Optical multiplexer

The present invention relates to optical multiplexers, and particularly, to a multimode-interference optical multiplexer.

Multimode-interference optical multiplexers are used as optical multiplexers in optical integrated circuits. Loss reduction and reflection reduction in the optical multiplexers are required along with the proceeding of integration of a plurality of light sources.

For instance, Japanese Patent Application Laid-Open No. 2010-237376 (Patent Document 1) discloses an optical modulator that includes a Mach-Zehnder waveguide disposed on a dielectric substrate. An example of the dielectric is lithium niobate. The Mach-Zehnder waveguide includes an output Y-multiplexer having a multimode waveguide through which light that has been multiplexed passes. A subsidiary output waveguide is connected to a location where the multimode waveguide is changed into a main output waveguide. The subsidiary output waveguide is a waveguide for higher-order mode where radiation-mode light is output.

Patent Document

Patent Document 1: Japanese Patent Application Laid-Open No. 2010-237376

Problem to be Solved by the Invention

The waveguides in the technique described in the above document are made of, for instance lithium niobate, which is a ferroelectric material. A semiconductor waveguide having a similar function has a design size that is one-tenth or less of that of the lithium niobate waveguide. Hence, the interval between the main output waveguide and the subsidiary output waveguide, which is a waveguide for higher-order mode, is narrowed. This leads to difficulty in fine processing through etching for forming these waveguides. The shapes of these waveguides consequently tend to deviate from design, thereby possibly increasing undesirable light reflection.

To solve this problem, it is an object of the present invention to provide an optical multiplexer that reduces light reflection.

Means to Solve the Problem

An optical multiplexer in the present invention includes two input waveguides, one output waveguide, a multi-mode-interference part, and two unwanted-light waveguides. The input waveguides are made of a semiconductor material. The output waveguide is made of a semiconductor material. The multi-mode-interference part is made of a semiconductor material. The multi-mode-interference part has an incoming end surface connected to the input waveguides, and an outgoing end surface opposite to the incoming end surface and connected to the output waveguide. The multi-mode-interference part has a waveguide width wider than the waveguide widths of the input waveguides and the waveguide width of the output waveguide. The unwanted-light waveguides are made of a semiconductor material. The unwanted-light waveguides are connected to the outgoing end surface of the multi-mode-interference part so as to sandwich the output waveguide. The unwanted-light waveguides each satisfy a single-mode condition.

Effects of the Invention

According to the present invention, the unwanted-light waveguides satisfy the single-mode condition. This reduces the waveguide widths of the unwanted-light waveguides when compared to an instance where the unwanted-light waveguides does not satisfy the single-mode condition. Accordingly, the interval between each unwanted-light waveguide and the output waveguide increases. Such an increased interval facilitates fine processing through etching for forming these waveguides. This enables shaping of the waveguides in conformity with design with accuracy, thereby preventing increase in light reflection resulting from processing error. That is, the light reflection reduces. The waveform of light consequently improves.

These and other objects, features, aspects and advantages of the present invention will become more apparent from the following detailed description of the present invention when taken in conjunction with the accompanying drawings.

FIG. 1 is a schematic plan view of the configuration of an optical multiplexer according to a first embodiment of the present invention.

FIG. 2 is a schematic perspective view of the configuration of the optical multiplexer according to the first embodiment of the present invention.

FIG. 3 is a plan view of an example of the calculation result of the light intensity distribution of transmitted light in the optical multiplexer according to the first embodiment of the present invention.

FIG. 4 is a graph showing an example of the calculation result of the relationship between reflectance and the angle between an output waveguide and an unwanted-light waveguide.

FIG. 5 is a plan view of the configuration of an optical multiplexer in a comparative example.

FIG. 6 is a plan view of the configuration of an optical multiplexer in a comparative example.

FIG. 7 is a schematic plan view of the configuration of an optical multiplexer according to a second embodiment of the present invention.

FIG. 8 is a schematic plan view of the configuration of an optical multiplexer according to a third embodiment of the present invention.

