JP4124189B2 - Planar optical circuit and optical circuit - Google Patents

Description

  The present invention relates to a planar optical circuit capable of reducing coupling loss that occurs when connecting to an optical fiber or connecting each other.

  Planar optical circuits are expected to be used more and more as major components that carry important functions such as optical signal routing in future ultra-high-speed and large-capacity optical communication systems. In particular, with the increase in communication capacity, construction of a larger-scale optical communication system is required. In order to realize such a large-scale optical communication system, it is essential to reduce the size of the planar optical circuit so that a large number of planar optical circuits can be connected.

  The schematic structure of this conventional planar optical circuit is shown in FIG. As shown in FIG. 30, an input waveguide 112 is formed on the substrate 111. The input waveguide 112 is connected to a slab waveguide 113 formed on the substrate 111. One end side of a plurality of arrayed waveguides 114 formed on the substrate 111 is connected to the slab waveguide 113. A half-wave plate 115 that eliminates polarization dependence is provided in the middle of the arrayed waveguides 114. The other end side of these arrayed waveguides 114 is connected to a slab waveguide 116 formed on the substrate 111. A plurality of output waveguides 117 formed on the substrate 111 are connected to the slab waveguide 116, respectively.

  In such a planar optical circuit 110, when a multiplexed optical signal having a plurality of different wavelengths enters from the input waveguide 112, the light enters the arrayed waveguide 114 via the slab waveguide 113, The polarization dependency is eliminated by the half-wave plate 115, and the wavelength is demultiplexed for each wavelength in the slab waveguide 116 due to the difference in the optical path length of each arrayed waveguide 114, and output from each output waveguide 117. The

In such a planar optical circuit 110, in order to achieve miniaturization, as shown by the following formula (1), the refraction of the _core with respect to the difference between the refractive index n _{core of the core} and the refractive index n _{clad of the} cladding. It is extremely effective to apply a waveguide (hereinafter referred to as “SHΔ waveguide”) in which the relative refractive index difference Δ, which is the ratio of the refractive index n _core , exhibits a high value of 1% or more. This is because the higher the relative refractive index difference Δ, the stronger the light confinement, and it can cope with a waveguide that bends rapidly.

Δ = (n _core −n _clad ) / n _core (1)
Related documents include the following documents.
Japanese Patent Laid-Open No. 11-52168 JP-A-8-171020 JP-A-5-60928 JP 63-280202 A Takahiro Nakamura et.al., 1998 IEICE Electronics Society Conference Proceedings 1, September 7, 1998, luminescence, p. 310, c-4-12

  However, the SHΔ waveguide has a difficulty in that the coupling loss (also referred to as connection loss) is very large. Here, the coupling loss will be described. As shown in FIG. 31, when an optical signal propagating through the core 120a of the optical fiber 120 is input to the core 112a of the input waveguide 112 of the planar optical circuit 110, an optical signal called coupling loss is attenuated. Coupling loss occurs when different types of waveguides are connected, and is caused by different field distributions between different types of waveguides, and increases as the number of connection points increases. For example, at a wavelength of 1.5 μm used in an optical communication system, a large coupling loss of about 3.5 dB occurs between a single-mode optical fiber having a core diameter of 9 μm and a SHΔ waveguide having a 5 μm square.

  In the planar optical circuit 110 used in the optical communication system, there is a contradictory relationship between downsizing and reduction in coupling loss. That is, when Δ is increased, the circuit is reduced in size, but the coupling loss increases. Therefore, there was a limit to the construction of a practical system. In other words, unless the coupling loss of the SHΔ waveguide is reduced, it will be difficult to cope with the desired increase in the capacity of the communication path, and the function and scale of the optical communication system will be restricted. It is.

  As one method for reducing the coupling loss between the SHΔ waveguide and the optical fiber, it is known to use a spot size converter having a narrow core width in the vicinity of the substrate end face as shown in FIG. 4).

  That is, if the core width is narrowed to some extent, there is a region where the spot size of light becomes wider, and it becomes possible to reduce the coupling loss of both by matching the spread spot size with the spot size of the optical fiber. .

  However, narrow taper spot size conversion with a reduced core width is known to cause significant changes in coupling loss due to slight fabrication errors such as core width and Δ, and so has been applied to practical circuits. There wasn't.

  An object of the present invention is to provide a planar optical circuit capable of suppressing coupling loss while reducing the size. In particular, it is an object of the present invention to provide a planar optical circuit and an optical circuit using a narrow taper spot size converter having a large tolerance for manufacturing errors.

In order to solve the above-described problems, a planar optical circuit according to the present invention includes a planar optical circuit in which an input waveguide and an output waveguide are formed on a substrate, and a core having a high refractive index is surrounded by a cladding having a low refractive index. And the input waveguide side of the input waveguide is formed so that the core width of the input end side of the input waveguide or the output end side of the core of the output waveguide gradually decreases toward the substrate end surface. edge side and the output input to with any different edge side of the edge side of the substrate positioned in the output end of the waveguide of the substrate is formed to the position of the end, the core width of the input end side A first monitor waveguide in which a core formed so as to become gradually narrower as it approaches the substrate end surface is surrounded by a clad , wherein the input end of the first monitor waveguide is cut, and the input end and the optical fiber after cutting Result of By measuring the loss, the first monitor waveguide for obtaining an optimum core width used as the core width of the input end or the output end of the output waveguide of said input waveguide, said input waveguide An input end located on the edge side of the substrate located on the input end side and having the same shape as the input end side, and an edge side of the substrate located on the output end side of the output waveguide A second monitor waveguide having an output end positioned at the same shape as the output end for estimating a coupling loss at the input end of the input waveguide and at the output end of the output waveguide The second monitor waveguide is formed on the substrate.
The planar optical circuit according to the present invention is a planar optical circuit in which an input waveguide and an output waveguide each having a high refractive index core surrounded by a low refractive index cladding are formed on a substrate. The core width on the input end side of the core or on the output end side of the core of the output waveguide is formed so as to gradually narrow toward the end face of the substrate, and the edge of the substrate on which the input end of the input waveguide is located An input end is formed on the edge side different from both the end side and the edge side of the substrate where the output end of the output waveguide is located, and the core width on the input end side gradually becomes narrower as it approaches the substrate end surface. the formed core as a first monitor waveguide surrounded by a cladding, to cut the input of the first monitor waveguide, measuring the coupling loss between the input end and the optical fiber after cutting Ri, a first monitor waveguide for obtaining an optimum core width used as the core width of the input end or the output end of the output waveguide of said input waveguide, said input end of said input waveguide An input end located on the edge side of the substrate located and having the same shape as the input end side, and an output located on the edge side of the substrate located on the output end side of the output waveguide A waveguide having an output end having the same shape as the end, an edge side of the substrate positioned on the input end side of the input waveguide, and the substrate positioned on the output end side of the output waveguide in the have a both ends edge side, a second monitor waveguide and a linear waveguide, the core width is constant, the input end and the output end of the output waveguide of said input waveguide second monitor waveguide and, said group for estimating the coupling loss It may be configured as being formed on.

