JP2008281639A - Optical deflection element, optical deflection module, optical switch module, and optical deflection method - Google Patents

Abstract

Provided is a small optical deflection element that can be operated at high speed and can perform large-scale multi-channel switching and can be easily manufactured.
An optical deflection in which an input channel waveguide 0001, an input slab waveguide 0002, a first waveguide array 0005, a second waveguide array 0006 having the same optical path length, and an output slab waveguide 0004 are sequentially connected. The second waveguide array 0006 includes refractive index modulation regions 0007 having different lengths, and the core width of the channel waveguide of the second waveguide array 0006 is equal to the channel guide of the first waveguide array 0005. An optical deflection element wider than the core width of the waveguide.
[Selection] Figure 1

Description

  The present invention relates to an optical deflection element, an optical deflection module, an optical switch module, and an optical deflection method.

  The demand for higher capacity, higher speed, and higher functionality of optical communication systems is increasing rapidly, and as such an optical signal processing device used in an optical communication system, an optical switch that switches an optical signal transmission path is indispensable. It is. In a 1 × N type multi-channel switching element that outputs light input from one port to a desired port from among a plurality (N) of output ports, a channel that can be switched from an increase in capacity of an optical communication system The demand for increasing numbers is increasing.

  As a 1 × N type optical switch element, an optical switch using a micromachine technology called MEMS (Micro Electro Mechanical System), or an oil having a refractive index equal to that of the waveguide is filled at the intersection of the waveguide, and bubbles are formed by heating. Optical switches that switch the optical paths by generating light and reflecting light at intersections have been developed, but the switching time in these devices is as slow as about milliseconds, that is, the path switching speed is slow There's a problem. The speeding up of the network is further advanced, and a high-speed optical switch is desired.

As a method of changing the refractive index at high speed, there is a method using an electro-optic effect (electro-optic effect method). Electro-optical crystal materials such as LiNbO ₃ (abbreviated as LN) and LiTaO ₃ (abbreviated as LT) have an extremely fast response speed to voltage application, and an electric field of about femtosecond (fs). Responds instantly to changes. By utilizing the refractive index modulation by the electro-optic effect, it is possible to switch the light path in an extremely short time on the order of nanoseconds (ns). That is, an optical switch that is driven at a very high speed is realized.

However, an electro-optic optical switching element using a bulk crystal is large in size and has a high voltage required for switching, and is not suitable for a large-capacity ultrahigh-speed switch. A waveguide type optical switch element is generally used as an ultrafast switching element. For example, in Patent Document 1, a refractive index is applied by applying a voltage from above and below the thin film only to one arm of a Y-branch waveguide in which a PLZT ((Pb, La) (Zr, Ti) O ₃ ) thin film is patterned. A 1 × 2 type switch has been reported that provides modulation and outputs light only to the desired arm. Further, in Patent Document 2, a 2 × 2 type using a Mach-Zehnder interference type device is utilized by utilizing a huge second-order electro-optic effect (Kerr effect) possessed by KTN (KTa _1-x Nb _x O ₃ ). A switch has been reported.

  However, in order to increase the switching capacity of a large-scale matrix switch or the like, it is necessary to increase the number of branches by connecting the above 1 × 2 type or 2 × 2 type switching elements in multiple stages. For example, when a switching element (1 × 8 switch element) of about 8 channels is used, if a basic element 1 × 2 switch element is used, it can be connected in three stages, and the number of basic elements is seven. In order to increase the number of channels and realize a 256-channel switching element, it is necessary to connect the basic element 1 × 2 switch elements in eight stages, and the number of necessary basic elements is as large as 255. At present, the element length of one basic element is about several millimeters, and the element area becomes very large when configuring a large-scale multi-channel switch. Further, in order to switch the output port, it is necessary to independently apply a voltage to each basic element, and it is difficult to realize a large-scale 1 × N switching element by the waveguide type switching element.

  As another example of the switching element using high-speed refractive index modulation by the electro-optic effect, in Patent Document 3, a prism is generated in the slab optical waveguide by the electro-optic effect, and the beam is directed in an arbitrary direction by controlling the applied voltage. An optical switch element that switches paths by deflecting has been reported. This method allows easy integration of switch elements, and in principle it is possible to configure a large-scale multi-channel switch.

  However, since a small change in refractive index due to the electro-optic effect is used, the deflection angle of the light beam is small. Therefore, in order to switch ports in a large-scale multichannel switch, it is necessary to increase the propagation length in the waveguide to about 100 mm, which increases the element area. Further, since the deflected light beam propagates in the slab waveguide for a long distance without being confined in the lateral direction, the deflected light beam is easily affected by the beam spread and scattering due to the impurities in the waveguide material and the nonuniformity of the refractive index. Due to the above problems, it is difficult to realize a large-scale 1 × N switching element even in a system using the prism effect by refractive index modulation.

  As described above, in an optical deflection element using a 1 × 2 type or 2 × 2 type switching element, a large-scale multi-channel switching (1 × N switching) operation exceeding 100 ports and ultra-high-speed path switching It is difficult to realize these simultaneously.

  On the other hand, Patent Document 4 proposes a switching element that switches a wavelength channel using an arrayed waveguide grating (AWG) type wavelength demultiplexer. The AWG duplexer has a plurality of output ports corresponding to wavelength channels to be demultiplexed with respect to at least one input port. The wavelength-multiplexed light is demultiplexed to the corresponding output port according to each wavelength. Here, the output port from which a certain wavelength signal is output depends on the optical path length difference between adjacent waveguides in the arrayed waveguide in the AWG element. Therefore, by applying refractive index modulation to the waveguide by the thermo-optic effect, the electro-optic effect, etc., it becomes possible to change the optical path length difference and change the output port. Since AWG has a plurality of output ports, it is attractive as a candidate for a 1 × N large-scale multi-channel switching element.

In a multi-channel switching device using AWG, it is inevitably required to integrate many optical waveguides. Here, as an optical circuit used in an optical communication device, a planar lightwave circuit (PLC) made of a quartz-based material is widely used, and the optical waveguide in the optical switching device proposed so far is used. The relative refractive index difference of the waveguide is often about the same as the relative refractive index difference of the silica-based optical waveguide constituting the PLC. The relative refractive index difference is defined by the relative refractive index difference Δ = (n _core ² −n _clad ² ) / (2n _core ² ) where n _core is the core refractive index and n _clad is the cladding refractive index. It is a parameter. Since the silica-based waveguide couples light from the optical fiber to the waveguide with high efficiency, the relative refractive index difference between the core and the clad is usually as small as about 1%. Therefore, the cross-sectional dimension of the waveguide for satisfying the single mode condition must be increased to about 5 to 10 μm square. Furthermore, the radius of curvature when the waveguide is bent to change the propagation direction must be increased to 1 to 25 mm. Therefore, the device size cannot be reduced and is not suitable for integration of high-density optical circuits. Similarly, when a large-scale multi-channel switching is realized by integrating waveguide switching elements having a relative refractive index difference comparable to that of a PLC, the device size becomes very large.

  Patent Document 4 proposes a switching element using LN as a constituent material of an arrayed waveguide. However, in order to change the optical path length sufficient for switching, a waveguide array having a linear waveguide as a constituent element is inserted. ing. This is because the change in refractive index due to the electro-optic effect of LN is small, and a relatively long propagation distance of 10 mm or more is necessary to obtain a significant change in optical path length. The LN waveguide used in Patent Document 4 realizes channel-type optical confinement by thermally diffusing Ti inside the LN, but its optical confinement strength is very high compared to the PLC. Therefore, it is weak to the bending of the waveguide. Therefore, a PLC waveguide element is used for the array portion including the bend, and the element is configured by hybrid connection with a linear waveguide array using LN.

  On the other hand, a minute light control device using a material having a very high refractive index as a core of an optical waveguide for a PLC waveguide has been studied. For example, a relative refractive index difference of about 40% or more can be realized by using a semiconductor material having a refractive index of 3 or more or a dielectric material having a refractive index of about 2 as a core and a cladding as air. Thereby, the device size can be drastically reduced as compared with a general quartz-based PLC device. Specifically, the bending radius of a waveguide that can propagate without loss is several μm, which is about 1/1000 or less that of a conventional silica-based waveguide. In this way, an optical waveguide having an extremely high relative refractive index difference as compared with a conventional total reflection type optical waveguide, specifically, an optical waveguide having a high index difference (HIC) of about 10% or more is used. It is called a waveguide.

  A typical core material constituting the HIC waveguide is silicon. Silicon is transparent to light in the wavelength 1550 nm band used for optical communication, and its refractive index is as high as about 3.5. Silicon originally has a large thermo-optic effect, but since the core is further miniaturized, refractive index modulation is possible only by heating a small region, so that the power consumption required for switching the element is reduced by the conventional PLC waveguide. Compared to, it is extremely small.

In addition, high-speed light modulation using the plasma dispersion effect of free carriers is becoming possible. In addition, an electro-optic crystal material such as LN, LT, or KTN, or a ceramic material such as PLZT has a large refractive index of 2.0 or more, and is suitable as a core material for an HIC waveguide. It is known that an HIC optical waveguide can be formed by processing an electro-optical material into a channel shape and surrounding the periphery with a low refractive index clad.
JP 2006-58837 A WO 2004-083953 JP 2006-194993 A Japanese Patent No. 3053294

  In the configuration of the switching element in Patent Document 4, in addition to the lengthening of the element due to the PLC, the tolerance of the optical axis when connecting different types of waveguides is small, and it is difficult to finely adjust the connection, and there is an excess loss at the connection part. There is a problem that it is likely to occur. That is, there are problems such as an increase in manufacturing cost of the switching element and an increase in excess loss due to switching.

  The HIC waveguide has a small propagation mode diameter, and one of the problems is to guide the propagation light from the optical fiber and the spatial propagation light into the waveguide with high efficiency. There are several methods for solving this problem. For example, the present inventors have found a method for guiding light that has been spatially propagated to the HIC waveguide with high efficiency. That is, by appropriately disposing a lens having a high NA in the optical axis direction that is not parallel to the light propagation direction, light can be collected on the output end face of the HIC waveguide.

  Further, a method for forming an electric field on a channel-shaped electro-optic material has been found. Utilizing the fact that the HIC waveguide by the electro-optic material enables strong optical confinement to the core, it is possible to place an electrode for forming an electric field close to the core. Therefore, it is possible to reduce the driving voltage when modulating the refractive index of the material. Thus, by using the HIC optical waveguide made of an electro-optic material as the basic wiring of the switching element, it is possible to integrate the elements at a high density and perform a low voltage switching operation.

  With the above technology, light can be guided with high efficiency to the HIC waveguide made of a semiconductor material or an electro-optic material, and light switching, that is, light path switching can be performed with low power consumption or low driving voltage. By applying the HIC waveguide to the above-described AWG type switching element, a compact switching element can be realized. Furthermore, since the HIC waveguide can be bent rapidly, the channel waveguide and the slab waveguide constituting the AWG can be integrated and formed. That is, unlike Patent Document 4, it is not necessary to hybrid-connect waveguides made of different core materials. Therefore, the switching element can be manufactured at a lower cost, and there is no excessive loss at the connection portion.