FIG. 9 is a schematic plan view of the configuration of an optical multiplexer according to a fourth embodiment of the present invention.

FIG. 10 is a schematic plan view of the configuration of an optical multiplexer according to a fifth embodiment of the present invention.

FIG. 11 is a schematic plan view of the configuration of an optical multiplexer according to a sixth embodiment of the present invention.

Embodiments of the present invention will be described with reference to the drawings. Identical or corresponding components are denoted by the same reference numerals, and will not be elaborated upon.

First Embodiment

FIG. 1 is a schematic plan view of the configuration of an optical multiplexer 100 according to a first embodiment of the present invention. FIG. 2 is a perspective view of the same. The optical multiplexer 100 includes a substrate 50 , a core layer 51 , and cladding layers 52 and 53 . The core layer 51 is disposed between the cladding layers 52 and 53 in its thickness direction. The core layer 51 and the cladding layer 52 and 53 constitute a stack, and this stack is disposed on the substrate 50 . The optical multiplexer 100 is a multi-mode-interference (MMI) optical multiplexer. Specifically, the optical multiplexer 100 includes two input waveguides 21 and 22 , an output waveguide 31 , a multi-mode-interference part 10 , and two unwanted-light waveguides 41 and 42 , each of which is formed of the aforementioned stack.

The input waveguides 21 and 22 are made of a semiconductor material. The output waveguide 31 is made of a semiconductor material. The multi-mode-interference part 10 is made of a semiconductor material. The unwanted-light waveguides 41 and 42 are made of a semiconductor material. These semiconductor materials may be the same. A semiconductor optical multiplexer is smaller than an optical multiplexer of lithium niobate (LN), which is a dielectric material, by one-tenth or less. Examples of these semiconductor materials include indium-phosphide (InP), indium-gallium-arsenide-phosphide (InGaAsP), aluminum-gallium-indium-arsenide (AlGalnAs), and indium-gallium-arsenide (InGaAs).

The multi-mode-interference part 10 has an incoming end surface 10 a and an outgoing end surface 10 b opposite the incoming end surface 10 a . The incoming end surface 10 a is a surface through which light is incident on the multi-mode-interference part 10 . The incident light may be any of single-mode light and multi-mode light. The outgoing end surface 10 b is a surface through which the light from the multi-mode-interference part 10 is emitted. The emitted light may be any of single-mode light and multi-mode light. The multi-mode-interference part 10 also has an end surface 10 c and an end surface 10 d opposite the end surface 10 c . A straight line passing through the center of an incoming end surface and the center of an outgoing end surface in plan view is defined as a center line CL. In the plan view of FIG. 1 , the straight line passing through the center of the incoming end surface 10 a and the center of the outgoing end surface 10 b is a center line CL. The incoming end surface 10 a , the outgoing end surface 10 b , the end surface 10 c , and the end surface 10 d form a substantially rectangular shape in FIG. 1 . In other words, the multi-mode-interference part 10 is substantially rectangular in plan view. The size of the incoming end surface 10 a (the size in the vertical direction in FIG. 1 ) is substantially the same as the size of the outgoing end surface 10 b (the size in the vertical direction in FIG. 1 ).

The input waveguides 21 and 22 are connected to the incoming end surface 10 a . In other words, the incoming end surface 10 a is provided with input ports P 21 and P 22 respectively connected to the input waveguides. 21 and 22 . The input ports P 21 and P 22 are preferably disposed on the incoming end surface 10 a so as to be almost symmetrical with respect to the center line CL.

The multi-mode-interference part 10 has a waveguide width W 2 that is wider than the waveguide width W 0 a of the input waveguide 21 , the waveguide width W 0 b of the input waveguide 22 , and the waveguide width W 0 c of the output waveguide 31 . The waveguide width W 2 corresponds to the aforementioned size of the incoming end surface 10 a.

