as a finer structure in plan view, the beam can be deformed with a smaller driving voltage, thereby reducing power consumption.

Japanese Patent No. 3957187 Japanese Patent No. 4429711 International Publication No. 20111/24968

N. Feng et al., "30GHz Ge electro-absorption modulator integrated with 3μm silicon-on-insulator waveguide", Optics Express, vol.19, no.8, pp.7062-7067, 2011.

  However, in the above-described technique, the space between the thin portion in plan view and the support portion cannot be accurately set due to manufacturing restrictions, and the place where the photoactive element composed of germanium using the quantum effect is disposed In addition, there is a problem that it is difficult to concentrate stress efficiently.

  In the technique described above, stress concentration is important in order to effectively control the band edge. As a stress concentration method, a narrower neck as described above is provided at the fixed end of the cantilever. However, even if a constriction is provided, the effect of stress concentration cannot be expected unless the position of the fixed end coincides with the position of the constriction.

  Silicon photonics is expected as a platform that can be used to reduce the size and cost of optical communication devices. In order to realize the stress applied variable band modulator as described above on silicon photonics, a silicon optical waveguide is loaded on a beam made of silicon, and the electric field absorption effect of germanium formed on the silicon optical waveguide is used. It is promising.

  In order to realize such a beam structure, a manufacturing method is used in which an SOI (Silicon on Insulator) substrate is used, a beam shape is formed with a surface silicon layer, and a buried oxide layer (BOX layer) is removed as a sacrificial layer. In general, wet etching using hydrofluoric acid is used to remove the BOX layer. In dry etching, since the silicon oxide film (BOX layer) under the surface silicon layer cannot be removed, the above-described wet etching capable of wraparound etching is applied.

  Here, the wet etching of the silicon oxide film with hydrofluoric acid proceeds isotropically. For this reason, when fabricating a beam structure with a surface silicon layer, if a sufficiently thick BOX layer underneath is to be etched completely, the beam structure is etched deeper than the fixed end of the beam (undercut). It was. The undercut increases the distance between the fixed end of the beam and the constricted portion of the beam, which hinders efficient stress concentration on the constricted portion.

  As a method for reducing the undercut, there is a method in which a large number of small holes are formed in the beam, and hydrofluoric acid is infiltrated from these holes to etch the lower BOX layer. According to this method, the undercut can be minimized. However, for example, when holes are arranged on an axis orthogonal to the main stress direction, stress concentrates near the hole, and stress cannot be concentrated near the beam fixing end (neck portion), which was the original purpose. A problem occurs. In this state, stress cannot be efficiently concentrated on the optical active element.

  The present invention has been made to solve the above-described problems, and is a light that modulates light by applying stress by displacing the beam structure to an optical active element formed on the beam structure. It is an object of the modulator to allow stress to be concentrated efficiently on the optical active element.

  An optical modulator according to the present invention is an optical modulator formed by processing a support layer formed on a substrate and a movable part forming layer formed on the support layer, and the movable part forming layer is A movable structure formed by processing and spaced apart on a substrate and provided with a plurality of through holes, a fixed portion formed by processing the movable portion forming layer, and a movable portion forming layer are processed. The constricted portion that is formed by connecting the one end of the movable structure and the fixed portion, and the support layer is processed by selective etching with respect to the movable portion forming layer to be formed and fixed in the region below the fixed portion. A support part for supporting the part on the substrate, a movable electrode formed on the movable structure, a fixed electrode formed on the substrate for drawing the movable electrode toward the substrate, and a constriction from the fixed part Light formed over the movable structure via the part A path and an optical active element that is formed on the optical waveguide at the constricted portion and modulates light by the applied stress, and the width of the constricted portion is smaller than the width of the movable structure in plan view. The plurality of through holes are arranged in a state of being arranged at lattice points of a lattice having a regular triangular shape, a regular square shape, or a regular hexagonal shape as a basic lattice shape.

  In the optical modulator, the interval between adjacent through holes may be made smaller as the distance from the fixed portion is closer.

In the optical modulator, to the direction in which the movable structure when viewed from the fixed portion is formed, that is a state in which the sides of the lattice element is inclined.

  As described above, according to the present invention, in an optical modulator that modulates light by applying stress by displacing the beam structure to the optical active element formed on the beam structure, the optical active element has an efficiency. Therefore, an excellent effect that stress can be concentrated can be obtained.