As described above, the AWG type switching element using the HIC waveguide enables a compact, low-cost and high-speed multi-channel switching element. However, compared to the PLC waveguide, the HIC waveguide has a very small core size for the waveguide to operate in a single mode, and the waveguide has a large equivalent refractive index even if the core size is slightly changed. Change. A switching element using AWG gives a phase difference between adjacent channel waveguides by changing the equivalent refractive index of a waveguide array having a propagation length of the order of millimeters. Therefore, it is important that the equivalent refractive index of each channel waveguide when the refractive index modulation is not performed is uniform. However, in the HIC waveguide, even if the core width varies by about 10 nm, the amount of change given to the equivalent refractive index is about 10 ⁻³ . This is larger than the refractive index modulation amount due to the electro-optic effect such as LN. That is, in order to realize a switching element with an HIC waveguide, only a core width variation of about 1 nm is allowed, and the tolerance for the dimensional accuracy of the core width is very small.

  In order to solve these problems, the present invention is capable of high-speed operation and large-scale multi-channel switching, and is easy to manufacture, and an optical deflection module using the optical deflection element. It is an object of the present invention to provide an optical switch module and to provide an optical deflection method capable of high-speed operation and large-scale multi-channel switching and capable of deflecting even with a minute element.

In order to solve the above problems, the present inventors have completed the following invention.
The present invention is an optical deflection element in which an input slab waveguide having at least one input port formed on a substrate, a waveguide array composed of a plurality of channel waveguides, and an output slab waveguide are sequentially connected. The waveguide array includes a first waveguide array and a second waveguide array in which the optical path lengths of the channel waveguides are equal to each other, and the second waveguide array includes each channel waveguide. The refractive index modulation regions having different lengths are included therein, and the core width of the channel waveguide constituting the second waveguide array is larger than the core width of the channel waveguide constituting the first waveguide array. It is an optical deflection element characterized by being wide.

  In a preferred aspect of the present invention, the core width of the channel waveguide constituting the second waveguide array is changed in an inclined manner at the connection portion between the first waveguide array and the second waveguide array. The optical deflection element is characterized in that

  The present invention is an optical deflection element in which an input slab waveguide having at least one input port formed on a substrate, a waveguide array composed of a plurality of channel waveguides, and an output slab waveguide are sequentially connected. The waveguide array includes a first waveguide array and a second waveguide array in which the optical path lengths of the channel waveguides are equal to each other, and the second waveguide array includes each channel waveguide. The refractive index modulation regions having different lengths are provided therein, and the relative refractive index difference between the core and the clad of the channel waveguide constituting the second waveguide array is the channel guide constituting the first waveguide array. An optical deflecting element characterized by being smaller than a relative refractive index difference between a core and a clad of the waveguide.

  In a preferred aspect of the present invention, a clad constituting the second waveguide array is formed in an inclined shape in the vicinity of a connection portion between the first waveguide array and the second waveguide array. The light deflection element.

  These optical deflecting elements have a small element size, can deflect a light beam into multichannels at high speed, and have a large and easy tolerance for the waveguide.

  The present invention is an optical deflection element in which an input slab waveguide having at least one input port formed on a substrate, a waveguide array composed of a plurality of channel waveguides, and an output slab waveguide are sequentially connected. The waveguide array includes a first waveguide array and a second waveguide array having the same optical path length of the channel waveguide, and the second waveguide array has a photonic crystal array. This is an optical deflecting element. This optical deflection element can shorten the optical propagation length required for optical switching or optical deflection.

  In a preferred embodiment of the present invention, the waveguide array includes an electroabsorption layer between channel waveguides. This light deflection element has little stray light to an adjacent route in route switching.

  In a preferred aspect of the present invention, the optical deflection element is characterized in that a plurality of output channel waveguides are connected to an optical path exit of the output slab waveguide. This optical deflection element is capable of large-scale multi-channel switching.

  In a preferred aspect of the present invention, the optical deflection element is characterized in that a mirror for reflecting light in the optical path is provided in the channel waveguide, and the input slab waveguide also serves as the output slab waveguide. This optical deflection element can further reduce the element size.

  Preferably, the present invention includes an input channel waveguide connected to a connection portion between the input slab waveguide and the waveguide array and / or an input port of the input slab waveguide, and the input slab waveguide and the input channel waveguide. The optical deflection element is characterized in that, in the connecting portion, the core width of the channel waveguide and / or the input channel waveguide is widened. This optical deflection element has less excess loss due to path switching.

  In a preferred aspect of the present invention, the output channel waveguide and / or channel waveguide core width is widened and connected at the connection between the output slab waveguide and the output channel waveguide and / or the waveguide array. The light deflecting element is characterized in that: This optical deflection element has less excess loss due to path switching.

  In a preferred aspect of the present invention, the input slab waveguide and the output slab waveguide have different optical path lengths. This optical deflection element has less excess loss due to path switching.

  According to a preferred aspect of the present invention, in the channel waveguide constituting the second waveguide array, the core of the refractive index modulation region is formed of an optical crystal material having an electrooptic effect, and the electrodes are opposed to each other with the core interposed therebetween. At least one pair is arranged, and the clad is formed between the core and the electrode. This light deflection element can change the light output position at a low voltage.

  The present invention is an optical deflection module configured by sequentially coupling a light source, an optical coupling unit, and an optical deflection element, wherein the optical deflection element is any one of the optical deflection elements described above. Is an optical deflection module. This optical deflection module is capable of large optical deflection or large-scale multi-channel switching with a small size.

  A preferred aspect of the present invention is the optical deflection module comprising a temperature control device for adjusting the temperature of the optical deflection element. This optical deflection module is capable of large optical deflection or large-scale multi-channel switching with a small size.

  The present invention is an optical switch module including an optical amplifier before or after the optical deflection element, wherein the optical deflection element is any one of the optical deflection elements described above. It is. These optical switch modules have a small size and can perform large optical deflection or large-scale multi-channel switching.

  The present invention deflects light by sequentially passing an input slab waveguide having at least one input port formed on a substrate, a waveguide array composed of a plurality of channel waveguides, and an output slab waveguide. In the optical deflection method, the waveguide array is provided with a first waveguide array and a second waveguide array in which the optical path lengths of the channel waveguides are equal to each other, and the second waveguide array includes: Refractive index modulation is performed with different lengths in the respective channel waveguides, and the core width of the channel waveguides constituting the second waveguide array is determined by the channel waveguides constituting the first waveguide array. The light deflection method is characterized in that it is wider than the core width.

  The present invention deflects light by sequentially passing an input slab waveguide having at least one input port formed on a substrate, a waveguide array composed of a plurality of channel waveguides, and an output slab waveguide. In the optical deflection method, the waveguide array is provided with a first waveguide array and a second waveguide array in which the optical path lengths of the channel waveguides are equal to each other, and the second waveguide array includes: Refractive index modulation is performed with different lengths in the respective channel waveguides, and the relative refractive index difference between the core and the clad of the channel waveguides constituting the second waveguide array is determined by the first waveguide array. The optical deflection method is characterized in that it is smaller than the relative refractive index difference between the core and the clad of the channel waveguide to be formed.

  According to the present invention, it is possible to provide a minute optical deflection element capable of high-speed operation and large-scale multi-channel switching and easy to manufacture, and an optical deflection module and an optical switch module using the optical deflection element. In addition, it is possible to provide an optical deflection method that can perform high-speed operation and large-scale multi-channel switching and can be deflected even by a minute element.

  The best mode for carrying out the present invention will be described with reference to the drawings as necessary. Note that it is easy for a person skilled in the art to make other embodiments by changing or correcting the present invention within the scope of the claims, and these changes and modifications are included in the scope of the claims. The following description is an example of a preferred embodiment of the present invention, and does not limit the scope of the claims.

(Operation Principle of AWG Type Optical Deflection Element in the Present Invention)
First, the operation principle and performance of the AWG type optical deflection element in the optical deflection element of the present invention will be described. FIG. 1 schematically shows the configuration of an AWG type optical deflection element. The optical deflection element is configured by sequentially connecting an input channel waveguide 0001, an input slab waveguide 0002, a waveguide array 0003, and an output slab waveguide 0004 so that the optical paths are connected. The waveguide array 0003 includes a first waveguide array 0005 and a second waveguide array 0006. The waveguide array 0003 is composed of a plurality of channel waveguides. In the first waveguide array 0005, it is preferable to provide a certain optical path length difference between adjacent channel waveguides. However, in the second waveguide array 0006, adjacent channel waveguides have the same optical path length. Furthermore, the second waveguide array 0006 includes a refractive index modulation region 0007. The refractive index in the refractive index modulation region 0007 is variable, and the equivalent refractive index of the channel waveguide in the region can be arbitrarily changed within a certain range. The refractive index modulation region 0007 is arranged so as to add a certain optical path length difference between adjacent channel waveguides.

  In this optical deflecting element, light incident on the input port of the input slab waveguide 0002 from the input channel waveguide 0001 is diffracted and spreads in the input slab waveguide 0002 and propagates. Then, it couple | bonds with the several channel waveguide which comprises the waveguide array 0003 in the same phase, and propagates a waveguide array. The propagation light of each channel waveguide interferes with each other in the output slab waveguide 0004, and is condensed into a minute light spot on the termination surface (also an exit port) of the output slab waveguide. Here, the condensing position in the output slab waveguide is determined by the phase difference between the channel waveguides given by propagating through the waveguide array 0003. Therefore, by controlling the refractive index modulation amount, the phase difference between the channel waveguides constituting the waveguide array can be changed, and the final light condensing position can be specified. That is, light deflection can be achieved by refractive index modulation of the element material of the channel waveguide in the refractive index modulation region 0007.

  The size of the light spot condensed on the terminal surface of the output slab waveguide 0004 can be reduced to the electromagnetic field distribution of the propagation mode propagating through the input channel waveguide 0001. Therefore, for example, when the element is constituted by a typical quartz optical waveguide having a relative refractive index difference of about 1%, the output spot diameter is about several μm. Furthermore, if the element is constituted by an HIC optical waveguide in which the difference in refractive index between the core and the cladding is extremely large, the propagation mode diameter of the waveguide can be reduced to 1 μm or less. The beam spot diameter at the exit port can also be made very small to 1 μm or less.

  As shown in FIG. 1, it is simplest to configure the second waveguide array 0007 with straight waveguides having the same adjacent waveguide structure, but the present invention is not necessarily limited to this. A waveguide may be included in the waveguide array 0007. Further, the second waveguide array may be configured such that the optical path length difference between adjacent channel waveguides is an integral multiple of the wavelength of light.

  Since the light beam collected at the end face of the output slab waveguide 0004 is diffused and spreads again after exiting from the end face, a collimating element 0008 is disposed immediately after the output slab as schematically shown in FIG. It is effective to convert the outgoing beam, which is convergent light, into parallel light. This makes it possible to scan with parallel beam light. As the collimating element 0008, a method of arranging an optical lens in the subsequent stage of the output slab is simple.