The multi-mode-interference part 10 is formed in such a manner that convergent light can be obtained at one location on the outgoing end surface 10 b . The output waveguides 31 is connected to the outgoing end surface 10 b . In other words, the outgoing end surface 10 b is provided with an output port P 31 connected to the output waveguide 31 . The output port P 31 is a port for outputting (taking out) the aforementioned convergent light from the multi-mode-interference part 10 . The output port P 31 is preferably disposed in the middle of the outgoing end surface 10 b . In other words, a distance X 3 is about half the waveguide width W 2 in FIG. 1 . Here, the distance X 3 is the distance from the upper end of the incoming end surface 10 a in FIG. 1 (i.e., from the end surface 10 c ) to the center in the width direction of the output port P 31 (the vertical direction in FIG. 1 ).

The input ports P 21 and P 22 are disposed in such a manner that the convergent light is positioned in the vicinity of the middle of the outgoing end surface 10 b . Specifically, the input port P 21 is disposed to be away from the upper end of the incoming end surface 10 a in FIG. 1 (i.e., from the end surface 10 e ) by about a quarter of the size of the incoming end surface 10 a (the size in the vertical direction in FIG. 1 ). Moreover, the input port P 22 is disposed to be away from the upper end of the incoming end surface 10 a in FIG. 1 by about three-quarters of the size of the incoming end surface 10 a . In other words, the input port P 21 is disposed to be away from the lower end of the incoming end surface 10 a in FIG. 1 (i.e., from the end surface 10 d ) by about three-quarters of the size of the incoming end surface 10 a . Moreover, the input port P 22 is disposed to be away from the lower end of the incoming end surface 10 a in FIG. 1 by about a quarter of the size of the incoming end surface 10 a.

In other words, distances X 1 and X 2 are each about a quarter of the waveguide width W 2 . Here, the distance X 1 is the distance from the upper end of the incoming end surface 10 a in FIG. 1 (i.e., from the end surface 10 c ) to the center in the width direction of the input port P 21 (the vertical direction in FIG. 1 ). Moreover, the distance X 2 is the distance from the lower end of the incoming end surface 10 a in FIG. 1 (i.e., from the end surface 10 d ) to the center in the width direction of the input port P 22 (the vertical direction in FIG. 1 ).

The unwanted-light waveguides 41 and 42 are connected to the outgoing end surface 10 b of the multi-mode-interference part 10 . In other words, the outgoing end surface 10 b is provided with unwanted-light ports P 41 and P 42 respectively connected to the unwanted-light waveguides 41 and 42 . The unwanted-light waveguides 41 and 42 are connected to the outgoing end surface 10 b so as to sandwich the output waveguide 31 . In other words, the respective unwanted-light waveguides 41 and 42 are connected onto the outgoing end surface 10 b on one and the other sides of the output waveguide 31 . In other words, the output waveguide 31 is connected to the outgoing end surface 10 b , between the unwanted-light waveguides 41 and 42 . The unwanted-light ports P 41 and P 42 are preferably disposed on the outgoing end surface 10 b so as to be almost symmetrical with respect to the center line CL.

The unwanted-light waveguides 41 and 42 respectively have waveguide widths W 1 a and W 1 b narrower than the waveguide width W 2 of the multi-mode-interference part 10 . The unwanted-light waveguides 41 and 42 each satisfy a single-mode condition. Thus, light entering the unwanted-light waveguides 41 and 42 propagates through the unwanted-light waveguides 41 and 42 not in higher-order mode, but in fundamental mode. The unwanted-light waveguides 41 and 42 preferably, respectively extend at angles θa and θb greater than 0 degrees and smaller than 85 degrees with respect to the output waveguide 31 . In the following description, the angles θa and θb as a whole can be referred to as an angle θ.