Further, the present invention comprises a substrate, a waveguide in which a core having a high refractive index is embedded in a cladding having a low refractive index, and a spot size converter disposed as a part of the waveguide near the end face of the substrate. an optical circuit that monitor waveguide having a spot size converter is formed in a plurality on the substrate in the substrate end face side, further, the spot size converter in said plurality of monitor waveguides, those the monitor waveguide The plurality of monitor waveguides are arranged to be shifted from each other by a predetermined distance in the longitudinal direction of the plurality of monitor waveguides after the spot size converters of the plurality of monitor waveguides are cut at a predetermined cutting position. By measuring the coupling loss between the spot size converter of the waveguide and the optical fiber, it is used as the core width on the substrate end face side of the spot size converter of the waveguide In order to obtain an appropriate core width, the spot size converter of the waveguide has a distance between the substrate end face and the spot size converter larger than a distance between the substrate end face and the predetermined cutting position. Each of the spot size converter of the waveguide and the spot size converter of the plurality of monitor waveguides is formed so that the core width is minimized at the end face of the substrate, and a core width fine-tuning part of the substrate end face side and the core width converting part which follows, the mean taper angle theta ₁ of the core width fine-tuning part, the average taper angle of greater said core width converting part than 0 ° theta You may comprise so that it may be smaller than ₂ .

The present invention also includes a substrate, a waveguide in which a core having a high refractive index is embedded in a clad having a low refractive index, and a spot size converter disposed as a part of the waveguide near the end face of the substrate. an optical circuit, said any of the edge side of the substrate positioned in the edge side and the output end of the waveguide of the substrate located in the input end of the waveguide at different edge side, the core width substrate A first monitor waveguide having a spot size converter formed so as to become gradually narrower as it approaches the end surface , wherein the input end of the spot size converter of the first monitor waveguide is cut, and the input end after cutting and by measuring the coupling loss between the optical fiber, a first monitor waveguide for obtaining an optimum core width used as the core width of the substrate end face side of the spot size converter of the waveguide, the waveguide A second monitor waveguide comprising a spot size converter at both edge side of the substrate located in the input and output ends, the coupling losses at the input end and the output end of the waveguide of the waveguide second and monitor waveguide, an optical circuit is formed on the substrate, a spot size converter of the waveguide, the spot size converter of the first monitor waveguide, and the second monitor to estimate Each of the spot size converters at both ends of the waveguide is formed such that the core width is minimized at the substrate end surface, and has a core width fine adjustment unit on the substrate end surface side and a core width conversion unit following the core width conversion unit. You may comprise so that the average taper angle (theta) ₁ of a width fine adjustment part may be larger than 0 degree and smaller than the average taper angle (theta) ₂ of this core width conversion part .
The present invention also includes a substrate, a waveguide in which a core having a high refractive index is embedded in a clad having a low refractive index, and a spot size converter disposed as a part of the waveguide near the end face of the substrate. an optical circuit, said any of the edge side of the substrate positioned in the edge side and the output end of the waveguide of the substrate located in the input end of the waveguide at different edge side, the core width substrate A first monitor waveguide having a spot size converter formed so as to become gradually narrower as it approaches the end surface , wherein the input end of the spot size converter of the first monitor waveguide is cut, and the input end after cutting and by measuring the coupling loss between the optical fiber, a first monitor waveguide for obtaining an optimum core width used as the core width of the substrate end face side of the spot size converter of the waveguide, the waveguide And with a waveguide spot size converter at both edge side of the substrate located in the input and output ends comprises a spot size converter, possess both ends on both edges end side of the substrate is the core width is constant A second monitor waveguide comprising a straight waveguide, the second monitor waveguide for estimating a coupling loss at the input end of the waveguide and the output end of the waveguide formed on the substrate an optical circuit which is, the waveguide of the spot size converter, each of the first monitor waveguide spot size converter, and a spot size converter of both ends of the spot size converter with waveguides, the core width is formed to minimize the substrate end face, and a core width fine-tuning part of the substrate end face side and the core width converting part which follows, the mean taper angle theta ₁ of the core width fine-tuning part, ° may be configured to be less than ₂ increased the average taper angle of said core width converting part θ than.

  According to the optical circuit of the present invention, it is possible to suppress the coupling loss of the SHΔ waveguide while reducing the size. In particular, by using the monitor waveguide, it is possible to eliminate a cutting error that occurs when the substrate is cut, and to always realize a low coupling loss. For this reason, many optical circuits, in particular, a low-loss and highly integrated optical circuit can be efficiently applied to an optical communication system, so that a large-capacity optical communication system with increasing demand can be constructed.

As explained in the prior art, a spot size converter using a narrow taper has a very narrow fabrication tolerance of the core width that minimizes the coupling loss with the optical fiber. That is, the optimum value of the core width varies depending on the manufacturing conditions such as the relative refractive index difference Δ between the core and the clad and the core thickness. In the present invention, in order to obtain an optimum core width by finely adjusting the core width, the core width in the input / output waveguide is made narrower as it approaches the substrate end face. Provided is a planar optical circuit that realizes an optimum core width with good reproducibility by adjusting the end face position. Hereinafter, although each embodiment in the present invention is described, the present invention is not limited to these embodiments.
[First embodiment]
<First example>
A first example of the first embodiment of the planar optical circuit according to the present invention will be described with reference to FIGS. 1 is a schematic configuration diagram of a planar optical circuit, FIG. 2 is an enlarged view of extraction of cores of an input waveguide and an output waveguide of the planar optical circuit of FIG. 1, and FIG. 3 is a planar optical circuit of FIG. FIG. 4 is a result of calculation of the relationship between the core width and the coupling loss at the substrate end face, and FIG. 5 is an explanatory diagram of the cut-out error.

  As shown in FIG. 1, an input waveguide 12 made of silica glass in which a core having a high refractive index is surrounded by a clad having a low refractive index is formed on a silicon substrate 11. As shown in FIG. 2, the taper is obtained by gradually changing the core width so that the core 12a has a tapered shape with a smaller core width on the input end side located on the edge side of the substrate 11 toward the input end side. It has a part 12aa.

  In each embodiment, the portion of the waveguide whose width is narrow from the vicinity of the substrate end surface to the substrate end surface is referred to as a spot size converter. For example, the tapered waveguide portion in FIG. 2 is a spot size converter.