  FIG. 2 schematically shows another form of the AWG type optical deflection element. Although the optical deflection element is configured by sequentially connecting the input channel waveguide 0001, the input slab waveguide 0002, the waveguide array 0003, and the output slab waveguide 0004, it is the same as the embodiment shown in FIG. In this embodiment, as further enlarged and shown in FIG. 2B, a plurality of output channel waveguides 0009 are connected to the end of the output slab waveguide 0004. The principle of deflecting the focused beam at the output slab end surface by refractive index modulation is the same as described above. In this embodiment, the output channel waveguide 0009 is disposed at the position of the exit port where the light is collected, and the condensed beam is guided to the waveguide. When a plurality of output channel waveguides 0009 are arranged, the light condensing position is changed by the refractive index modulation. Therefore, the condensing position of the output port, that is, the output waveguide is moved by the refractive index modulation in the refractive index modulation region. be able to. Therefore, according to this embodiment, a 1 × N port multi-channel switch can be realized.

  In the configuration shown in FIG. 2, if the waveguide array 0006 including the second refractive index modulation region does not exist, the configuration is the same as that of the AWG type wavelength demultiplexer widely used for optical fiber communication. . At that time, in principle, it is possible to switch the output port by refractive index modulation of the entire first waveguide array 0005 or a part thereof. However, in this case, if the length of the waveguide to be modulated by the refractive index is short and the amount of change in the refractive index of the waveguide material is small, a phase difference large enough to enable port switching is given between adjacent channel waveguides. It is difficult.

  On the other hand, in the optical deflection element of the present invention, it is possible to give a phase difference sufficient for port switching between adjacent channel waveguides by increasing the propagation distance of the second waveguide array. Accordingly, as will be described later, a sufficient deflection angle can be obtained even when refractive index modulation is performed by an electro-optic effect or the like that is high speed but has a small refractive index modulation amount, and the number of ports that can be switched in a 1 × N type switch. Can be increased. In addition, when the refractive index modulation is not performed, in the second waveguide array, the same optical path length is propagated in all the channel waveguides in the array. Therefore, in principle, the light collecting performance compared with the conventional AWG element. There is no deterioration.

  Furthermore, if the propagation distance of the waveguide array is increased, the element size generally increases, but the element size is dramatically reduced by the HIC optical waveguide that has a small core size and can be bent rapidly. Can be reduced.

(Advantages over other waveguide type deflection elements)
A feature of the AWG as a wavelength demultiplexer is that wavelength multiplexed light is demultiplexed into a plurality of output ports at once. Therefore, the AWG has a plurality of output ports corresponding to the number of wavelength channels. By using this structure as a switching element, it is a feature of this element that 1 × N switching can be performed with one element. For example, there are Mach-Zehnder interferometers, Y-branches, ring resonators, directional couplers, and the like as optical waveguide elements applicable to the switching elements, all of which are 1 × 2 or 2 × 2 switching elements. Therefore, for 1 × N switching, 1 × 2 or 2 × 2 switches, which are basic elements, need to be cascaded in multiple stages, which increases the device size. For example, when a 1 × 256 switch having 256 output ports is configured with the above 1 × 2 switch as a basic element, it is necessary to combine eight stages and 255 basic elements.

  On the other hand, in the case of increasing the number of output ports, the optical deflection element of the present invention only needs to increase the number of output channel waveguides coupled to the output slab. There is no need to couple deflection elements in multiple stages. In order to construct a 1 × N switch having a large number of channels, it is required to increase the distance that the focused beam scans the output end face of the output slab waveguide. This is the propagation of the slab radius, that is, the propagation of the slab waveguide. This is easily realized by increasing the length. Note that increasing the slab radius does not affect the focused beam diameter. As described above, the optical deflection element according to the present invention can easily increase the scale of the multichannel switch. Although the result of study on a specific device size will be described later, a 1 × 256 switch can be realized with a device size of 10 mm square or less. This means that it is realized with an area of 1/100 or less as compared with a normal PLC element. Furthermore, the optical deflection element of the present invention can switch the path only by adjusting the refractive index change amount of one refractive index modulation region. Therefore, the refractive index modulation control at the time of path switching can be simplified as compared with a conventional optical switch that requires individual modulation of many basic element devices.

(Advantages of configuring an optical deflection element with an HIC waveguide)
Here, the merit of configuring the optical deflecting element of the present invention with an HIC waveguide will be described in comparison with a conventional optical deflecting element with a PLC waveguide as reported in Patent Document 4. An AWG de-light deflection element formed by a typical PLC waveguide has a large element area of about several cm square. This is because the relative refractive index difference of the PLC in the channel waveguide is as small as about 1% or less, and in the light propagation including the bending as in the first waveguide array, the bending radius of curvature can only be reduced to about several millimeters. Due to that. Here, the relative refractive index difference Δ is a parameter defined by Δ = (n _core ² −n _clad ² ) / 2n _core ² . Furthermore, in the PLC element, the light confinement in the core is weak, and the diameter of the light beam spot condensed in the output slab waveguide is as large as several μm to 10 μm. Therefore, in the arrayed waveguide and the output channel waveguide in which the channel waveguides are arranged adjacent to each other, it is necessary to widen the distance between the channel waveguides to about 10 μm. The above makes it extremely difficult to reduce the element size.

For such a PLC optical circuit, if the optical deflection element is configured by an HIC waveguide in which the refractive index difference between the core and the clad is extremely large, the element area can be dramatically reduced. For example, the HIC waveguide can be bent slightly by several μm, and this property directly leads to miniaturization of the element. Furthermore, since the light is strongly confined in a minute core of 1 μm ² or less, the interval between adjacent channel waveguides in the waveguide array can be reduced to about 1 μm. Since the beam spot diameter of the light collected in the output slab is substantially equal to the propagation mode diameter of the input channel waveguide, the output light is collected in a minute spot in the output slab waveguide by applying the HIC waveguide to the optical deflection element. Can be light. Therefore, the interval between the adjacent exit ports can be reduced, and the interval between the output channel waveguides can be reduced. On the other hand, if the scanning length of the light beam by the optical deflection element is constant, the number of channels that can be switched can be increased by closely packing the output waveguide. As described above, the HIC waveguide is extremely effective for reducing the element area and increasing the number of channels in the multichannel switch.

(Operation example of deflection element: FDTD simulation)
Specific parameters were designed for the optical deflection element as shown in FIG. 2, and the deflection characteristics by refractive index modulation were confirmed. For the performance evaluation, numerical simulation based on a two-dimensional FDTD method (Finite Difference Time Domain method) was used. Since the optical deflecting element of the present invention has an AWG wavelength demultiplexer as a basic configuration, parameter design is determined in accordance with AWG design guidelines. Further, here, the performance is evaluated assuming that a silicon thin wire waveguide having silicon as a core and air as a cladding is applied as a waveguide constituting the element.

A channel waveguide with silicon as the core and air as the cladding is two-dimensionalized by approximation of the equivalent refractive index, and the refractive index n _core of the waveguide core in the simulation is 3.06, and the refractive index n _{clad of the} cladding is 1.0. Were determined. The core width w of the channel waveguide was 480 nm. At this time, the equivalent refractive index n _eq of the waveguide with respect to the wavelength of 1550 nm is 2.644. Further, for the channel waveguides constituting the waveguide array, the waveguide pitch d at the connection portion with the slab waveguide was set to 1.0 μm. The number of channel waveguides constituting the array was 24. Similarly, for a plurality of output waveguides, the waveguide interval D at the connection with the slab waveguide is set to 1.0 μm. The focal length of the slab waveguide (same as the slab radius) was 25 μm. The path difference ΔL between adjacent channel waveguides in the first waveguide array was 5.8 μm. Assuming that the wavelength focused at the center of the end face of the output slab waveguide is the center wavelength λ ₀ , n _eq, ΔL, and λ ₀ satisfy the relationship of n _eq ΔL = mλ ₀ . Here, m is the diffraction order. By setting the diffraction order m to 10 with respect to the above values of n _eq and ΔL, the center wavelength becomes approximately 1550 nm.

A second waveguide array section was provided for the AWG wavelength multiplexer / demultiplexer designed as described above. The second waveguide array was constituted by a linear channel waveguide as shown in FIG. 2, and the length of the linear array was 100 μm. The length of refractive index modulation was set to 100 μm for the outermost waveguide and 0 μm for the innermost waveguide, and the length of the waveguide for refractive index modulation was changed during this period. The refractive index modulation amount dn _core of the core in the refractive index modulation region was changed by 0.005 in the range of −0.025 to +0.010. The wavelength of the incident light was 1540 nm.

4 shows the output port No. 0-No. It is the result of plotting how the output electric field intensity at +4 changes according to the refractive index modulation amount. Here, as shown in FIG. 0 and the right channel waveguide from there in order of port No. +1, No. +2,..., The output channel waveguide from the center to the left is connected to the port No. -1, No. 1 -2,... From FIG. 4, it can be seen that when the refractive index of the core in the refractive index modulation region is changed by 0.018, the output position changes by one port. In the AWG designed in this example, the change in the emission position for one port corresponds to a deflection angle of 2.04 degrees. Accordingly, when the length of the refractive index modulation region is 100 μm, a deflection angle of 1.13 degrees is obtained with respect to a refractive index modulation amount of 0.01. 5A and 5B show output field distributions when dn _core is set to -0.015 and +0.005, respectively. +1 and No. It can be confirmed that output light is emitted at +2. As described above, the light condensing position on the end face of the output slab waveguide is changed by the difference in the refractive index modulation amount, and the output port can be switched.

Next, in the design of the AWG element, the amount of change in deflection angle due to refractive index modulation when the diffraction order was changed and m = 20 was confirmed. The design value of the AWG element is almost the same as the above value. FIG. 6 shows the result of plotting the emission intensity from each port while changing dn _core in the range of −0.04 to +0.04 with respect to input light having a wavelength of 1540 nm. The length of the linear waveguide array is 100 μm as in the case of m = 10. From FIG. 6, it can be confirmed that the change amount of dn _core of 0.02 corresponds to the change of the emission position for one port. As in the case of m = 10, the change in emission position for one port corresponds to a deflection angle of 2.04 degrees. Therefore, a deflection angle of 1.02 degrees is obtained with respect to the refractive index modulation amount of 0.01, and almost the same result is obtained as compared with the result of the deflection characteristic of m = 10. From the above results, it can be confirmed that even if the diffraction order of the AWG element is different, the deflection angle change amount accompanying the refractive index modulation does not change.

(Design example of optical deflection element using LN thin wire waveguide)
As a method for enabling high-speed light modulation, it is effective to use a material having an electro-optic effect as a waveguide material and change its refractive index by applying a voltage. Therefore, the deflection characteristics when an electro-optic crystal material is applied as the core material of each waveguide in the AWG type optical deflection element were confirmed by two-dimensional FDTD simulation. The parameters used for the simulation are described below. Here, with respect to the parameters of the waveguide, the refractive index is assumed that lithium niobate (LN: LiNbO ₃ ), which is a typical electro-optic crystal, is used for the core of the HIC waveguide, and the surrounding cladding is SiO ₂ . Furthermore, the HIC waveguide was determined by two-dimensional approximation. The wavelength was set to 650 nm suitable for the sensitivity characteristic of the photosensitive drum, assuming that the element of the present invention was applied to an optical scanning device such as a laser printer.