An example of the single-mode condition will be described. Consider a slab waveguide that includes the core layer 51 having a refractive index of n 1 , the cladding layer 52 having a refractive index of n 2 , and the cladding layer 53 having a refractive index of n 3 . Let the thickness of the core layer 51 be denoted by d; moreover, the wavelength of light in a vacuum, by λ 0 . Further, let the following equations be defined:

a ( n 3 2 −n 1 2 )/( n 2 2 −n 3 2 ); and

V =(2π/λ 0 )· d ·( n 2 2 −n 3 2 ) 1/2 .

In this case, the aforementioned waveguide satisfies the single-mode condition when the following expression is satisfied:

V <(π/2)+(½)·arctan( a 1/2 ).

Light propagates through a waveguide satisfying the single mode in the fundamental mode.

The unwanted-light waveguides 41 and 42 prevent light reflection in a portion of the outgoing end surface 10 b where the output port P 31 is not disposed. In other words, the unwanted-light waveguide 41 or 42 takes out light that would reflect without the unwanted-light waveguides 41 and 42 , to the outside of the multi-mode-interference part 10 via the unwanted-light port P 41 or P 42 .

FIG. 3 is a plan view of an example of the calculation result of the light intensity distribution of transmitted light when a beam of transverse-electric-field (TE) polarized light having a wavelength of 1.295 μm enters from the input port P 21 of the optical multiplexer 100 ( FIG. 1 ). The “transmitted light” in this calculation is light passing from the input port P 21 toward the output port P 31 in the optical multiplexer 100 . In the drawing, a brighter site indicates a higher light intensity. Furthermore, the white lines denote the outline of the optical multiplexer 100 , and the rectangular portion corresponds to the multi-mode-interference part 10 . Here, “calculation” means calculation using a computer or other things, and will be similarly defined in the following description.

According to the calculation result ( FIG. 3 ), unwanted light has a high intensity in positions of the outgoing end surface 10 b , corresponding to the input ports P 21 and P 22 of the incoming end surface 10 a . In other words, the unwanted light has a high intensity in positions away from the upper end of the outgoing end surface 10 b in FIG. 3 (i.e., from the end surface 10 c in FIG. 1 ), by about a quarter and three-quarters of the size of the incoming end surface 10 a (the size in the vertical direction in FIG. 3 ). This result reveals that the unwanted-light port P 41 is preferably disposed to be away from the upper end of the outgoing end surface 10 b in FIG. 3 (i.e., from the end surface 10 c in FIG. 1 ) by about a quarter of the size of the outgoing end surface 10 b (the size in the vertical direction in FIG. 1 ), and that the unwanted-light port P 42 is preferably disposed to be away from the upper end of the outgoing end surface 10 b in FIG. 3 by about three-quarters of the size of the outgoing end surface 10 b . In other words, the unwanted-light port P 41 is preferably disposed to be away from the lower end of the outgoing end surface 10 b in FIG. 3 (i.e., from the end surface 10 d in FIG. 1 ) by about three-quarters of the size of the outgoing end surface 10 b ; moreover, the unwanted-light port P 42 is preferably disposed to be away from the lower end of the outgoing end surface 10 b in FIG. 3 by about a quarter of the outgoing end surface 10 b . In other words, in FIG. 1 , distances X 4 and X 5 are each preferably about a quarter of the waveguide width W 2 . Here, the distance X 4 is the distance from the upper end of the outgoing end surface 10 b in FIG. 1 (i.e., from the end surface 10 c in FIG. 1 ) to the center in the width direction of the unwanted-light port P 41 (the vertical direction in FIG. 1 ). Moreover, the distance X 5 is the distance from the lower end of the outgoing end surface 10 b in FIG. 1 (i.e., from the end surface 10 d in FIG. 1 ) to the center in the width direction of the unwanted-light port P 42 (the vertical direction in FIG. 1 ).