  As shown in FIG. 1, the input waveguide 12 is connected to a quartz glass slab waveguide 13 formed on a substrate 11. One end side of an arrayed waveguide 14 made of quartz glass formed on the substrate 11 is connected to the slab waveguide 13. A half-wave plate 15 that eliminates polarization dependence is provided in the middle of the arrayed waveguide 14. The other end side of the arrayed waveguide 14 is connected to a quartz glass slab waveguide 16 formed on the substrate 11.

  A quartz glass output waveguide 17 formed on the substrate 11 is connected to the slab waveguide 16, and the output waveguide 17 has a core 17 a formed on the substrate 11 as shown in FIG. 2. A taper portion 17aa in which the core width is gradually changed so as to have a taper shape with a smaller core width toward the output side is provided on the output end side located on the edge side.

  The waveguides 12 and 17 of the planar optical circuit 10 are formed as follows.

First, a lower clad glass soot mainly composed of SiO ₂ is deposited on a silicon substrate 11 by a flame deposition method, and then a core glass soot obtained by adding GeO ₂ to SiO ₂ is flame-flamed on the lower clad glass soot. After being deposited by the deposition method, the glass soot is made transparent by high-temperature heat treatment (1000 ° C. or higher) to form the lower clad glass 12ba and the core glass 12a on the substrate 11 (FIG. 3A). At this time, the thickness at the time of deposition by the flame deposition method is adjusted so that the lower clad glass 12ba and the core glass 12a have target thicknesses.

  Next, an etching mask 100 is formed on the core glass 12a by a photolithographic technique so that the edge of the substrate 11 has a taper shape with a smaller width toward the edge (FIG. 3B), and then the reaction. The core glass 12a is patterned by reactive ion etching (FIG. 3C), and then the etching mask 100 is removed (FIG. 3D).

Finally, a SiO ₂ -based upper clad glass 12bb having a glass transition temperature lowered by adding a dopant such as B ₂ O ₃ or P ₂ O ₅ is deposited on the lower clad glass 12ba and the core glass 12a by a flame deposition method. The waveguides 12 and 17 can be formed on the substrate 11 by depositing and allowing the upper cladding glass 12bb to enter even between the adjacent core glasses 12a (FIG. 3E).

  In such a planar optical circuit 10, when a multiplexed optical signal having a plurality of different wavelengths enters from the input waveguide 12, the light enters the arrayed waveguide 14 via the slab waveguide 13, The half-wave plate 15 eliminates the polarization dependency, and the light is demultiplexed for each wavelength in the slab waveguide 16 due to the difference in the path length of each arrayed waveguide 14 and output from each output waveguide 17. The

  Here, FIG. 4 shows the result of calculating the relationship between the core width and the coupling loss at the end face of the substrate for a single mode optical fiber having a core diameter of 9 μm. However, the core thickness was 5 μm and Δ was 1.5%. As can be seen from the figure, when the core width is gradually reduced from 5 μm, the spot size in the waveguide decreases accordingly, and the coupling loss with the fiber once increases. However, when the core width is 4 μm or less, light confinement to the core is weakened and the spot size is widened, so that the coupling loss with the fiber is low. When the core width is further reduced, the coupling loss is minimized in the vicinity of 1.2 μm. However, the coupling loss increases rapidly thereafter. This is because the spot size increases rapidly because the core width is too narrow.

  As shown in FIG. 4, when the core width is 1.2 μm, the coupling loss is about 0.5 dB, and the total of the coupling loss on the both sides of the input and output can be suppressed to about 1.0 dB. On the other hand, since the waveguide is designed to propagate single mode light, the width and thickness of the core are determined by the material and the like.

  Therefore, in the planar optical circuit 10 of the first example of the present embodiment, the tapered portions 12aa and 17aa as described above are formed on the edge side of the substrate 11 of the cores 12a and 17a of the waveguides 12 and 17, respectively. Thus, the coupling loss can be reduced while satisfying the propagation condition of the single mode light.

  As described above, in the vicinity of the core width of 1.2 μm, the coupling loss is greatly increased even if the core width is slightly shifted. Therefore, due to an error at the time of cutting out the planar optical circuit 10, the input waveguide 12 and There is a possibility that the core width of the output waveguide 17 is shifted and the coupling loss is increased.

  Here, as shown in FIG. 5, assuming that the taper angle of the taper portions 12aa and 17aa of the cores 12a and 17a is θ and the shift amount of the cutout position is Δx, the shift amount Δa on one side of the core width is expressed by the following formula ( 2).

Δa = Δx · tan θ (2)
For example, when [theta] is 1.5 [deg.] And [Delta] x is 5 [mu] m (normal size of cut-out error), [Delta] a is 0.13 [mu] m, so the deviation 2 [Delta] a on both sides is 0.26 [mu] m. Therefore, the core width is 1.2 ± 0.26 μm, and as can be seen from FIG. 4, the coupling loss is about 0.7 dB, which is only about 0.2 dB higher than when Δx is 0. . Accordingly, even if an error occurs when the planar optical circuit 10 is cut out, the coupling loss is not greatly affected, and the planar optical circuit 10 with reduced coupling loss can be easily manufactured on a regular basis.

  The taper angle θ of the taper portions 12aa and 17aa of the cores 12a and 17a of the waveguides 12 and 17 is preferably greater than 0 ° and 5 ° or less. This is because if the taper angle θ is 0 °, the effect of the present invention cannot be obtained, and if the taper angle θ exceeds 5 °, the core width deviation amount Δa due to the above-described cutting error becomes too large. This is because the coupling loss becomes too large.

In the planar optical circuit 10 described above, the quartz-based waveguides 11 to 17 are formed on the silicon substrate 11. However, the present invention is not limited to this, and materials such as polyimide, silicon, semiconductor, and LiNbO ₃ are used. The waveguides 11 to 17 may be formed on the substrate 11 made of various materials.

<Second example>
Instead of providing a simple taper as shown in FIG. 2, for example, as shown in FIG. 6, the input / output end side located on the edge side of the substrate 11 has a smaller core width toward the input / output end side. The cores 32a and 37a formed with a plurality of step portions 32aa and 37aa that gradually change the core width may be applied.

  At this time, if the number of the step portions 32aa and 37aa is increased so as to make the difference Δb between the adjacent step portions 32aa and 37aa as small as possible, the shape of the cores 32a and 37aa is approximated to the taper shape in the case described above. Thus, since the effect of reducing the coupling loss can be improved, it is preferable to increase the number of the step portions 32aa and 37aa as much as possible in consideration of the cut-out error Δx.

  In other words, since the core width is made smaller toward the input / output end side by the plurality of step portions 32aa and 37aa, even if a low coupling loss is not obtained at the cut-out position, it is used by cutting again at another position. Even if the exact value of the optimal core width is not known and only the estimated value can be obtained by analysis, the optimal core width can be searched by changing the cutout position. It can be done.