A channel waveguide with LN (refractive index = 2.17) as the core and SiO ₂ as the cladding is two-dimensionalized by approximation of the equivalent refractive index, and the refractive index n _core of the waveguide core in the simulation is 1.93. The refractive index n _clad of 1.46 was determined to be 1.46. The core width w of the channel waveguide was 300 nm. For the channel waveguides constituting the waveguide array, the waveguide pitch d at the connection portion with the slab waveguide was set to 1.2 μm. The number of channel waveguides constituting the array was 24. Similarly, for a plurality of output waveguides, the waveguide interval D at the connection portion with the slab waveguide is set to 1.2 μm. The focal length (slab radius) f of the slab waveguide was 30 μm. The path difference ΔL between adjacent channel waveguides in the first waveguide array was 5.4 μm. By setting the diffraction order m in the AWG type wavelength demultiplexer to 15 for the value of ΔL, the center wavelength becomes approximately 650 nm.

With respect to the optical deflection element having the design value as described above, a change in the condensing position due to a change in refractive index was confirmed. Here, as shown in FIG. 2, a linear portion that gives refractive index modulation is provided at the center of the optical deflection element, and the length of the linear array is 100 μm. The length of refractive index modulation was 100 μm for the outermost waveguide and 0 μm for the innermost waveguide, and the length of the linearly modulated waveguide was changed during this period. The refractive index modulation amount dn _core of the core in the refractive index modulation region was changed by 0.005 in the range of −0.060 to +0.050. The wavelength of incident light was 650 nm.

  FIG. -3 to No. It is the result of plotting how the output electric field intensity at +3 changes according to the refractive index modulation amount. FIG. 8 shows the output electric field distribution when the light is most concentrated at each port. From FIG. 7, it can be read that the output channel waveguide changes by one port when the refractive index modulation amount of the core is 0.019. In the deflection element designed here, the change in the emission position for one port corresponds to a deflection angle of 2.08 degrees. Accordingly, when the length of the refractive index modulation region is 100 μm, a deflection angle of 1.1 degrees is obtained with respect to the refractive index modulation amount 0.01.

Next, the relationship between the deflection angle and the linear array length in the refractive index modulation region was similarly confirmed by a two-dimensional FDTD simulation. Specifically, it was confirmed how the output port of the output channel waveguide changes with respect to the refractive index modulation amount when the length of the linear waveguide is increased to 200 μm. The design parameters and simulation conditions are the same as those described above except that the refractive index modulation length is 200 μm for the outermost waveguide and 0 μm for the innermost waveguide, and the modulation length is changed linearly during that time. This is exactly the same as when the waveguide length is 100 μm. Output channel waveguide port No. for the refractive index change amount. -3 to No. FIG. 9 shows the result of calculating the change in output intensity up to +3. The refractive index modulation amount dn _core of the core in the refractive index modulation region was changed by 0.0025 in the range of −0.03 to +0.025.

  Comparing the result shown in FIG. 9 with the result shown in FIG. 7 when the linear array length is 100 μm, it can be seen that an equivalent output port change can be obtained with exactly half the refractive index modulation amount. This suggests that a large diffraction angle can be obtained by lengthening the linear array even when the refractive index modulation amount is small. Therefore, even if the switching element uses an electro-optic crystal having a small refractive index modulation amount as a core, the switching amount can be increased by lengthening the linear array. For example, even if the refractive index modulation amount is 0.001, if the length of the refractive index modulation region is 1 mm, a deflection angle of about 1 degree can be obtained. Furthermore, if the radius of the slab waveguide is increased, the size of the arc that can be changed according to the deflection angle can be increased even if the deflection angle is small, and in principle, the condensing spot diameter at the output end of the output slab waveguide is slab. Since it does not depend on the radius, the number of channels can be increased in a multi-channel switching element.

  Based on the simulation results of the basic deflection characteristics of the optical deflection element of the present invention, when an electro-optic crystal material such as LN is assumed as a core material, switching is performed over how many ports with the amount of change in refractive index. Verify that it is possible. As a basic precondition, in a 100 μm long linear array, the condensing position is deflected once with respect to the refractive index change amount 0.01, and the deflection amount linearly changes with respect to the length of the linear array ( If the array length is doubled, the deflection angle is also doubled), and the spot size of the beam focused at the output position does not depend on the slab radius (even if the slab radius increases, light is collected in a small spot). Assuming that

The causes of the phase difference that contributes to the change in the condensing position are collectively expressed as a change in the optical path length in the linear array. That is, in the precondition, the linear array length L and the refractive index change amount dn _core are considered as parameters that give the phase difference. These are summarized as dS = L × dn _core as the change in optical path length. For example, a refractive index change amount 0.01 in a 100 μm long linear array is dS = 1 [μm]. The unit of the optical path length change dS is a unit of length, but it should be noted that it includes a dimensionless value of the refractive index change. For example, in order to obtain dS = 1 [μm], it is also possible to give a refractive index change amount of 0.001 to an array length of 1000 μm. Therefore, even for a material with a small amount of change in refractive index such as LN, it is possible to increase the amount of change in optical path length by increasing the linear array length.

  Next, from the precondition, when the optical path length variation dS = 1 [μm], the deflection angle is 1 degree. Further, dS and the deflection angle dθ are in a proportional relationship, and dθ = 1 × dS [deg]. Increasing the amount of change in the optical path length provides a large deflection angle, but the greater the deflection angle, the greater the intensity of light that is condensed into higher-order diffracted light. It is therefore preferable to avoid extremely large deflection angles. Specifically, it is usually preferable that the deflection angle is in the range of about 5 degrees or less.

The performance of the deflecting element determines the moving distance of the light beam scanned by light deflection rather than the deflection angle itself. Even when the deflection angle is the same, if the slab radius f is different, the scanning length L _c due to deflection is also different, and the value is given by L _c = f × dθ. Here, f is a slab radius, and dθ needs to be converted into a unit of rad. The scan length can be increased in proportion to the slab radius. When a 1 × N type multi-channel switch element is constituted by the optical deflection element of the present invention, the number N of channels that can be switched is given by the above-mentioned scanning length divided by the output waveguide interval D. That is, N = L _c / D = (f / D) × (dS · π / 180). Even when the scanning length _Lc is long, the number of channels N is small if the waveguide interval D is wide. Therefore, in the output waveguide, it is effective to increase the number of channels to confine light as much as possible and narrow the waveguide interval.

  Assuming a specific value, the number of channels that can be switched is estimated. For example, when dn = 0.001 and L = 1000 μm, dS = 1.0. Assuming that D = 1.2 μm, when f = 600 μm, 1200 μm, and 1800 μm, the number of channels N is 8, 17, and 26, respectively. Therefore, for example, when f = 600 μm, switching of 8 channels becomes possible. This means that an 8-channel switching element can be realized with a minute element size of about 1 mm square. It is also possible to reversely calculate the design value from the desired number of channels. For example, if N = 256 is first determined and the value of dS is 5.0 and the waveguide interval D is 1.2 μm, the slab radius f is found to be 3.52 mm. For example, dS = 5.0 is a value that can be achieved by setting the linear array length to 5 mm and the refractive index change amount to 0.001. As described above, it is estimated that a 1 × 256 multi-channel switch element can be realized with a small area of about 10 mm square or less.

  In the description so far, the description has been made on the assumption that the optical deflection element has the waveguide array 3 that is symmetrical to the left and right. However, the waveguide array 3 does not have to be symmetrical. For example, the waveguide arrays 005 upstream and downstream of the waveguide array 006 may be linear channel waveguides or may be curved with different curvatures. Further, there may be no waveguide array 005 on the downstream side of the waveguide array 006, and the waveguide array 006 may be directly connected to the output slab waveguide 004. In this case, it can be considered that a waveguide array 005 having a length of 0 exists here. In addition, a mirror is provided in the waveguide on the downstream side after the waveguide array 006, the light entering from the entrance side is reflected, and the optical path is changed in the original direction so that the size of the light deflection element is approximately Can be halved. The position of this mirror may be anywhere in the waveguide array 006 or in the waveguide array 003 downstream of the waveguide array 006, but it is preferable in terms of design and manufacture to be at the center of the optical path of the waveguide array 006.

(Embodiment 1)
As described above, the optical deflection element according to the present invention gives a phase difference between adjacent channel waveguides by changing the equivalent refractive index of a linear waveguide array of the order of millimeters. Therefore, it is important that the equivalent refractive index of each channel waveguide when the refractive index modulation is not performed is uniform. When the non-uniformity of the equivalent refractive index of the channel optical waveguide is larger than the refractive index modulation amount of the material, the phase is not aligned when the diffracted light is collected in the output slab waveguide, and the focused spot cannot be reduced.

  The value of the equivalent refractive index of the optical waveguide depends on the refractive index of the material and the size of the core, but when an electro-optic crystal such as LN or silicon is applied to the core, the core is formed by processing from a bulk crystal. The uniformity of the refractive index is extremely high. Therefore, it is the dimensional variation of the core that affects the non-uniformity of the equivalent refractive index. FIG. 10 shows the result of calculating the equivalent refractive index of the optical waveguide by the mode simulation by the three-dimensional beam propagation method (BPM) for the channel waveguide structure assuming that the core material is LN. Here, the relationship between the core width and the equivalent refractive index when the core thickness is fixed at 200 μm was calculated. The refractive indexes of the core and the clad were assumed to be 2.2818 and 1.4567, respectively, and the light source was TM polarized light with a wavelength of 650 nm. Further, in FIG. 11, the amount of change in equivalent refractive index when the core width is changed by 1 nm is plotted against the core width.

In a normal waveguide type device, the core size is designed so that the channel waveguide satisfies the single mode condition. Assuming the above parameters, about 300 nm is appropriate as the core width that satisfies the single mode condition. In such a core size, it can be seen from FIG. 11 that when the core width changes by 1 nm, the equivalent refractive index changes by about 6 × 10 ⁻⁴ . For an electro-optic crystal material such as LN, the refractive index that can be changed by applying a voltage is expected to be on the order of about 10 ^{−3 at} maximum. Therefore, when a linear waveguide array is constituted by a normal HIC optical waveguide, it is required to manufacture the core with an accuracy of about 1 nm or less. However, this does not necessarily require that the accuracy of 1 nm or less be maintained over the entire linear waveguide, and it means that the average value of the accuracy of the core width of the linear portion must be 1 nm or less.

  In order to increase the tolerance of the core width in the linear waveguide array, it is effective to reduce the change in the equivalent refractive index of the channel waveguide due to the change in the core width. A simple method is to increase the core width of the channel waveguide only in the linear array portion. From FIG. H, for example, if the core width is increased to 800 nm, the amount of change in the equivalent refractive index with respect to the core width change of 1 nm can be reduced to about 1/10 compared with the case where the core width operating in the single mode is 300 nm. This corresponds to an increase in the tolerance for the core width by 10 times.

  With respect to the deflection element in which the core width of the linear array portion is increased, only the waveguide array portion is enlarged and schematically shown in FIG. In the first embodiment of the optical deflecting element of the present invention, as shown in FIGS. 2 and 12, the first waveguide array 0102 and the second waveguide array 0103 are channel guides constituting the respective waveguide arrays. The waveguides are coupled to each other. As a feature of the first embodiment, the channel waveguides constituting the first waveguide array 0102 have a core width that satisfies the single mode condition, and the channel waveguides constituting the second waveguide array 0103 Has a wider core width. Further, the respective channel waveguides having different core widths are coupled via the core width conversion unit 0104.