FIG. 1 is a perspective view showing a configuration of an optical modulator according to an embodiment of the present invention. FIG. 2 is a plan view showing a partial configuration of the optical modulator according to the embodiment of the present invention. FIG. 3 is a plan view showing a partial configuration of the optical modulator according to the embodiment of the present invention. FIG. 4 is a plan view showing a partial configuration of the optical modulator according to the embodiment of the present invention. FIG. 5 is a plan view showing a partial configuration of the optical modulator according to the embodiment of the present invention. FIG. 6 is a plan view showing a partial configuration of the optical modulator according to the embodiment of the present invention. FIG. 7 is a plan view showing a partial configuration of the optical modulator according to the embodiment of the present invention.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a perspective view showing a configuration of an optical modulator according to an embodiment of the present invention. This optical modulator is formed by processing the support layer 102 formed on the substrate 101 and the movable part forming layer 103 formed on the support layer 102.

  First, a movable structure 104 formed by processing the movable part forming layer 103 and spaced apart from the substrate 101, a fixed part 106 formed by processing the movable part forming layer 103, A constricted portion 107 that is formed by processing the movable portion forming layer 103 and connects one end of the movable structure 104 and the fixed portion 106 is provided.

  Further, the support layer 102 is formed in a region below the fixed portion 106 by processing the support layer 102 by selective etching with respect to the movable portion forming layer 103, and includes a support portion 105 that supports the fixed portion 106 on the substrate 101. The fixing unit 106 is fixed on the support unit 105.

  The movable structure 104, the fixed portion 106, and the constricted portion 107 are integrally formed by processing the movable portion forming layer 103. The width of the constricted portion 107 is smaller than the width of the movable structure 104 in plan view.

  For example, the substrate 101, the support layer 102, and the movable part forming layer 103 are well-known SOI substrates, and a surface silicon layer having a thickness of about 0.2 μm is the movable part forming layer 103, and has a thickness of about 2 to 3 μm. The support oxide layer 102 is the buried oxide layer. Therefore, according to wet etching with hydrofluoric acid, the support layer 102 can be processed by selective etching with respect to the movable part forming layer 103.

  In addition, a movable electrode 108 formed on the movable structure 104 is provided. Here, the substrate 101 can be used as an electrode by forming the substrate 101 from silicon into which impurities are introduced. In this case, the substrate 101 serves as a fixed electrode for attracting the movable electrode 108 to the substrate 101 side. The movable electrode 108 is connected to the wiring 110 drawn on the fixed portion 106 via a part of the constricted portion 107. Note that in the case where the fixed electrode is formed separately from the substrate 101, it is only necessary that the fixed electrode be formed on the substrate 101 in a position overlapping the movable electrode 108 in plan view, in other words, in a region where the movable structure 104 is projected. .

  In addition, the optical waveguide 111 formed on the movable structure 104 from the fixed portion 106 through the constricted portion 107, and the light modulation by the applied stress formed on the optical waveguide 111 at the constricted portion 107. The optical active element 112 which performs this. The optical active element 112 modulates an optical signal guided through the optical waveguide 111.

  The traveling direction of the optical waveguide 111 is bent on the movable structure 104, and both ends of the optical waveguide 111 are disposed on the fixed portion 106. The optical waveguide 111 is a rib-type waveguide. An optical signal incident from one end of the optical waveguide 111 first passes through a portion where the optical active element 112 is formed on one of the constricted portions 107 formed in a divided manner. Next, the traveling direction is changed by 180 ° in the bent portion of the optical waveguide 111 on the movable structure 104. Next, the light passes through a portion where the optical active element 112 is formed on any other constricted portion 107, and is emitted from the other end of the optical waveguide 111.

  The optical active element 112 has a quantum well structure composed of a well layer made of germanium with a thickness of about 10 nm and a barrier layer made of a mixed crystal of silicon and germanium with a thickness of about 10 nm. When stress is applied to the optical active element 112 configured as described above, the band gap of germanium constituting the well layer can be converted to a wavelength and varied by 100 nm. Therefore, the optical active element 112 can change the light wavelength to be absorbed by applying stress.

  In the optical modulator described above, by applying a voltage so that a potential difference is generated between the movable electrode 108 and the substrate 101, an electrostatic attractive force is generated between the movable electrode 108 and the substrate 101, and the movable structure 104. Can be displaced closer to the substrate 101 side. At this time, the movable structure 104 and the constricted portion 107 are deformed to generate stress. The stress generated here is concentrated in the constricted portion 107 which is disposed at a location closer to the fixed portion 106 and has a narrow width. In addition, the amount of displacement of the movable structure 104 can be changed by changing the voltage to be applied.