  Therefore, if the cores 32a and 37a are applied, the target core width can be easily obtained even if the cutting error Δx occurs.

  Here, the difference (step size) Δb between the adjacent step portions 32aa and 37aa of the cores 32a and 37a is preferably greater than 0 μm and not more than 5 μm. This is because if Δb is 0 μm, the effect of the present invention cannot be obtained, and if Δb exceeds 5 μm, the propagation condition for single mode light cannot be satisfied.

  In the case shown in FIG. 6, the difference between the adjacent step portions 32aa and 37aa (the size of the step portion) Δb and the length of the step portions 32aa and 37aa are all constant, but the adjacent step portions 32aa and 37aa. (Step size) Δb and the lengths of the step portions 32aa and 37aa can be set to be different.

<Third example>
Further, for example, as shown in FIG. 7, the core width is gradually reduced so that the input / output end side on the input end side or the output end side positioned on the edge side of the substrate 11 has a tapered shape with a smaller core width. It is also possible to apply cores 42a, 47a in which a plurality of taper portions 42aa, 47aa changed and straight portions 42ab, 47ab formed along the axial direction so as to have a constant core width are alternately formed. It is.

  When such cores 42a and 47a are applied, the coupling loss can be reduced while satisfying the propagation condition of the single mode light, as well as the cutting error by cutting out at the straight portions 42ab and 47ab. Can be buffered, and the desired core width can be easily obtained.

  Here, the length s of the straight portions 42ab and 47ab of the cores 42a and 47a is preferably 1 μm or more. This is because, if s is less than 1 μm, it becomes difficult to buffer the clipping error.

  In the case shown in FIG. 7, the lengths of the tapered portions 42aa and 47aa and the straight portions 42ab and 47ab are all constant, but these lengths may be set differently.

<Fourth example>
For example, as shown in FIG. 8, markers La1 and La2 indicating the cutting positions of the waveguides 12 and 17 are provided in the vicinity of the input / output end side of the waveguides 12 and 17 on the surface of the planar optical circuit 10. If the markers Lb1 and Lb2 indicating the positions of changing points of the core widths of the cores 12a, 17a, 42a, and 47a as described above are formed at the time of manufacture, the cutting accuracy can be easily improved. .

[Second embodiment]
<First example>
A first example of the second embodiment of the planar optical circuit according to the present invention will be described with reference to FIGS. FIG. 9 is a schematic configuration diagram of the planar optical circuit, and FIG. 10 is an enlarged view of the cores of the input waveguide and the output waveguide of the planar optical circuit of FIG. However, the same members as those in the first embodiment described above are denoted by the same reference numerals as those used in the description of the first embodiment, and the description thereof is omitted.

  In this embodiment, the present invention is applied to an arrayed waveguide grating type optical wavelength demultiplexing circuit (AWG) that multiplexes and demultiplexes optical signals of different wavelengths in an optical wavelength multiplexing communication system as a planar optical circuit. This is the case. Note that AWG is an example of a waveguide type optical circuit.

  As shown in FIG. 9, a plurality of input waveguides 12 are formed on the substrate 11, and as shown in FIG. 10, the input waveguides 12 have a core width at the input end face of the tapered portion 12aa of the core 12a. Are different for each input waveguide 12.

  In addition, a plurality of output waveguides 17 are formed on the substrate 11, and as shown in FIG. 10, each of the output waveguides 17 has a core width at the output end face of the tapered portion 17aa of the core 17a. Each one is different.

  The waveguides 12 and 17 of the AWG 50 can be manufactured basically in the same manner as the planar optical circuit 10 of the first embodiment described above. For example, as shown in FIG. Then, reactive ion etching is performed by forming a tapered etching mask having a smaller taper angle θ on the edge side of the substrate 11 and having a different taper angle θ, or as shown in FIG. By forming a taper-shaped etching mask 100 having the same taper angle θ which is smaller on the edge side and with different taper sections, reactive ion etching is performed, and the substrate 11 is cut out linearly. The cores 12a and 17a having the tapered portions 12aa and 17aa having different core widths on the input end face and the output end face can be easily formed.

  In the AWG 50 having such a structure, when a multiplexed optical signal having a plurality of different wavelengths enters from an arbitrary input waveguide 12, the light enters the arrayed waveguide 14 via the slab waveguide 13. , The polarization dependency is eliminated by the half-wave plate 15, and the light is demultiplexed for each wavelength in the slab waveguide 16 due to the difference in the path length of each arrayed waveguide 14 and output from each output waveguide 17. Is done.

  According to such an AWG 50, since the core widths of the input / output end faces of the tapered portions 12a and 17a of the cores 12a and 17a are different from each other, an increase in coupling loss due to a manufacturing error can be eliminated. The reason for this will be described below.

  The multiple wavelength signal light incident from an input port (not shown) connected to the input waveguide 12 is demultiplexed for each wavelength and connected to each output waveguide 17 as described above. Is output from. At this time, since the core width of each output port varies due to the cutting error, the coupling loss differs for each output port.

  Therefore, by providing a plurality of input waveguides 12 with different core widths at the input end face, and connecting the input ports to these input waveguides 12, an input port that can reduce the coupling loss is selected for each output port. It is something that can be done. In addition, since the sum of the coupling loss between the input port and the output port can be made constant, the amount of reduction in the coupling loss of all the ports can be made uniform.

<Second example>
Further, for example, as described in the second example of the first embodiment described above, the core width is smaller toward the input / output end side on the input end side or output end side located on the edge end side of the substrate 11. It is also possible to apply the cores 32a and 37a formed with a plurality of step portions 32aa and 37aa that gradually change the core width so that

<Third example>
Further, as described in the third example of the first embodiment, the taper having a smaller core width toward the input / output end side on the input end side or output end side located on the edge end side of the substrate 11. A plurality of taper portions 42aa and 47aa whose core widths are gradually changed so as to have a shape and straight portions 42ab and 47ab formed along the axial direction so as to have a constant core width are formed alternately. It is also possible to apply the cores 42a and 47a.

<Fourth example>
Further, in the first example of the present embodiment, among the input ports connected to the plurality of input waveguides 12, the input port with the smallest coupling loss is selected and the signal light is incident. When it is necessary to use all the input waveguides 12, for example, as shown in FIG. 13, an input port i having a plurality of input waveguides 12 as described in the first embodiment described above. Is connected to each input waveguide 12 of the AWG 50, and an output port j having a plurality of output waveguides 17 as described in the first embodiment is connected to each output waveguide 17 of the AWG 50, If an optimal input waveguide 12 is selected at each input port i and signal light is incident on each input waveguide 12 of the AWG 50, the output waveguide 17 of the output port j of each output waveguide 17 of the AWG 50 is selected. It is possible to output a signal light, irrespective of the manufacturing error, all ports i, it is possible to reduce the coupling loss in j.