  In the second waveguide array 0103 with the expanded core width, the channel waveguide is a multi-mode waveguide, so that higher-order propagation modes may be excited. However, when only the fundamental mode matching the propagation mode of the waveguide is input to the center of the multi-mode waveguide, the higher-order mode is not excited in principle if it is a straight waveguide. Therefore, if the core width of the channel waveguide in the first waveguide array 0102 satisfying the single mode condition is gradually increased and coupled to the second waveguide array 0103, the excitation of the higher-order mode can be suppressed to the minimum. it can. In FIG. 12, the core width is increased by using a linear taper shape, but the shape for increasing the core width is not limited to this, and for example, a parabolic shape is effective in suppressing the occurrence of higher-order modes.

  The change of the light collection performance to the output port due to the structure for expanding the tolerance was confirmed by the two-dimensional FDTD method. Specifically, a comparison was made as to how much the output light intensity from each output port in the optical deflection element differs depending on the presence or absence of the tolerance expansion structure.

  The parameters used for the simulation are the same values as when the above-mentioned LN is used for the waveguide core. Here, the core width in the second waveguide array was expanded to 800 nm, and the channel waveguide in the first waveguide array having a core width of 300 nm was coupled by a linear taper having a taper length of 10 μm. The linear array length was 100 μm. The other parameters in the simulation were set on the assumption that the optical deflection element was constituted by an LN thin wire waveguide. FIG. 14 shows the result of confirming whether the output intensity from each output channel waveguide changes depending on the presence or absence of the tolerance relaxation structure. The condensing performance of a conventional deflecting element, that is, a deflecting element in which the second waveguide array is composed of a single mode waveguide, is indicated by a broken line, and the deflecting element having an increased tolerance by introducing a multimode waveguide array The light condensing performance is shown by a solid line. In both cases, port no. Light is condensed at +1, and there is almost no difference in output light intensity. Also, the difference from the output at the adjacent output port is 20 dB or more, and there is no deterioration of the light collecting performance due to the tolerance expansion configuration. This is due to the fact that high-order guided modes are hardly excited in the linear array portion.

  Furthermore, in the above-described expanded form of tolerance, the movement of the output port due to the refractive index modulation of the core of the linear arrayed waveguide was confirmed by a two-dimensional FDTD method. The calculation results are shown in FIG. For example, output port No. -1 to No. In order to change the output position to +1 by two ports, the refractive index of the core should be changed by 0.035. When the 1-port output position is different, the deflection angle is about 2.08 degrees, and therefore the deflection angle per refractive index modulation amount 0.01 is estimated to be 1.19 degrees. This value is about the same as the deflection angle (1.1 degrees) when the linear array is formed of a single mode waveguide. However, by increasing the core width of the channel waveguide for refractive index modulation, the optical confinement to the core with respect to the fundamental mode becomes stronger, so the amount of refractive index modulation of the core compared to the case of a linear array with a single mode waveguide The deflection angle due to is slightly larger.

(Embodiment 2)
As another method for increasing the tolerance against the change in the core width of the channel waveguide, it is effective to reduce the relative refractive index difference of the channel waveguides constituting the linear array. If the refractive index difference between the core and the clad is reduced, the range of the maximum value and the minimum value that can be taken by the equivalent refractive index is narrowed, so that the amount of change in the equivalent refractive index when the core width changes is also reduced. However, if the relative refractive index difference is reduced, it is necessary to increase the radius of curvature in the bent waveguide, so that the element size increases in a normal AWG device. However, in the optical deflecting element according to the second embodiment, only the second waveguide array portion configured by the linear channel waveguide is configured by the waveguide having a low relative refractive index difference, and the tolerance in the channel waveguide is increased. It is possible to avoid an increase in device size.

  FIGS. 13A and 13B are enlarged views of only the waveguide array portion of the optical deflecting element in the case where the channel waveguide constituting the waveguide array that performs the refractive index modulation is constituted by a low relative refractive index difference waveguide. It is shown. Refer to FIG. 2 for the entire optical deflection element. The first waveguide array 0202 and the second waveguide array 0203 in the optical deflecting element of this embodiment are such that channel waveguides constituting each waveguide array are coupled to each other. Here, by making the refractive index of the clad of the second waveguide array higher than the refractive index of the clad of the first waveguide array, it is possible to easily integrate channel waveguides having different equivalent refractive indexes. . The first waveguide array and the second waveguide array are coupled via a core width changing unit 0204. Since channel waveguides having different relative refractive index differences have different core widths that satisfy the single mode condition, it is normal that the channel waveguides in the first waveguide array and the channel waveguides in the second waveguide array have different core widths. It is. Therefore, it is preferable to combine them by gradually changing the core width. Similarly, in the first and second waveguide arrays, since the refractive indexes of the clads are different, it is preferable to form the clads so that they are gradually replaced. As a method of changing the clad, for example, the form of linear change is shown in FIGS. 13A and 13B, but other forms such as a parabolic shape may be used. Further, in both FIGS. 13A and 13B, the core width and the cladding are changed by a linear taper, but it is also effective to change the core width and the cladding region by, for example, a parabolic shape or an elliptical shape. It is.

  FIGS. 16A and 16B schematically show channel waveguide cross sections in the second waveguide array. A channel waveguide core 0211 is formed on the first cladding 0212. The first clad is a clad common to all the waveguides in the optical deflection element of this embodiment. Therefore, the difference between the refractive index of the core and the refractive index of the first cladding is large. The channel waveguide shown in FIG. 12A has a structure in which the upper portion and the side portion of the core 0211 are surrounded by the second cladding 0213. By making the refractive index of the second cladding close to the refractive index of the core, the relative refractive index difference of the optical waveguide is reduced. In the channel waveguide shown in FIG. 12B, the side of the core is buried with the second clad 0213, and the upper part of the core is covered with the third clad 0214. In the structure shown in FIG. 12B, by making the refractive indexes of the first clad 0212 and the third clad 0214 the same, the electromagnetic field distribution of the propagation mode in the channel waveguide can be made vertically symmetrical. The clad that sandwiches the core from above and below has a low refractive index, and the confinement of the channel waveguide in the film thickness direction becomes strong. On the other hand, since the left and right claddings of the core have a refractive index close to that of the core, the lateral confinement becomes weak. Since both the electromagnetic field distributions of the propagation modes in the first and second waveguide arrays are vertically symmetrical, the transition of the propagation modes when they are coupled is facilitated, and the core width in the channel waveguides constituting both arrays The taper length when changing the length can be shortened.

The effect of increasing the tolerance by reducing the relative refractive index difference of the channel waveguides constituting the second arrayed waveguide is estimated. FIG. 17 shows the result of calculating the change in the equivalent refractive index of the waveguide due to the change in the core width of 1 nm for the channel waveguide having the cross-sectional structure shown in FIG. The horizontal axis represents the core width of the waveguide, and the effect of expanding the tolerance by changing the refractive index of the second cladding 0213 from 1.6 to 2.2 was compared. The refractive index and thickness of the core are 2.2818 and 200 nm, respectively, and the refractive index of the first and third claddings is 1.4567. The wavelength was 650 nm. For example, when the refractive index of the second cladding is 2.2, if the core width is increased to about 600 nm or more, the equivalent refractive index change per 1 nm of the core width change becomes 5 × 10 ⁻⁴ or less, and a typical LN thin wire Compared with the waveguide, the tolerance can be increased to about 10 times. Furthermore, since the channel waveguide at this time satisfies the single mode condition, the higher order mode is not excited, and there is no deterioration in the element light condensing performance due to the increased tolerance.

(Embodiment 3)
FIG. 18 schematically shows the second waveguide array in the present invention. The waveguide array 0401 includes a first waveguide array 0402 and a second waveguide array 0403. Furthermore, the second waveguide array has a photonic crystal arrangement. The photonic crystal array has a periodic refractive index distribution that is about the wavelength of light, and a photonic band gap that is a forbidden body of photons can be formed by adjusting the periodic structure. At this time, no light can exist in the region where the photonic crystal array exists in the second waveguide array. As a specific photonic crystal arrangement, it is preferable to provide periodic circular holes in the waveguide material as shown in FIG. When a line defect waveguide which is a straight line portion without a circular hole is introduced into the photonic crystal arrangement, light is strongly confined in the line defect waveguide and propagates. Therefore, an optical waveguide can be formed by this photonic crystal line defect waveguide 0405.

  With the photonic crystal line defect waveguide, light can be strongly confined in the line defect, and a minute optical circuit can be formed. Furthermore, by appropriately designing the photonic crystal array, the group velocity of light propagating through the photonic crystal line defect waveguide can be reduced to about 1/10 to 1/100 of the light propagating in the vacuum. is there. This corresponds to the fact that the equivalent refractive index of the photonic crystal line defect waveguide is extremely larger than that of a normal waveguide. Therefore, it is considered that the amount of phase difference imparted between the waveguides due to the change in the refractive index of the core material is also larger than that of the channel waveguide using the optical confinement by total reflection. By utilizing this effect, the waveguide length and refractive index modulation region of the second waveguide array in the optical deflection element of the present invention can be reduced, and the device size of the deflection element can be extremely reduced.

(Embodiment 4)
FIG. 19 schematically shows a second waveguide array according to the fourth embodiment of the optical deflection element of the present invention. The second waveguide array 0503 is coupled to the first waveguide array 0502 via the core width changing unit 0504. The core 0505 of the channel waveguide constituting the waveguide array has a larger core width in the second waveguide array. Further, a light absorption layer 0506 that is an electroabsorption layer is provided between the cores of the plurality of channel waveguides.

  As described above, in the second linear array, the tolerance for the core width can be increased by increasing the core width of the channel waveguide, but at the same time, a higher-order propagation mode may be excited in the channel waveguide. . Unexpected stray light may be generated when higher-order modes excited in the respective channel waveguides constituting the array interfere with each other in the output slab waveguide. Therefore, in order to suppress the generation of stray light with respect to the condensed beam due to the interference of the fundamental mode, it is preferable to quickly remove the higher order mode when it occurs in the waveguide array. The light absorption layer 0506 is added to absorb and remove the higher-order propagation mode.

  Compared with the fundamental mode, the higher-order mode has weak optical confinement in the core, and the propagation field is widely distributed throughout the cladding. Therefore, by disposing the light absorption layer in a region away from the core to such an extent that the electromagnetic field in the fundamental propagation mode is sufficiently attenuated and the electromagnetic field distribution in the higher order propagation mode is not attenuated completely, only the higher order mode is absorbed. Can be attenuated by layers. Therefore, even if a higher-order mode is generated in the second waveguide array, it is possible to reduce the deterioration of light collection in the output slab due to this. As a specific material for the light absorption layer, a dielectric having a higher refractive index than the cladding, a metal that absorbs an electromagnetic field, or the like is preferable.