The following details a size condition used in the aforementioned calculation. The waveguide width W 0 a of the input waveguide 21 , the waveguide width W 0 b of the input waveguide 22 , and the waveguide width W 0 c of the output waveguide 31 are 1.4 μm. The waveguide width W 1 a of the unwanted-light waveguide 41 and the waveguide width W 1 b of the unwanted-light waveguide 42 are 2.1 μm. The waveguide width W 2 of the multi-mode-interference part 10 is 12.0 μm. The distances X 1 and X 2 are each 10 μm. The distance X 3 is 6.0 μm. The distances X 4 and X 5 are each 3.0 μm. The length L of each of the end surfaces 10 c and 10 d is 179 μm.

The angle θa between the unwanted-light waveguide 41 and the output waveguide 31 and the angle θb between the unwanted-light waveguide 42 and the output waveguide 31 , are greater than 0 degrees and smaller than 85 degrees, and are angles at which etching can be performed with sufficient accuracy in the production of the optical multiplexer 100 . Here, the angle θ has a positive symbol when, as indicated by the arrows in FIG. 1 , the interval between the output waveguide 31 and each of the unwanted-light waveguides 41 , 42 increases along with distance from the outgoing end surface 10 b . In the aforementioned calculation, the angles θa and θb are equal.

In the aforementioned calculation, the input ports P 21 and P 22 are disposed on the incoming end surface 10 a so as to be almost symmetrical with respect to the center line CL. Further, the output port P 31 is disposed on the outgoing end surface 10 b so as to be located on the center line CL. The unwanted-light ports P 41 and P 42 are disposed on the outgoing end surface 10 b so as to be almost symmetrical with respect to the center line CL. For a size equivalent to the aforementioned specific size, a margin of error about 0.2 μm is considered to not have a great adverse effect on the positions of the individual ports. Such a margin of error is considered to be allowable in the other embodiments.

FIG. 4 is a graph showing an example of the calculation result of the relationship between reflectance, and the angle θ (the angles θa and θb in FIG. 1 ) in a range of 0 degrees≤θ≤85 degrees. Here, the reflectance is the ratio of the intensity of reflection light (reflected return light) that reflects on the output end surface 10 b and then joins to the input waveguide 21 , to the intensity of incident light. FIG. 5 is a plan view of the configuration of an optical multiplexer 100 A without the unwanted-light waveguides 41 and 42 , in a comparative example. As indicated by the solid line in FIG. 4 , the reflectance of the optical multiplexer 100 A in the comparative example is calculated to be −12.71 dB. Meanwhile, as indicated by the plot dots in FIG. 4 , the graph reveals that the reflectance of the optical multiplexer 100 in the present embodiment, which includes the unwanted-light waveguides 41 and 42 , is noticeably lower than the reflectance of the optical multiplexer optical 100 A, i.e., −12.71 dB, particularly when the angle θ is smaller than 85 degrees.

FIG. 6 is a plan view of the configuration of an optical multiplexer 100 B in a comparative example that corresponds to an instance where the angle θ is 0 degrees. The reflectance of the optical multiplexer 100 B corresponds to the angle θ=0 degrees in FIG. 4 , and is calculated to be −56.65 dB. The optical multiplexer 100 B thus has a sufficiently low reflectance. However, the angle θ of 0 degrees leads to difficulty in fine processing through etching for forming the output waveguide 31 and the unwanted-light waveguides 41 and 42 . The following describes the details. It is noted that a direction from the incoming end surface 10 a toward the outgoing end surface 10 b (the rightward direction in FIG. 1 ) is also referred to as a light propagation direction.

The interval Da between the unwanted-light waveguide 41 and the output waveguide 31 in a position away from the outgoing end surface 10 b in the light propagation direction by a distance Z ( FIG. 1 ) is expressed by the following equation:

Da={W 2/4− W 1 a /(2·cos θ a )− W 0/2}+ Z ·tan θ a.

Likewise, the interval Db between the unwanted-light waveguide 42 and the output waveguide 31 in a position away from the outgoing end surface 10 b in the light propagation direction by the distance Z is expressed by the following equation:

Db={W 2/4− W 1 b /(2·cos θ b )− W 0/2}+ Z ·tan θ b.