[Third embodiment]
A third embodiment of the planar optical circuit according to the present invention will be described with reference to FIG. FIG. 14 is a schematic configuration diagram of a planar optical circuit. However, members similar to those in the first and second embodiments described above are omitted by using the same reference numerals as those used in the description of the first and second embodiments described above. To do.

  As shown in FIG. 14, a plurality of first monitor waveguides 68 and second monitor waveguides 69 are formed on the substrate 11. The first monitor waveguide 68 has an input end 68a located on the edge side different from the edge side of the substrate 11 where the input end of the input waveguide 12 and the output end of the output waveguide 17 are located. The 68a side is tapered so that the core width becomes smaller toward the input end 68a, and the core width of the input end 68a is different for each waveguide 68, while the output end 68b is the output guide. It is located on the edge side of the substrate 11 where the output end of the waveguide 17 is located.

  Further, the input end 69 a and the output end 69 b of the second monitor waveguide 69 have the same structure as the input waveguide 12 and the output waveguide 17. That is, the input end 69a is located on the edge side of the substrate 11 where the input end of the input waveguide 12 is located, and the input end 69a side is tapered so that the core width becomes smaller toward the input end 69a side. In addition, the core width of the input end 69a is different for each waveguide 69, while the output end 69b is located on the edge side of the substrate 11 where the output end of the output waveguide 17 is located. The end 69b side is tapered so that the core width is smaller toward the output end 69b, and the core width of the output end 69b is different for each waveguide 69.

  In the AWG 60 in which the monitor waveguides 68 and 69 are formed, the input end 68a of the monitor waveguide 68 of the substrate 11 is cut and the coupling loss is measured repeatedly, and the coupling loss depends on the core width. After determining the optimum core width by examining the characteristics, the edges of the input waveguide 12 side and the output waveguide 17 side of the substrate 11 are cut so as to be the core width, whereby the input end face of the input waveguide 12 is obtained. The core width of the output end face of the output waveguide 17 can be set to an optimum size.

  If the coupling loss of the monitor waveguide 69 is measured, the cutting error of the input waveguide 12 and the output waveguide 17 can be examined.

  According to such an AWG 60 of the present embodiment, it is possible to monitor a cutting error occurring for each individual substrate 11 and to obtain an optimum core width for each individual substrate 11.

  In the second and third embodiments described above, the case where the present invention is applied to the AWG has been described. However, the present invention is not limited to the AWG, and a plurality of input / output waveguides having different core widths are provided to select an optimum port. If used, the coupling loss can be reduced in any planar optical circuit regardless of manufacturing errors.

[Fourth embodiment]
Next, a fourth embodiment will be described with reference to FIGS.

  FIG. 15 is an enlarged view of a spot size converter provided in the input / output waveguide of the planar optical circuit, FIG. 16 is a diagram showing the relationship between the taper angle and excess loss, and FIG. 17 is a spot in this embodiment. It is a figure which shows the example of application of a size converter. In addition, about the member similar to embodiment mentioned above, the description is abbreviate | omitted by using the code | symbol same as the code | symbol used by description of embodiment mentioned above.

  In this embodiment, the spot size converter 21 having the configuration shown in FIG. 15 is provided in the input / output waveguide in the configuration of the planar optical circuit shown in FIG. In addition, since the spot size converter provided in an input waveguide and an output waveguide is the same, below, especially the code | symbol in a figure is not distinguished.

As shown in FIG. 15, the spot size converter 21 of this embodiment includes a core width fine adjustment unit 21 and a subsequent core width conversion unit 22. The main difference between the two is the taper angle. Here, the taper angle is defined as twice the angle represented by θ ₁ or θ ₂ in FIG. The taper angle 2θ ₁ of the core width fine adjustment unit 21 is set smaller than the taper angle 2θ ₂ of the core width conversion unit 22. Moreover, you may provide the marker 23 which shows the cutting position of a waveguide. Details of the marker will be described later.

  Both design parameters will be described below.

In the present invention, the optimum core width is realized with good reproducibility by making the input / output waveguide tapered and adjusting the end face position. However, since the waveguide end face position has only an accuracy of about ± 100 μm for the following reason, in this embodiment, in order to set the accuracy of the finally obtained core width to ± 0.1 μm, the taper angle 2θ _1. Is set to 0.057 °.

  In general, the optical waveguide and the optical fiber are connected and fixed with an adhesive. Since the refractive index of the adhesive is slightly different from the refractive index of glass, light is reflected at the connection surface. In order to prevent this reflected light from being coupled to the optical fiber or the optical waveguide, it is common to polish the end faces of both at an angle of about 5 ° to 10 °. In the present situation where the input / output end faces are formed by this oblique polishing, it is difficult to obtain the input / output end face positions with high accuracy. The input / output end face position accuracy obtained experimentally was ± 100 μm.

For this reason, it is desirable that the taper angle for finely adjusting the optimum core width is a very small value. Therefore, as the above, in the present embodiment, the accuracy of the finally obtained core width in order to ± 0.1 [mu] m, are set to a taper angle 2 [Theta] ₁ and 0.057 °.
In this case, if the core width is simply narrowed from 5 μm to 0.8 μm with a single taper, a taper length of 4.2 mm is required. Considering that the purpose of adopting the SHΔ waveguide is to reduce the size of the planar optical circuit, it is not preferable that the input / output waveguide is as long as 4.2 mm.

  Therefore, in the present invention, the spot size converter 21 includes a core width conversion unit 22 that rapidly reduces the core width within a range where no loss occurs, and a core width fine that is set with a small taper angle in consideration of errors in the end face formation position. It is divided into the adjustment unit 21.

Taper angle 2 [Theta] ₂ of the core width converting part is larger value is preferred to the extent that excess loss does not occur. From the relationship between the taper angle and excess loss shown in FIG. 16, it can be seen that the smaller the taper angle, the lower the loss. In the present embodiment sets the 2 [Theta] ₂ 1 ° and was the core width of the core width converting part so as to reduce from 5μm to a 1.5 [mu] m. At this time, the length of the core width converting portion is about 200 μm.