(Embodiment 5)
As can be seen from FIGS. 1 and 2, the optical deflecting element of the present invention can be symmetric except for the form of the entrance port and the exit port. In the fifth embodiment, using this characteristic, for example, in FIG. 2A, the left side is removed from the central portion of the second waveguide array 0006, and a mirror that reflects light is connected to the removed portion. . In order to make the input slab waveguide 0002 function as the output slab waveguide 0004, an output channel waveguide 0009 is provided. In the optical deflection element of this form, the light incident from the input channel waveguide 0001 reaches the mirror via the input slab waveguide 0002, the first waveguide array 0005, and the second waveguide array 0006. The light reaching the mirror is reflected by the mirror and reaches the incident surface of the input slab waveguide 0002 through a path opposite to that incident from the second waveguide array 0006. At this time, the emitted light can be shifted from the position of the input channel waveguide 0001 by the refractive index modulation action of the waveguide array. If the output channel waveguide 0009 is provided at that position, the input slab waveguide 0002 can be used as the output slab waveguide 0004, and light can be deflected by the refractive index modulation action. This form of light deflection element can be made of approximately half the size and material of the light deflection element described above.

(Embodiment 6)
FIGS. 20A and 20B schematically show the connection portion between the input slab waveguide or output slab of the optical deflection element of the present invention and the waveguide array. Here, the end surface of the input or output slab waveguide 0701 is a curved surface represented by an arc having a certain curvature. A plurality of channel waveguides constituting the waveguide array 0702 are connected perpendicularly to the curved surface of the input or output slab waveguide 0701, respectively.

  When the input or output slab waveguide 0701 is an input slab waveguide, light incident from the input channel waveguide spreads by diffraction in the slab waveguide, and is input to the channel waveguides that constitute the waveguide array in substantially the same phase. Is done. Here, if the light diffracted in the slab waveguide is scattered without being coupled to the arrayed waveguide, it leads to an increase in the loss of the element, and if an extra diffracted wave due to the scattering is generated, the cause of the deterioration of the crosstalk level It also becomes. Therefore, it is preferable to couple the propagation light with as low a loss as possible at the connection portion between the slab waveguide and the arrayed waveguide.

Therefore, as shown in FIGS. 20A and 20B, it is effective to increase the core width of the channel waveguide at the connection portion between the slab waveguide and the channel waveguide. If the core width other than the end of the channel waveguide is w, the core width of the end face incident on the slab waveguide is w _2, and the propagation length when the core width changes is l ₂ , the core width w ₂ of the end face is As the size is increased, the light propagated through the slab waveguide can be input to each channel waveguide, so that the excessive loss of light of the optical deflection element can be reduced. Further, if the length of the propagation length l ₂ when the core width changes is set to about several microns or more, excess loss due to changing the core width can be made substantially zero. In addition, the linear taper shape as shown in FIG. 20 (a), the parabolic shape as shown in FIG. 20 (b), and other elliptical shapes are also effective as the core width.

(Embodiment 7)
FIGS. 21A and 21B schematically show a connection portion between an input channel waveguide and an input slab waveguide. The input channel waveguides 0711 are connected perpendicularly to the curved surface of the input slab waveguide 0712. Further, the core width is widened at the connection portion between the input channel waveguide and the input slab waveguide.

The core width of a portion other than the end portion of the input channel waveguides 0711 w, the core width of the end face of the input channel waveguides when entering the input slab waveguide 0712 and w _1, the propagation length when the core width changes l _Set to ₁ . If the core of the input channel waveguide is coupled to the slab waveguide while w remains, the light that is strongly confined in the input waveguide is diffracted in the slab waveguide. . Therefore, when the diffracted light propagates through the slab waveguide and reaches the arrayed waveguide, the phase of the light incident on each arrayed waveguide is aligned. However, since diffraction extends over a wide range, if the number of arrayed waveguides is small, diffracted light in the slab waveguide cannot be picked up, leading to excess loss of incident light. On the other hand, the effect of narrowing the diffraction angle in the input slab waveguide 0712 by widening the core width w ₁ of the end face of the input channel waveguide and widening the field width when coupled to the input slab waveguide 0712 in advance. There is. As a result, the number of channel waveguides constituting the waveguide array can be reduced, which is effective for reducing the size of the optical deflection element. However, since the spot width of the light collected at the output port of the optical deflection element is about the same as the field width at the input port, if the input channel waveguide width is too wide, the number of resolution points at the output port cannot be obtained. . Actually, the value of w ₁ should be several times larger than w.

  For increasing the core width in the input channel waveguide 0711, a linear taper shape as shown in FIG. 21A, a parabolic shape as shown in FIG. By expanding the core in a parabolic manner, it is possible to suppress the generation of higher-order modes due to changes in the core width, and to prevent higher-order modes from being excited in the slab waveguide.

(Embodiment 8)
FIGS. 22A and 22B schematically show a connection portion between the output slab waveguide and the output channel waveguide in the embodiment shown in FIG. The output slab waveguide 0721 has an end face represented by an arc having a certain curvature, and a plurality of output channel waveguides 0722 are vertically connected to the end face. Further, the core width is widened at the connection portion between the output channel waveguide and the output slab waveguide.

  The light beam is collected at the end (exit port) of the output slab waveguide 0721, but the spot diameter is substantially equal to the light beam incident on the input slab waveguide from the input channel waveguide. Therefore, in order to efficiently couple the light focused on the output slab waveguide 0722 to the output channel waveguide 0722, the core width of the end face of the output channel waveguide is set to the input channel waveguide at the connection portion with the input slab. It is desirable to have the same width as the core width of the end face. Further, when the output light coupled from the output slab waveguide 0721 is guided to the output channel waveguide 0722, the waveguide width is narrowed in a tapered shape in order to suppress light emission and high-order mode excitation. As a shape for changing the waveguide width, in addition to the linear taper shape shown in FIG. 22A, a parabolic shape shown in FIG. 22B is preferable in order to suppress excitation of higher-order modes. .

In the two-dimensional FDTD simulation shown above, in consideration of the above matters, the core width of the channel waveguide is increased at the connection portion between each channel waveguide and the slab waveguide. Specifically, in the connection portion, the core width w ₁ of the end face of the input channel waveguide is 1.5 μm, the core width w ₂ of the end face of the waveguide array is 1.0 μm, and the core width of the end face of the output channel waveguide is 1 It was set to 0.0 μm. Further, the taper lengths l _{1 to} l ₃ accompanying the change in the core width were all set to 3.0 μm.

(Embodiment 9)
FIG. 23 schematically shows a coupling portion between the output slab waveguide and the output channel waveguide in the deflection element that guides the deflected light beam by the output channel waveguide as shown in FIG. A plurality of channel waveguides 0802 are coupled vertically to the output curved surface of the output slab waveguide 0801, respectively. FIG. 23 shows a state in which the core width of the channel waveguide is expanded in a parabolic manner at the coupling portion with the output slab waveguide 0801, but this connection portion is not limited to this shape but is tapered. An inclined structure such as

  The input and output slab waveguides have a fan-shaped shape, and the center of curvature is at the input channel waveguide end or the center waveguide end of the plurality of output channel waveguides. They are arranged radially to pass through the center of curvature. The distance from the center of curvature to the exit end face of the slab waveguide is referred to as the slab radius, and is represented by f here. Usually, the plurality of output channel waveguides are arranged on a curved surface having a radius of curvature of the slab radius f, and the center of curvature is the center channel waveguide end in the waveguide array coupled to the output slab waveguide. When the center of curvature is arranged as described above, in principle, the end of the output channel waveguide is positioned where the output light is most concentrated. However, in reality, the light diffracted in the input slab waveguide does not completely enter the waveguide array in the same phase, or the optical path length difference given to each channel waveguide of the waveguide array is not completely constant, etc. For the reason, the output light is not always collected most at the output channel waveguide position. That is, the output light may be collected on a curved surface inside or outside the ideal light collection position. In such a case, the beam spot diameter of the output light becomes large at the end face of the output slab waveguide, and in the adjacent output channel waveguide, crosstalk between channels, that is, light to a port other than the original output port Leakage increases.

  In order to solve this problem, it is effective to move the position of the output channel waveguide from the above position. Specifically, there is a method of making the slab radius f ′ in the output slab waveguide different from the input slab radius f. When f ′ <f, the output channel waveguide is located inside than usual. Conversely, when f ′> f, the output position can be set outside the normal position. Further, even when f ′ = f, that is, when the slab radius of the output slab waveguide is equal to the input slab radius, the curvature center of the curved surface on which the output channel waveguide is arranged is moved from the end of the normal waveguide array. Can change the output position. In FIG. 23, 0803 is the amount of change in the center of curvature, and is represented here by δ. The exit curved surface can be moved inward or outward as a whole depending on the value of δ.

  FIG. 24 shows how the interference light is condensed on the output channel waveguide in the output slab waveguide. In this case, the output beam is output port no. The optical path length difference between adjacent channel waveguides in the refractive index modulation region was adjusted so as to be output from 1. FIG. 24A shows a state in which diffracted light is collected when no adjustment is made with respect to the output end face position of the output slab, that is, when the output slab waveguide has the same structure as the input slab waveguide. It can be seen that the point where the output light is most focused is slightly inside the position of the output waveguide (plus side in the z coordinate in the figure). Therefore, there is a possibility that crosstalk between adjacent ports of output light can be reduced by moving the z coordinate of the output waveguide. FIGS. L (b) and (c) show how the output light is collected when the centers of curvature of the output slab waveguide end face are moved inward by 1 μm and 2 μm, respectively (plus side in the z coordinate). Here, the displacement amount of the center of curvature is expressed as δ = −1 μm and δ = −2 μm, respectively. By moving the position of the output slab waveguide, the leakage of light to the adjacent port can be reduced.

  FIG. 25 shows these results in numerical form. The horizontal axis represents the output port number, and the vertical axis represents the value obtained by normalizing the output intensity from each port with the input light intensity. When the output slab waveguide position is not moved, the port No. The output intensity from 1 is as small as −5.0 dB, and the light leakage to the adjacent port is as large as about −16 dB. On the other hand, by moving the position of the output slab waveguide by 2 μm, the port no. The output intensity from 1 could be increased to -3.2 dB, and light leakage to the adjacent port could be reduced to -30 dB or less. Here, in FIG. The output intensity from -7 is larger than the output intensity from other ports. This is because the first-order diffracted light is condensed at -7. By increasing the radius f of the slab waveguide and bringing the condensing position of the first-order diffracted light outside the output port, it is possible not to be affected by such higher-order diffracted light. .

(Embodiment 10)
FIGS. 26A to 26C show examples of the cross-sectional structure of the waveguide array in the refractive index modulation region. The core 0902 of the channel waveguide causes a refractive index change by the electro-optic effect. By sandwiching the channel waveguide core 0902 from above and below by the electrode 0904 and applying a voltage, an electric field can be formed in a direction perpendicular to the channel waveguide core 0902. The electrode material may be a general metal material, and metals such as Cr, Au, Cu, and Al are suitable. However, since the propagation light of the channel waveguide is absorbed by the electrode material when the core and the electrode are brought into contact with each other, the channel waveguide cladding 0903 serving as a buffer is formed between the electrode and the core. The cladding material, low dielectric material preferably having a refractive index, specifically and SiO _2, mixed glass material and _Ta 2 _{O 5} and TiO ₂ and SiO _2, silicon oxynitride (SiON), preferably polymers, and It is. The closer the distance between the electrode and the core, the lower the voltage required for refractive index modulation. However, since the propagation light in the waveguide is not affected by the electrode, it is preferable to separate the two by about 1 μm. The upper electrode may be disposed so as to sandwich the entire waveguide array as shown in FIG. 26 (a), or the channel leads constituting the waveguide array as shown in FIG. 26 (b) or 26 (c). You may arrange | position only on the upper part of the core of a waveguide. Furthermore, as shown in FIG. 26 (c), the clad material sandwiching the core from above and below may be different from the clad material of the side surface of the core. In FIGS. 26 (a) to 26 (c), voltage sources are arranged so that equal voltages are applied between all the electrodes. In FIGS. 26 (b) and 26 (c), Different voltages may be applied independently to each channel waveguide.