When the angles θa and θb are smaller than 0 degrees, the intervals Da and Db get smaller as the distance Z increases. In other words, the interval between each of the unwanted-light waveguides 41 , 42 and the output waveguide 31 decreases along with distance from the outgoing end surface 10 b . This leads to difficulty in fine processing through etching for forming the output waveguide 31 and the unwanted-light waveguides 41 and 42 . Further, when the angles θa and θb are 0 degrees, each of the unwanted-light waveguides 41 and 42 and the output waveguide 31 extend in an outgoing direction with the interval Da and Db kept minimum, as illustrated in FIG. 6 . This also leads to difficulty in fine processing for forming the output waveguide 31 and the unwanted-light waveguides 41 and 42 . Furthermore, the angles θa and θb, even though designed to be 0 degrees, can be actually smaller than 0 degrees due to manufacture error. Accordingly, as illustrated in FIG. 1 , the angles θa and θb are preferably greater than 0 degrees.

The configuration of the optical multiplexer 100 in FIG. 1 is one example. Any change may be made to, for instance, the sizes of the multi-mode-interference part, the input waveguides, the output waveguide, and the unwanted-light waveguides, and the positions of the ports.

According to the present embodiment, the unwanted-light waveguides 41 and 42 satisfy the single-mode condition. This reduces the waveguide widths W 1 a and W 1 b of the unwanted-light waveguides 41 and 42 when compared to an instance where the unwanted-light waveguides 41 and 42 do not satisfy the single-mode condition. Accordingly, the interval between each of the unwanted-light waveguides 41 , 42 and the output waveguide 31 expands. Such an expanded interval facilitates fine processing through etching for forming these waveguides. In one example, an experimental result was obtained that a mask margin improved by 10%. This improvement enables shaping of the waveguides in conformity with design with accuracy, thereby preventing increase in light reflection resulting from processing error. That is, the light reflection reduces. The waveform of light consequently improves.

Each of the unwanted-light waveguides 41 and 42 preferably extends at the angle θ greater than 0 degrees and smaller than 85 degrees with respect to the output waveguide 31 . For the angle θ greater than 0 degrees, the interval between each of the unwanted-light waveguides 41 , 42 and the output waveguide 31 increases along with distance from the outgoing end surface 10 b . In this case, the interval between each of the unwanted-light waveguides 41 , 42 and the output waveguide 31 , in a location away from the outgoing end surface 10 b is greater than the interval between each of the unwanted-light waveguides 41 , 42 and the output waveguide 31 , on the outgoing end surface 10 b . This facilitates fine processing for forming the waveguides not only in a location on the outgoing end surface 10 b , but also in a location away from the outgoing end surface 10 b , thereby enabling shaping of the waveguides in conformity with design with accuracy. Increase in light reflection resulting from processing error is consequently further prevented. That is, the light reflection reduces. In addition, the angle θ smaller than 85 degrees further reduces the light reflection, as illustrated in FIG. 4 .

Second Embodiment

FIG. 7 is a schematic plan view of the configuration of an optical multiplexer 101 according to a second embodiment of the present invention. The configuration of the optical multiplexer 101 will be outlined firstly.

The optical multiplexer 101 includes a multi-mode-interference part 11 instead of the multi-mode-interference part 10 ( FIG. 1 ). The multi-mode-interference part 11 is made of a semiconductor material similar to that of the multi-mode-interference part 10 ( FIG. 1 : the first embodiment). The multi-mode-interference part 11 is provided with an incoming end surface 11 a , an outgoing end surface 11 b , an end surface 11 c , and an end surface 11 d , which respectively correspond to the incoming end surface 10 a , outgoing end surface 10 b , end surface 10 c , and end surface 10 d of the multi-mode-interference part 10 . A straight line passing through the center of the incoming end surface 11 a and the center of the outgoing end surface 11 b in plan view in FIG. 7 is a center line CL.