For the core width fine-tuning part, 0.8 [mu] m in order to ± 0.1 [mu] m the formation accuracy of the previously described optimal core width, the taper angle 2 [Theta] ₁ is set to 0.057 °, the core width is from 1.5μm The taper is reduced to a minimum. The length of the core width fine adjustment portion at this time is 700 μm. As a result, the total length of the core width fine adjustment portion and the core width conversion portion is 900 μm, which is 4.2 mm when formed with a single taper. In comparison, the length is about 1/5.
Incidentally, the taper angle 2 [Theta] ₂ of the core width converting part is larger than 0.08 °, preferably a 10 ° or less. This is because if the taper angle 2θ ₂ is 0.08 ° or less, the length of the core width conversion portion becomes long, and the effect of downsizing cannot be obtained. Further, if the taper angle 2θ ₂ is larger than 10 °, the excessive loss becomes too large. Further preferably taper angle 2 [Theta] ₁ of the core width fine-tuning part is 0 ° greater than 0.08 ° or less. This is because the taper angle 2 [Theta] ₁ is shortens the length of the larger the taper portion than 0.08 °, it becomes impossible to obtain sufficient accuracy due to an error of the polishing.

  The spot size converter 20 shown in FIG. 15 can be applied to the AWG described in the second embodiment as shown in FIG. 17 in addition to the planar optical circuit shown in FIG.

  In this case, the spot size converter of the present invention is provided at the input end of the AWG 70 so that the core width becomes narrower as it approaches the input end, and the spot of the present invention whose core width becomes narrower toward the output end of the AWG. A size converter is provided.

Further, in the same manner as in the second embodiment, the input end face, the spot size converter taper angle theta ₁ or to differ at the output end face, or in advance by shifting the position of each spot size converter to each other The core width of the end face can be made different for each spot size converter by cutting the substrate linearly.

  Furthermore, the cutting position of the end face can be determined using the markers shown in FIGS. As shown in FIG. 17, the marker may be provided only on the upper side of the spot size converter, or may be provided on the upper side and the lower side.

The above optical waveguide can also be formed by a method similar to that of the above-described embodiments. As in the previous embodiments, the waveguide material is not limited to quartz glass, but may be polyimide, silicon, semiconductor, LiNbO ₃ or the like.

  The present embodiment can also be applied to an input / output port as described in FIG. Further, in this embodiment, an arrayed waveguide type diffraction grating (AWG) is taken up as a target planar optical circuit. However, since the point of the present invention is an input / output waveguide including a spot size converter, It does not depend on the type of optical circuit. Therefore, the planar optical circuit is not specified by AWG.

[Fifth embodiment]
A spot size converter 75 according to a fifth embodiment of the present invention is shown in FIG. 18, and enlarged views of the core width fine adjustment unit 76 in FIG. 18 are shown in FIGS.

This embodiment is substantially the same as the fourth embodiment, and the only difference is in the core width fine adjustment unit 76. As shown in FIG. 15, the core width fine-tuning part in the fourth embodiment it was composed of a single tapered waveguide taper angle 2 [Theta] _1. On the other hand, in this embodiment, for example, as shown in FIG. 18, a plurality of tapered waveguides having different taper angles are combined instead of a single tapered waveguide.

For example, as shown in FIG. 19, the core width fine adjustment unit may be configured by alternately connecting tapered waveguides 82 and linear waveguides 81, or as shown in FIG. 20, the linear waveguides are stepped. You may connect and comprise. Since the end face of the waveguide is not a tapered part but a straight waveguide part, the coupling loss with the optical fiber is reduced.
Here, an average value (referred to as an average taper angle) of the taper angle 2θ ₁ is defined by the following equation. However, the core width on the side connected to the fiber is W1, the core width on the side connected to the core width conversion section is W2, and the length of the core width fine adjustment section is L (FIG. 21).

コア幅微調整部の具体的なパラメータは、平均テーパ角度２θ_１を０．０５７°、直線導波路を２００μmとし、それを７回繰り返す構成とすると、コア幅は１．５μmから０．８μmまで低減させることができる。

Specific parameters of the core width fine-tuning part, the average taper angle 2 [Theta] ₁ of 0.057 °, a straight waveguide and 200 [mu] m, when configured to repeat it 7 times, the core width to 0.8μm from 1.5μm Can be reduced.

  Note that the definition of the average taper angle can also be used for the core width converter. In this case, W2 is used as the W1, the core width of the waveguide portion outside the spot size converter is used as the W2, and the length of the core width conversion unit is used as the L.

  Furthermore, the definition of the average taper angle described above can also be used for the spot size converter in the first embodiment having a configuration without the core width conversion unit. In this case, the core width of the waveguide portion outside the spot size converter is used as W2, and the length of the spot size converter is used as L.

  Also in this embodiment, the configurations shown in FIGS. 17 and 13 can be adopted, and this embodiment can also be applied to various optical circuits as in the fourth embodiment.

[Sixth embodiment]
Next, as a sixth embodiment, a planar optical circuit that can form an optimum core width when various conditions change depending on process conditions will be described with reference to FIGS. explain. In addition, in each figure, about the member similar to embodiment mentioned above, the description is abbreviate | omitted by using the code | symbol same as the code | symbol used by description of embodiment mentioned above.

  In this embodiment, as in the third embodiment, a monitor having the spot size converter of the present invention on the same substrate as the planar optical circuit, separately from the planar optical circuit intended for manufacture. A waveguide is arranged.

  FIG. 22 shows an example in which a monitor waveguide and an AWG 90 as shown in FIG. 17 are provided in the same substrate, and FIG. 23 shows a portion of this monitor waveguide. As shown in FIG. 23, in this embodiment, seven monitor waveguides 87 are arranged by shifting the position of the spot size converter having the core width fine adjustment unit 88 and the core width conversion unit 89 by 100 μm from each other. Yes. These monitor waveguides are cut in a lump at the cutting position to become monitor waveguides having different core widths at the end faces. By evaluating the coupling loss between each monitor waveguide and the optical fiber, it is possible to experimentally determine the core width that minimizes the coupling loss with the optical fiber. FIG. 24 shows the shape after cutting.

  In the configuration shown in FIG. 22, the spot size converter provided on the input / output end side of the AWG is inside the cutting position of the monitor waveguide. It can be cut again.

  Also in this embodiment, the same configuration as that of the monitor waveguide shown in FIG. 14 can be adopted. FIG. 25 shows an enlarged view of a portion including the monitor waveguide in this case. The right end of the first monitor waveguide 92 in the figure is a straight line, the optimum core width is clarified by the first monitor waveguide 92, and this circuit is cut. The second monitor circuit 93 can estimate the AWG coupling loss. As shown in the figure, the second monitor waveguide 93 can be constituted by two spot size converters whose ends are the same as the input / output ends of the AWG and two waveguides whose ends are straight. In this case, the difference between the total excess losses of the two becomes the excess loss reduction amount of the spot size converter.

  Note that a planar optical circuit having a monitor waveguide and a planar optical circuit to be manufactured may be provided separately.