(Method for Manufacturing Optical Deflection Element of the Present Invention)
The manufacturing process of the optical deflection element of the present invention will be described below. In addition to an optical crystal material such as LN or LT, a ceramic material such as PLZT or a polymer material having an electro-optic effect can be applied as the core material. Here, assuming a case where an electro-optic crystal is used as a core, a thin film is formed from a bulk crystal material, and the obtained optical crystal thin film is patterned to form a channel. The manufacturing process can be broadly divided into two steps: optical crystal thinning process, waveguide patterning and electrode formation.

  First, a process for forming an optical crystal thin film will be described. Here, the electrode layer and the cladding layer are sandwiched between the thin film and the supporting substrate holding the thin film. These are hereinafter referred to as the lower electrode and the lower cladding. The simplest process for forming a thin film from a bulk optical crystal is a precision polishing method. By high-precision polishing, for example, thinning can be performed from an optical crystal substrate having a thickness of about 0.5 mm to 1.0 mm to a thickness of 1 μm or less. FIG. 27 shows a thinning process using polishing. First, a lower clad material and a lower electrode material are sequentially formed on an optical crystal substrate. The electrode material is preferably a metal material, and a vacuum deposition method or a sputtering method is suitable as the film formation method. Further, as a method for forming the lower clad, a sputtering method, a chemical vapor deposition (CVD) method, a vacuum deposition method, or the like is preferable. Next, a support substrate is directly bonded or bonded to the lower electrode side of the optical crystal substrate on which the lower cladding and the lower electrode are formed. The support substrate material is preferably the same optical crystal as the core material, but may be, for example, silicon or synthetic quartz. As a bonding method at this time, heat bonding under pressure or room temperature bonding can be applied. At this time, it is also effective to add a dielectric layer or a metal layer having a low melting point for facilitating mutual bonding to the lower electrode layer or the support substrate surface. In addition, the optical crystal substrate and the support substrate may be bonded together with an adhesive more simply. For the bonded substrate thus obtained, the optical crystal portion is thinned by polishing, so that the optical crystal thin film is finally formed on the support substrate so that the lower electrode layer and the lower cladding layer are sandwiched between them. To do.

  As another method for obtaining a thin film from a bulk optical crystal, there is an ion slicing (Crystal Ion Slicing: CIS) method shown in FIG. The ion slicing method is a method of forming a thin film from a bulk crystal by the following steps. First, hydrogen ions or helium ions having a small mass number are implanted into the optical crystal with high energy. Ions penetrate the substrate surface and form an implantation layer at a depth corresponding to the implantation energy from the surface. In the injection layer, innumerable minute cracks (microcavities) are generated in a direction parallel to the substrate surface. When the ion-implanted substrate is heated at an appropriate temperature and time, the microcavities formed by ion implantation are dissolved into each other by an effect called Ostwald ripening to grow into large cavities. By utilizing the growth of cracks in the implantation layer near the substrate surface, the thin film is transferred from the bulk substrate to the support substrate by heating the ion-implanted substrate bonded to the support substrate. It is possible.

  The film thickness of the thin film obtained by the ion slicing method is determined by the ion implantation depth from the surface of the bulk substrate, and this depth can be controlled by the ion implantation energy. Specifically, a lithium niobate thin film having a thickness of about 750 nm is obtained by injecting helium ions having an energy of 200 keV into a lithium niobate substrate. Further, since the implantation depth is uniform within the substrate surface, the finally obtained thin film is a uniform thin film with almost no variation in the film thickness within the surface. FIG. 29 shows an SEM image of the LN thin film formed by the ion slice method. A thin film having a thickness of 750 nm is formed. Thus, by controlling the implantation energy, it is possible to form an LN thin film having a film thickness of 1 μm or less with extremely high film thickness accuracy.

  FIG. 28 shows a process for forming an optical crystal thin film using the ion slice method. First, ions are implanted into the optical crystal substrate to form an ion implantation layer. Next, a lower cladding layer and a lower electrode layer are sequentially formed. However, the lower cladding layer may be formed before ion implantation. The ion implantation substrate and the support substrate are bonded. The above film formation and bonding are performed at a relatively low temperature such that the thin film does not peel off. By heating the bonded substrates, the thin film is peeled off with the ion implantation layer as a boundary. Since the surface of the thin film immediately after peeling usually has a roughness of about 100 nm, it is preferable to smooth and polish the thin film surface last. As the polishing process, a method such as CMP (Chemical Mechanical Polishing) or laser processing may be used other than normal optical polishing.

  The deflection element is completed by performing fine patterning on the optical crystal thin film with the lower electrode and the lower clad formed by the above procedure. The manufacturing process is shown in FIGS. 30 and 31 show an example of a manufacturing process for forming the electrode configuration shown in FIGS. 27 and 28, respectively. In FIG. 30, a material to be used as a mask when an optical crystal thin film is channelized by etching is first formed. As the mask material, a metal material such as Cr or a polymer material such as a photoresist is suitable. Next, the deposited mask is patterned by a lithography technique that uses a semiconductor process. For example, a patterning process such as electron beam lithography or photolithography can be applied. A lift-off process can also be applied to mask patterning. The patterned mask is transferred to the optical crystal thin film by an etching process, and the channelization of the optical crystal is completed. Etching processes include RIE (Reactive Ion Etching), ICP (Inductively Coupled Plasma) etching, FIB (Focused Ion Beam) etching, NLD (Magnetic NeutralL Etching, etc.). A dry etching process is preferred. FIG. 32 shows an example in which a fine pattern is formed on the surface of lithium niobate. A line pattern having a line width less than the wavelength of light or a photonic crystal array pattern having a period less than the wavelength of light can be formed.

After channeling the optical crystal, the mask material is peeled off and an upper clad is formed. When the upper clad material is a glass material such as SiO _2, the film can be formed by a high frequency sputtering method, a CVD method or the like. Further, the upper clad may be formed of a polymer material. In that case, the upper clad can be formed by a spin coating method. Finally, an upper electrode is formed, and the upper electrode is patterned so that the region where the refractive index modulation is performed between adjacent channel waveguides changes constantly. The upper electrode can be formed by the same method as the lower electrode, and a lithography technique used in a semiconductor process can be applied to the patterning.

  In the process example shown in FIG. 31, first, an upper clad and a mask material are formed on an optical crystal thin film. Here, a metal thin film such as Cr is used as a mask material. The metal mask finally functions as the upper electrode. The formed mask material is patterned, and the formed pattern is transferred to the upper cladding and the optical crystal thin film by dry etching. Next, a cladding layer for embedding the channeled optical crystal is formed, and finally the upper electrode is patterned into a shape suitable for refractive index modulation. All of the individual steps used in these processes can apply the same technique as that used in the manufacturing process of FIG.

(Embodiment 11)
FIG. 33 shows a first configuration example of an optical deflection module using an optical deflection element according to the present invention. Light from the laser light source enters the light deflection element via the optical coupler, and deflects the light beam. Light may be guided between the light source and the optical coupler by, for example, coupling using an optical fiber, or in the air while maintaining the shape of the light beam via an optical element such as an appropriate mirror or lens. It may be propagated.

  When the light beam from the light source is guided by an optical fiber, the optical coupler performs a function of coupling the end face of the optical fiber and the optical deflection element with high efficiency. The input channel waveguide in the optical deflecting element according to the present invention is preferably a waveguide having a large refractive index difference between the core and the clad (hereinafter referred to as a high Δ waveguide). When the fiber is directly connected to the optical deflection element, the coupling loss increases. Therefore, the optical coupling element is an element that couples a waveguide having a small relative refractive index difference (hereinafter referred to as a low Δ waveguide) and a high Δ waveguide with high efficiency. The low Δ waveguide is formed by surrounding the end face of the high Δ waveguide with the low Δ waveguide, and gradually narrowing the core of the high Δ waveguide in a tapered shape to disappear in the region surrounded by the low Δ waveguide. Can be guided to the high Δ waveguide with high efficiency. In this case, the optical coupling element and the optical deflection element are formed integrally on a single substrate. By coupling the end face of the optical fiber with an optical coupling element, light can be guided to an input channel waveguide with high efficiency and a large relative refractive index difference.

  There is also a method for guiding light that has propagated in space to a high Δ waveguide with high efficiency. By disposing the lens having a high NA in the optical axis direction that is not parallel to the light propagation direction, it is possible to collect light on the output end face of the high Δ waveguide. Since the high Δ waveguide has a high NA, it is possible to guide spatially propagated light to the high Δ waveguide by optimally setting the NA of the waveguide and the NA of the lens. By applying this principle to the element of the present invention, the light emitted from the light source can be propagated in the air and guided to the input channel waveguide.

  The light guided to the light deflection element is given a deflection angle corresponding to the refractive index change amount in the refractive index modulation region, and is emitted from the light deflection element. The amount of change in refractive index to be applied can be controlled by the amount of voltage applied to the refractive index modulation region. By using an electro-optic material for the core material of the waveguide, high-speed refractive index modulation is possible, that is, high-speed variable light deflection is possible. Such an optical deflection module can be applied as a light source for optical writing such as a laser printer. The light source is not necessarily a continuous wave, and may be an optical pulse train subjected to intensity modulation, phase modulation, or the like. That is, the optical module according to the present invention can be applied as a switching element in optical fiber communication.

Embodiment 12
FIG. 34 shows a second configuration example of an optical deflection module using the optical deflection element of the present invention. The portion for guiding the light from the light source to the light deflecting element is the same as that of the first module. The light deflection element has a condensing position depending on the wavelength of the light source. That is, the condensing position in the output slab waveguide when the refractive index modulation is not performed differs depending on the wavelength of the input light. When the refractive index modulation is not performed, the emission position is preferably at the center of the output slab waveguide. If light is emitted from the center when refractive index modulation is not performed, light deflection at the same angle to the left and right is possible by simply changing the voltage application direction during refractive index modulation. Therefore, the temperature of the entire deflection element is controlled so that the light is emitted from the center of the output slab waveguide in the initial state, that is, in the state where refractive index modulation is not performed. When the temperature of the deflection element changes, the refractive index of the material constituting the element also changes, so that the optical path length difference given in the waveguide array changes. Therefore, the condensing position of the emitted light changes. By using this effect, the emission position in the initial state can be adjusted to be the center of the slab waveguide. As a method for giving a temperature change to the element, for example, it is convenient to place a Peltier element close to the optical deflection element. Furthermore, since the optical deflecting element of the present invention has a HIC waveguide as a constituent element, the optical deflecting element has a very small area and requires less power for temperature control. In addition, the fact that the temperature control in a minute region is sufficient leads to an increase in the stability of the module, and the overall size is also reduced.