  Since the optical active element 112 is formed on the constricted portion 107 where the stress is concentrated in this way, by generating an electrostatic attractive force on the movable structure 104 as described above, Stress (tensile strain) can be applied, and as a result, the wavelength of transmitted light in the optical active element 112 can be changed (modulated). For the configuration of the optical modulator described above, refer to Japanese Patent Application No. 2012-159617.

  In the optical modulator that operates as described above, in the embodiment, the movable structure 104 includes a plurality of through holes 141, and the plurality of through holes 141 are any one of a regular triangle, a regular square (square), and a regular hexagon. It is characterized in that it is arranged in a state of being arranged at lattice points of a lattice having a shape of a basic lattice. The through hole 141 is used for allowing an etchant (hydrofluoric acid) used when the support layer 102 is processed by selective etching to the movable part forming layer 103 to enter the lower layer than the movable part forming layer 103 (movable structure 104). It is a hole. In FIG. 1, the dotted lines shown between the through holes 141 are virtual lines for explaining the state of the lattice, and do not actually exist. The same applies to FIGS. 2, 3, 4, 5, 6 and 7.

  The through-hole 141 penetrates the movable structure 104, from which hydrofluoric acid enters and the support layer 102 made of silicon oxide can be etched. The diameter of the through hole 141 may be about 50 nm or more. With this hole diameter, hydrofluoric acid can enter. The plurality of through holes 141 may be arranged uniformly over the entire area of the movable structure 104.

  The manufacture of the above-described optical modulator will be briefly described. First, an SOI substrate is prepared. The base portion of the SOI substrate is the substrate 101, the buried oxide layer is the support layer 102, and the surface silicon layer is the movable portion forming layer 103. First, the optical waveguide 111 and the optical active element 112 are formed by a known lithography technique, etching technique, and film forming technique. A metal layer made of gold resistant to hydrofluoric acid is formed on the movable part forming layer 103 by a known film formation technique, and then the metal layer is patterned by a known lithography technique and etching technique, whereby the movable electrode 108 and the wiring are formed. 110 is formed.

  Next, a mask pattern for forming the movable structure 104, the fixed portion 106, and the constricted portion 107 is formed by a known lithography technique. At this time, it is assumed that a plurality of hole portions corresponding to the through holes 141 are provided in a region of the mask pattern which is the movable structure 104. By using the mask pattern thus formed and patterning the movable part forming layer 103 by a known dry etching technique, the movable structure 104, the fixed part 106, and the constricted part 107 can be formed. A plurality of through holes 141 can be formed. In this case, in the portion of the movable electrode 108, the through hole 141 also penetrates the movable electrode 108.

  After the movable structure 104, the fixed portion 106, and the constricted portion 107 are formed as described above, the used mask pattern is removed. Next, hydrofluoric acid is allowed to act on the support layer 102 through the opening region around the movable structure 104 and the constricted portion 107, and through the through hole 141, thereby selectively supporting the support layer 102 other than the lower region of the fixed portion 106. Etch away.

  By this etching, the movable structure 104 and the support layer 102 below the constricted portion 107 are removed, and the movable structure 104 and the constricted portion 107 are separated from the substrate 101. Further, since the wet etching described above is isotropically etched, the lower part of the end portion of the fixed portion 106 is also etched to some extent and undercut occurs.

  However, since a plurality of through holes 141 are formed in the movable structure 104 and hydrofluoric acid acts on the support layer 102 from here, the support layer 102 below the movable structure 104 does not have the through holes 141. Compared to the above, etching is removed more rapidly. As a result, the hollow structure of the movable structure 104 and the constricted portion 107 can be realized by a shorter etching time, and as a result, the amount of the undercut described above can be reduced. If the amount of undercut can be reduced, the stress concentration on the constricted portion 107 described above can be performed more efficiently.

  By the way, the through holes 141 are arranged in a hexagonal lattice as described above, for example. A honeycomb structure having a regular hexagonal basic lattice is known as a porous material having high strength and high rigidity, such as a building material. For this reason, even if the plurality of through holes 141 are formed in this arrangement, the strength and rigidity of the movable structure 104 are not significantly impaired.