[Seventh embodiment]
Next, various embodiments relating to the marker for indicating the end face formation position will be described.
As described above, in order to realize the core width obtained by using the monitor waveguide shown in FIGS. 22 and 25 by the input / output waveguide of the target planar optical circuit, the planar optical circuit is used. For example, the marker shown in FIG. 15 is provided in the fine adjustment portion of the core width in the input / output waveguide so that the end face formation position can be seen even during the polishing process.

  For example, if the monitor waveguide and the marker are appropriately provided, and the fifth monitor waveguide has the smallest loss in the measurement of the monitor waveguide, the end face processing is performed so that the fifth marker becomes the end face. For example, the loss of the target planar optical circuit can be minimized.

  The marker can be attached in various forms, for example, a marker as shown in FIGS.

  The marker (first example) shown in FIG. 26 is a marker in which vertical lines 95 are arranged at regular intervals like a ruler. Symbols (numbers) as shown in the figure may be attached. The cutting position is determined by reading a ruler memory.

  Markers (second example) shown in FIGS. 27A to 27C indicate the positions of the portions where the shape of the spot size converter changes. In this case, for example, an optimal core width can be obtained by cutting between two predetermined markers.

  The markers (third example) shown in FIGS. 28A and 28B indicate the cutting position, the width of the marker 96 on the upper side of each figure is made equal to the cutting width, and the lower marker 97 indicates the cutting word. This shows the position of the end face. Thereby, it is possible to perform cutting in consideration of the cutting width.

  The markers (fourth example) shown in FIGS. 29A and 29B are almost the same as the markers shown in FIGS. 28A and 28B, but the markers are removed in consideration of the amount to be removed by polishing. The point that has been placed is different.

  The present invention is not limited to the above-described embodiments, and various modifications and applications can be made within the scope of the claims.

  For example, as the shape of the tapered portion in the spot size converter of the present invention, not only a linear shape whose core width changes linearly but also a curved shape such as an exponential function or a parabola can be used. For example, the taper shape in the core width conversion part shown in FIG. .13, NO.7, JULY 2001, pp.675-677 ”, the coupling loss can be reduced more effectively by using an optimized taper. Of course, an optimized taper can also be used for the core width fine adjustment section.

  Further, although the embodiment of the present invention has been described by taking a planar optical circuit as an example, the spot size converter of the present invention is not limited to a planar optical circuit, and for example, a multilayer optical circuit or a waveguide It can be applied to any optical circuit such as an optical circuit in which are stacked. The term “optical circuit” in this specification refers to an optical circuit in a broad sense that is not limited to a planar optical circuit or a waveguide optical circuit.

  Furthermore, the main object of the present invention is to reduce the coupling loss between the optical circuit and the optical fiber. However, an optical device composed of a waveguide type optical circuit such as a semiconductor laser can be used as an optical fiber or a planar optical circuit. Even in the case of coupling, the coupling loss can be reduced by using the optimum core width. The present invention can also be applied when different optical circuits are connected to each other.

It is a schematic block diagram of the 1st example of 1st embodiment of the planar optical circuit by this invention. FIG. 2 is an extraction enlarged view of cores of an input waveguide and an output waveguide of the planar optical circuit of FIG. 1. It is explanatory drawing of the preparation procedure of the input waveguide of the planar optical circuit of FIG. 1, and an output waveguide. It is the result of having calculated | required the relationship between the core width in a board | substrate end surface, and coupling loss by calculation. It is explanatory drawing of a clipping error. FIG. 6 is an enlarged view of extraction of cores of an input waveguide and an output waveguide of a second example of the first embodiment of the planar optical circuit according to the present invention. FIG. 5 is an enlarged view of extraction of cores of an input waveguide and an output waveguide of a third example of the first embodiment of the planar optical circuit according to the present invention. FIG. 10 is an enlarged view of extraction around the input / output end sides of the input waveguide and output waveguide of the fourth example of the first embodiment of the planar optical circuit according to the present invention. It is a schematic block diagram of the 1st example of 2nd embodiment of the planar optical circuit by this invention. FIG. 10 is an extraction enlarged view of the cores of the input waveguide and the output waveguide of the planar optical circuit of FIG. 9. It is explanatory drawing of the manufacturing method of the core of FIG. It is explanatory drawing of the other manufacturing method of the core of FIG. It is a schematic block diagram of the 2nd example of 2nd embodiment of the planar optical circuit by this invention. It is a schematic block diagram of 3rd embodiment of the planar optical circuit by this invention. It is an enlarged view of the spot size converter provided in the input-output waveguide of the planar optical circuit of 4th embodiment by this invention. It is the result of having calculated | required the relationship between taper angle 2 (theta) and excess loss by calculation. It is a figure which shows the example of application of the spot size converter in 4th embodiment. It is an enlarged view of a spot size converter provided in an input / output waveguide of a planar optical circuit according to a fifth embodiment of the present invention. It is a figure which shows the example of the core width fine adjustment part in 5th embodiment. It is a figure which shows the example of the core width fine adjustment part in 5th embodiment. It is a figure for demonstrating the definition of an average taper angle. It is a schematic block diagram of the planar optical circuit in the 6th embodiment in this invention. It is a block diagram of the monitor waveguide in 6th embodiment. It is a figure which shows the shape after the cutting | disconnection of the monitor waveguide in 6th embodiment. It is a figure which shows the other example of the monitor waveguide in 6th embodiment. It is a figure which shows the 1st example of the marker in the planar optical circuit of this invention. It is a figure which shows the 2nd example of the marker in the planar optical circuit of this invention. It is a figure which shows the 3rd example of the marker in the planar optical circuit of this invention. It is a figure which shows the 4th example of the marker in the planar optical circuit of this invention. It is a schematic block diagram of an example of the conventional planar optical circuit. It is explanatory drawing of the connection state of the input waveguide and optical fiber of the conventional planar type optical circuit. It is a figure which shows the example of the spot size converter in a prior art.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Planar type optical circuit 11 Board | substrate 12 Input waveguide 12a, 22a, 32a, 42a Core 12aa, 22aa, 42aa Tapered part 32aa Step part 42ab Straight part 13 Slab waveguide 14 Array waveguide 15 1/2 wavelength plate 16 Slab waveguide 17 Output waveguides 17a, 27a, 37a, 47a Cores 17aa, 27aa, 47aa Tapered portions 37aa Stepped portions 47ab Straight portions 50, 60, 70, 90 Arrayed waveguide grating type optical wavelength demultiplexing circuit (AWG)
68 First monitor waveguide 68a Input end 68b Output end 69 Second monitor waveguide 69a Input end 69b Output ends La1, La2, Lb1, Lb2, 23, 78, Markers 20, 75 Spot size converters 21, 76, 80, 85, 88 Core width fine adjustment section 22, 77, 89 Core width conversion section 87 Monitor waveguide 81, 86 Straight section 82 Taper section 92 First monitor waveguide 93 Second monitor waveguide