(Embodiment 13)
FIG. 35 and FIG. 36 show an embodiment of an optical switch module using an optical deflection element according to the present invention. In the present embodiment, an optical amplifier is included in either or both of the front stage and the rear stage of the optical deflection element. In addition, the light from the light source is subjected to light modulation such as intensity modulation and phase modulation, and further includes the case where the light has propagated through a medium such as an optical fiber. Regarding such propagating light, the optical power before being input to the optical deflection element may be abnormally small. Also, optical loss is incurred when the path is switched by the optical deflection element. Therefore, before the light is input to the optical deflecting element, or after the light path is switched by the optical deflecting element, the light is amplified by the optical amplifier, thereby increasing the small input optical power and compensating for the excess loss due to the optical deflection. Is effective. As a specific optical amplifier, an optical fiber amplifier or a waveguide amplifier using an optical fiber doped with erbium or the like is suitable.

(Embodiments 14 and 15)
The optical deflection method of the present invention is based on optical deflection in a structure in which an input slab waveguide having at least one input port, a waveguide array composed of a plurality of channel waveguides, and an output slab waveguide are sequentially connected. And The waveguide array is composed of a part in which a certain optical path length difference is provided between adjacent channel waveguides and a part in which the optical path lengths of adjacent channel waveguides are made equal, and the refractive index modulation region is added to the waveguide array having the same optical path length. Is provided. The core width of the channel waveguide of the waveguide array provided with the refractive index modulation region is widened (Embodiment 14), or the relative refractive index difference between the core and the clad is reduced (Embodiment 15). . The effect of refractive index modulation can be increased, the element size can be reduced, and the tolerance for the width of the channel waveguide can be increased.

  When the refractive index modulation is not performed in the refractive index modulation region, the light input from the input channel waveguide is collected at a predetermined position at the output slab exit end. Here, by modulating the refractive index of the waveguide array, a constant phase difference is added between the adjacent channel waveguides, and the condensing position in the output slab waveguide can be changed. The change in the condensing position can be controlled by the refractive index modulation amount, and optical scanning is possible. Further, by coupling a plurality of output channel waveguides to the output slab waveguide, the deflected light can be guided to a desired output channel waveguide, and the deflection element functions as a 1 × N switch. By increasing the length of the arrayed waveguide having the refractive index modulation region, a large deflection angle can be obtained even when the refractive index modulation amount is small. By configuring the channel waveguide with a high refractive index difference waveguide, it is possible to reduce the size of the element, and by providing a structure that increases the core width tolerance of the linear array portion, manufacturing is also easy. By configuring the deflection element of the present invention with a waveguide material capable of high-speed refractive index modulation such as an electro-optic material, a 1 × N switch capable of high-speed path switching and an optical deflector capable of high-speed scanning are provided. Can be provided.

  The present invention relates to an optical deflector used in a scanner unit such as an optical switching element, a laser printer, a laser scanning microscope, and a barcode reader used for optical fiber communication, and more particularly to an optical deflector using a planar optical waveguide circuit, It can be used for a channel optical switching element.

Conceptual diagram of optical deflection element (1) Conceptual diagram of optical deflection element (2) Explanatory drawing of output port part of optical deflection element Relationship graph between refractive index modulation amount and output intensity at each port (m = 10) Change of output field by refractive index modulation Relationship graph between refractive index modulation amount and output intensity at each port (m = 20) Graph of relationship between refractive index modulation amount and output intensity at each port (second waveguide array length = 100 μm) Change of output field by refractive index modulation Relationship graph between refractive index modulation amount and output intensity at each port (second waveguide array length = 200 μm) Changes in the equivalent refractive index of a waveguide with changes in core width Change in equivalent refractive index of waveguide per 1 mm change in core width Conceptual diagram of the core width inclined enlarged portion of the waveguide array Example of a conceptual diagram of an inclined clad in the core width expansion part of a waveguide array Graph showing the change in light collection performance due to the tolerance expansion structure Graph showing the change in light output intensity due to refractive index modulation in the tolerance expansion structure Cross section of channel waveguide Graph showing equivalent refractive index change for core width change in low refractive index waveguide Conceptual diagram of photonic crystal waveguide array Conceptual diagram of waveguide array with electroabsorption layer Connection between input slab waveguide and waveguide array or connection between output slab waveguide and waveguide array Connection between input channel waveguide and input slab waveguide Connection between output channel waveguide and output slab waveguide Conceptual diagram of change in optical path length of output slab waveguide Condensing light to the output port Graph showing change in optical output intensity due to movement of output slab waveguide Structure diagram of waveguide array Optical crystal thin film fabrication process (1) Optical crystal thin film fabrication process (2) SEM photograph of LN thin film formed by ion slicing method Patterning to optical crystal thin film (1) Patterning to optical crystal thin film (2) SEM photo with fine pattern formed on LN thin film Optical deflection module (1) Optical deflection module (2) Optical switch module (1) Optical switch module (2)

Explanation of symbols

0001: input channel waveguide 0002: input slab waveguide 0003: waveguide array 0004: output slab waveguide 0005: first waveguide array 0006: second waveguide array 0007: refractive index modulation region 0008: collimating element 0009 : Output channel waveguide 0101: waveguide array 0102: first waveguide array 0103: second waveguide array 0104: core width changing portion 0105: core of channel waveguide 0106: clad of waveguide array 0201: waveguide Array 0202: First waveguide array 0203: Second waveguide array 0204: Core width changing part 0205: Channel waveguide core 0206: Cladding of the first waveguide array 0207: Cladding of the second waveguide array 0211: Channel waveguide core 0212: Channel waveguide core 1 clad 0213: channel waveguide second clad 0214: channel waveguide third clad 0401: waveguide array 0402: first waveguide array 0403: second waveguide array 0404: photonic crystal array 0405: Photonic crystal line defect waveguide 0501: Waveguide array 0502: First waveguide array 0503: Second waveguide array 0504: Core width changing portion 0505: Core 0506 of channel waveguide: Light absorption layer 0701 Input slab waveguide or output slab waveguide 0702: Waveguide array 0703: Core width change region 0711: Input channel waveguide 0712: Input slab waveguide 0713: Core width change region 0721: Output slab waveguide 0722: Output channel waveguide 0723: Core width change region 0801: Output slab waveguide 0802: Output channel waveguide 0803: the amount of change in the optical path of the output slab end faces 0901: substrate 0902: core channel waveguides 0903: the cladding of the channel waveguide 0904: electrode

Claims (17)

An optical deflecting element in which an input slab waveguide having at least one input port formed on a substrate, a waveguide array composed of a plurality of channel waveguides, and an output slab waveguide are sequentially connected,
The waveguide array comprises a first waveguide array and a second waveguide array in which the optical lengths of the channel waveguides are equal to each other,
The second waveguide array comprises refractive index modulation regions of different lengths in each channel waveguide;
An optical deflecting element, wherein a core width of a channel waveguide constituting the second waveguide array is wider than a core width of a channel waveguide constituting the first waveguide array. An optical deflecting element in which an input slab waveguide having at least one input port formed on a substrate, a waveguide array composed of a plurality of channel waveguides, and an output slab waveguide are sequentially connected,
The waveguide array comprises a first waveguide array and a second waveguide array in which the optical lengths of the channel waveguides are equal to each other,
The second waveguide array comprises refractive index modulation regions of different lengths in each channel waveguide;
The relative refractive index difference between the core and the clad of the channel waveguide constituting the second waveguide array is smaller than the relative refractive index difference between the core and the clad of the channel waveguide constituting the first waveguide array. An optical deflection element characterized by the above. In the connection between the first waveguide array and the second waveguide array,
2. The optical deflection element according to claim 1, wherein the core width of the channel waveguide constituting the second waveguide array is increased in a tilted manner. In the vicinity of the connection portion between the first waveguide array and the second waveguide array,
The optical deflection element according to claim 2, wherein a clad constituting the second waveguide array is formed in an inclined shape. An optical deflecting element in which an input slab waveguide having at least one input port formed on a substrate, a waveguide array composed of a plurality of channel waveguides, and an output slab waveguide are sequentially connected,
The waveguide array comprises a first waveguide array and a second waveguide array in which the optical lengths of the channel waveguides are equal to each other,
The optical deflection element, wherein the second waveguide array has a photonic crystal arrangement.   6. The optical deflection element according to claim 1, wherein the waveguide array has an electroabsorption layer between channel waveguides.   The optical deflection element according to any one of claims 1 to 6, wherein a plurality of output channel waveguides are connected to an optical path exit of the output slab waveguide.   The optical deflection element according to any one of claims 1 to 7, wherein a mirror that reflects light in the optical path is provided in the channel waveguide, and the input slab waveguide also serves as the output slab waveguide.   In the connection portion between the input slab waveguide and the input channel waveguide, the connection portion between the input slab waveguide and the waveguide array, and / or the input channel waveguide connected to the input port of the input slab waveguide. 9. The optical deflecting element according to claim 1, wherein the core width of the channel waveguide and / or the input channel waveguide is widened and connected.   The connection section between the output slab waveguide and the output channel waveguide and / or the waveguide array is connected with the core width of the output channel waveguide and / or channel waveguide being widened. Item 10. The light deflection element according to Item 7 or 9.   The optical deflection element according to claim 1, wherein the input slab waveguide and the output slab waveguide have different optical path lengths.   In the channel waveguide constituting the second waveguide array, the core of the refractive index modulation region is formed of an optical crystal material having an electrooptic effect, and at least a pair of electrodes are arranged so as to face each other with the core interposed therebetween, The optical deflection element according to claim 1, wherein the clad is formed between the core and the electrode. An optical deflection module configured by sequentially coupling a light source, an optical coupling unit, and an optical deflection element,
The optical deflection module according to claim 1, wherein the optical deflection element is an optical deflection element according to claim 1.   The optical deflection module according to claim 13, further comprising a temperature control device that adjusts a temperature of the optical deflection element. An optical switch module including an optical amplifier before or after the optical deflection element,
The optical deflection module according to claim 1, wherein the optical deflection element is the optical deflection element according to claim 1. An optical deflection method for deflecting light by sequentially passing an input slab waveguide having at least one input port formed on a substrate, a waveguide array composed of a plurality of channel waveguides, and an output slab waveguide. There,
The waveguide array is provided with a first waveguide array and a second waveguide array in which the optical path lengths of the channel waveguides are equal to each other,
The second waveguide array performs refractive index modulation with different lengths in each channel waveguide;
An optical deflection method characterized in that a core width of a channel waveguide constituting the second waveguide array is made wider than a core width of a channel waveguide constituting the first waveguide array. An optical deflection method for deflecting light by sequentially passing an input slab waveguide having at least one input port formed on a substrate, a waveguide array composed of a plurality of channel waveguides, and an output slab waveguide. There,
The waveguide array is provided with a first waveguide array and a second waveguide array in which the optical path lengths of the channel waveguides are equal to each other,
The second waveguide array performs refractive index modulation with different lengths in each channel waveguide;
The relative refractive index difference between the core and the clad of the channel waveguide constituting the second waveguide array is made smaller than the relative refractive index difference between the core and the clad of the channel waveguide constituting the first waveguide array. An optical deflection method.