  Here, in order to completely remove the silicon oxide film (support layer 102) having the thickness d, it is necessary to set the interval between the hydrofluoric acid intrusion through holes 141 to be equal to or less than the thickness d. Since the lattice spacing of a hexagonal lattice having a length of d on one side is √3d, when a hole for entering hydrofluoric acid is arranged at the apex of the hexagonal lattice, the lattice spacing x of the hexagonal lattice shown in the plan view of FIG. May be less than or equal to √3 times the thickness of the support layer 102.

  If the lattice spacing x is reduced, the rigidity is lowered, so that stress concentration can be promoted. Therefore, as shown in FIG. 3, first, a plurality of through holes 141 are arranged at a lattice size that is √3 times the thickness of the support layer 102 at a position far from the fixed portion 106 of the movable structure 104. In addition, a plurality of through holes 142 are arranged with a smaller lattice size as the fixing portion 106 is approached. The interval between the adjacent through holes 142 is made smaller as it is closer to the fixed portion 106. With these configurations, the stress concentration in the constricted portion 107 can be increased.

  In addition to the hexagonal lattice, the arrangement of the through holes 141 may be a rectangular lattice as shown in FIG. As described above, in the configuration in which the plurality of through holes 141 are arranged at the lattice points of the quadrangular lattice in which the shape of the basic lattice is a square, the side of the lattice is the direction in which the movable structure 104 is formed when viewed from the fixed portion 106. There are also parallel or orthogonal configurations. In this case, the main stress direction generated in the moving movable body 104 is the above-described direction, and the through holes 141 are arranged in a direction perpendicular to the main stress direction. In this state, since rigidity is lowered, as shown in FIG. 5, it is preferable that the sides of the square lattice (basic lattice) be inclined with respect to the direction in which the movable structure 104 is formed as viewed from the fixed portion 106.

  Further, the arrangement of the through holes 141 may be a triangular lattice as shown in FIG. As described above, in the configuration in which the plurality of through holes 141 are arranged at the lattice points of a triangular lattice in which the shape of the basic lattice is an equilateral triangle, the movable structure 104 is formed with the sides of the lattice as viewed from the fixed portion 106. Some configurations are orthogonal to the direction. In this case, the main stress direction generated in the moving movable body 104 is the above-described direction, and the through holes 141 are arranged in a direction perpendicular to the main stress direction. In this state, since the rigidity is lowered, as shown in FIG. 7, it is preferable that the sides of the triangular lattice (basic lattice) be inclined with respect to the direction in which the movable structure 104 is formed as viewed from the fixed portion 106.

  The present invention is not limited to the embodiment described above, and many modifications and combinations can be implemented by those having ordinary knowledge in the art within the technical idea of the present invention. It is obvious. For example, the through hole may be formed in the constricted portion.

  DESCRIPTION OF SYMBOLS 101 ... Board | substrate, 102 ... Support layer, 103 ... Movable part formation layer, 104 ... Movable structure, 105 ... Support part, 106 ... Fixed part, 107 ... Constriction part, 108 ... Movable electrode, 110 ... Wiring, 111 ... Optical waveguide 112 ... Photoactive element, 141 ... Through hole.

Claims (2)

An optical modulator formed by processing a support layer formed on a substrate and a movable part forming layer formed on the support layer,
A movable structure formed by processing the movable part forming layer and spaced apart on the substrate, the movable structure including a plurality of through holes;
A fixed part formed by processing the movable part forming layer;
A constriction formed by processing the movable part forming layer and connecting one end of the movable structure and the fixed part;
A support part that is formed in a region under the fixed part by processing the support layer by selective etching with respect to the movable part forming layer and supports the fixed part on the substrate;
A movable electrode formed on the movable structure;
A fixed electrode formed on the substrate for pulling the movable electrode toward the substrate;
An optical waveguide formed over the movable structure via the constricted portion from the fixed portion;
An optical active element that is formed on the optical waveguide at the constricted portion and modulates light by applied stress, and
In a plan view, the width of the constricted portion is formed smaller than the width of the movable structure,
The plurality of through-holes are arranged in a state of being arranged at lattice points of a lattice having a regular triangle, a regular square, or a regular hexagon as a basic lattice shape ,
An optical modulator characterized in that a side of the basic grating is inclined with respect to a direction in which the movable structure is formed as viewed from the fixed portion . The optical modulator of claim 1, wherein
An optical modulator characterized in that the interval between the adjacent through holes is made smaller as it is closer to the fixed part.

Images