Claims (5)

In a planar optical circuit in which an input waveguide and an output waveguide in which a core having a high refractive index is surrounded by a clad having a low refractive index are formed on a substrate,
Formed so that the core width of the input end side of the core of the input waveguide or the output end side of the core of the output waveguide gradually becomes narrower as it approaches the substrate end face,
An input end is formed on an edge side different from both the edge side of the substrate where the input end of the input waveguide is located and the edge side of the substrate where the output end of the output waveguide is located, A first monitor waveguide in which a core formed so that the core width on the input end side becomes gradually narrower as it approaches the substrate end surface is surrounded by a clad, and the input end of the first monitor waveguide is cut and cut. A first monitor for determining an optimum core width to be used as a core width of the input end of the input waveguide or the output end of the output waveguide by measuring a coupling loss between the input end of the input waveguide and the optical fiber A waveguide ;
An input end located on the edge end side of the substrate located on the input end side of the input waveguide and having the same shape as the input end side, and the substrate located on the output end side of the output waveguide A second monitor waveguide having an output end located on the edge end side of the output waveguide and having the same shape as the output end, the input end of the input waveguide and the output end of the output waveguide A second monitor waveguide for estimating the coupling loss;
Is formed on the substrate. In a planar optical circuit in which an input waveguide and an output waveguide in which a core having a high refractive index is surrounded by a clad having a low refractive index are formed on a substrate,
Formed so that the core width of the input end side of the core of the input waveguide or the output end side of the core of the output waveguide gradually becomes narrower as it approaches the substrate end face,
An input end is formed on an edge side different from both the edge side of the substrate where the input end of the input waveguide is located and the edge side of the substrate where the output end of the output waveguide is located, A first monitor waveguide in which a core formed such that the core width on the input end side becomes gradually narrower as it approaches the substrate end surface is surrounded by a clad, and the input end of the first monitor waveguide is cut and cut. A first monitor for determining an optimum core width to be used as a core width of the input end of the input waveguide or the output end of the output waveguide by measuring a coupling loss between the input end of the input waveguide and the optical fiber A waveguide ;
An input end located on the edge end side of the substrate located on the input end side of the input waveguide and having the same shape as the input end side, and the substrate located on the output end side of the output waveguide A waveguide having an output end located on the edge side of the substrate and having the same shape as the output end; an edge side of the substrate located on the input end side of the input waveguide; and the output waveguide wherein possess both ends on the edge end of the substrate located in the output end side, a second monitor waveguide and a linear waveguide, the core width is constant, the input end of said input waveguide And a second monitor waveguide for estimating a coupling loss at the output end of the output waveguide;
Is formed on the substrate. An optical circuit comprising a substrate, a waveguide in which a core having a high refractive index is embedded in a cladding having a low refractive index, and a spot size converter disposed as a part of the waveguide near the end face of the substrate. ,
Monitor waveguide having a spot size converter to the substrate end face side is formed in a plurality on the substrate, further, the spot size converter in said plurality of monitor waveguides, the longitudinal direction at a distance of those said monitor waveguide Are just offset from each other,
The plurality of monitor waveguides measure a coupling loss between the spot size converters of the plurality of monitor waveguides and the optical fiber after the spot size converters of the plurality of monitor waveguides are cut at a predetermined cutting position. Thus, for obtaining the optimum core width to be used as the core width on the substrate end face side of the waveguide spot size converter,
The spot size converter of the waveguide is disposed on the substrate such that the distance between the substrate end surface and the spot size converter is larger than the distance between the substrate end surface and the predetermined cutting position.
Each of the spot size converter of the waveguide and the spot size converter of the plurality of monitor waveguides is formed so that the core width is minimized on the substrate end surface, and the core width fine adjustment unit on the substrate end surface side and An optical circuit characterized in that an average taper angle θ ₁ of the core width fine adjustment unit is larger than 0 ° and smaller than an average taper angle θ ₂ of the core width conversion unit. . A substrate, a waveguide embedded high refractive index core with a low refractive index cladding, a composed optical circuit and a spot size converter disposed in the vicinity of the substrate end surface as part of the waveguide,
As the core width approaches the substrate end surface, the core width gradually narrows toward the edge side that is different from both the edge side of the substrate where the input end of the waveguide is located and the edge side of the substrate where the output end of the waveguide is located. A first monitor waveguide having a spot size converter formed as described above , wherein an input end of the spot size converter of the first monitor waveguide is cut, and the input end after cutting is coupled to the optical fiber A first monitor waveguide for determining the optimum core width to be used as the core width on the substrate end face side of the waveguide spot size converter by measuring the loss;
A second monitor waveguide comprising spot size converters on both edge sides of the substrate where the input end and output end of the waveguide are located, the input end of the waveguide and the output end of the waveguide a second monitor waveguide to estimate the coupling loss in a light circuit which is formed on said substrate,
Each of the spot size converters of the waveguide and the spot size converters at both ends of the second monitor waveguide is formed so that the core width is minimized at the substrate end surface, and the core width fine adjustment unit on the substrate end surface side and and a core width converting part which follows, the mean taper angle theta ₁ of the core width fine-tuning part, characterized in that 0 is smaller than the average taper angle theta ₂ larger of said core width converting part than ° Optical circuit . A substrate, a waveguide embedded high refractive index core with a low refractive index cladding, a composed optical circuit and a spot size converter disposed in the vicinity of the substrate end surface as part of the waveguide,
As the core width approaches the substrate end surface, the core width gradually narrows toward the edge side that is different from both the edge side of the substrate where the input end of the waveguide is located and the edge side of the substrate where the output end of the waveguide is located. A first monitor waveguide having a spot size converter formed as described above , wherein an input end of the spot size converter of the first monitor waveguide is cut, and the input end after cutting is coupled to the optical fiber A first monitor waveguide for determining the optimum core width to be used as the core width on the substrate end face side of the waveguide spot size converter by measuring the loss;
Possess said waveguide input end and the spot size converter with waveguide at both edge side comprises a spot size converter substrate located at the output end, the two ends on both edges end side of the substrate, core width And a second monitor waveguide for estimating a coupling loss at the input end of the waveguide and the output end of the waveguide. an optical circuit formed on a substrate,
The spot size converter of the waveguide and the spot size converters at both ends of the waveguide with the spot size converter are each formed so that the core width is the smallest at the substrate end face, and the core width on the substrate end face side is small. An adjustment section and a core width conversion section that follows the adjustment section, and an average taper angle θ ₁ of the core width fine adjustment section is larger than 0 ° and smaller than an average taper angle θ ₂ of the core width conversion section. An optical circuit .

Images