The multi-mode-interference part 11 includes a rectangular portion 11 n having a substantially rectangular shape, which is similar to the shape of the multi-mode-interference part 10 ( FIG. 1 ), and a tapered portion TP 1 having a substantially trapezoidal shape. The tapered portion TP 1 specifically has a shape tapering down toward the outgoing end surface 11 b . Here, the “shape tapering down toward the outgoing end surface 11 b ” means a shape such that a width size (the size in the vertical direction in FIG. 7 ) decreases toward the outgoing end surface 11 b (toward the rightward direction in FIG. 7 ).

The rectangular portion 11 n has the incoming end surface 11 a connected to the two input waveguides 21 and 22 In other words, the incoming end surface 11 a is provided with the input ports P 21 and P 22 . The input ports P 21 and P 22 are preferably disposed on the incoming end surface 11 a so as to be almost symmetrical with respect to the center line CL. The opposite side of the rectangular portion 11 n from the incoming end surface 11 a is connected to the tapered portion TP 1 . Both ends of the incoming end surface 11 a are connected to the end surfaces 11 c and 11 d opposite to each other.

The tapered portion TP 1 has the outgoing end surface 11 b connected to the output waveguide 31 and the two unwanted-light waveguides 41 , 42 . In other words, the outgoing end surface 11 b is provided with the output port P 31 and the unwanted-light ports P 41 , P 42 . The opposite side of the tapered portion TP 1 from the outgoing end surface 11 b is connected to the rectangular portion 11 n . Both ends of the outgoing end surface 11 b are connected to the end surfaces 11 c and 11 d , which are opposite to each other.

The end surfaces 11 c and 11 d are substantially parallel with each other in the rectangular portion 11 n ; moreover, their interval decreases toward the outgoing end surface 11 b in the tapered portion TP 1 . The size of the outgoing end surface 11 b (the size in the vertical direction in FIG. 7 ) is thus smaller than the size of the incoming end surface 11 a (the size in the vertical direction in FIG. 7 ).

Configuration other than that described above, which is almost similar to the configuration described in the first embodiment, will be detailed below.

The multi-mode-interference part 11 has a waveguide width W 2 wider than the waveguide width W 0 a of the input waveguide 21 , the waveguide width W 0 b of the input waveguide 22 , and the waveguide width W 0 c of the output waveguide 31 . The waveguide width W 2 corresponds to the aforementioned size of the incoming end surface 11 a . In the present embodiment, the waveguide width W 2 is the waveguide size of the rectangular portion 11 n of the multi-mode-interference part 11 .

The multi-mode-interference part 11 is formed in such a manner that convergent light can be obtained at one location on the outgoing end surface 11 b . The output port P 31 is a port for outputting (taking out) this convergent light from the multi-mode-interference part 11 . The output port P 31 is preferably disposed in the middle of the outgoing end surface 11 b . In other words, a distance X 3 is about half the waveguide width W 2 in FIG. 7 . Here, the distance X 3 is the distance from the upper end of the incoming end surface 11 a in FIG. 7 (i.e., from the end surface 11 c ) to the center in the width direction of the output port P 31 (the vertical direction in FIG. 7 ).

The input ports P 21 and P 22 are disposed in such a manner that the convergent light is positioned in the vicinity of the middle of the outgoing end surface 10 b . Specifically, the input port P 21 is disposed to be away from the upper end of the incoming end surface 11 a in FIG. 7 (i.e., from the end surface 11 c ) by about a quarter of the size of the incoming end surface 11 a (the size in the vertical direction in FIG. 7 ). Moreover, the input port P 22 is disposed to be away from the upper end of the incoming end surface 11 a in FIG. 7 by about three-quarters of the size of the incoming end surface 11 a . In other words, the input port P 21 is disposed to be away from the lower end of the incoming end surface 11 a in FIG. 7 (i.e., from the end surface 11 d ) by about three-quarters of the size of the incoming end surface 11 a . Moreover, the input port P 22 is disposed to be away from the lower end of the incoming end surface 11 a in FIG. 7 by about a quarter of the size of the incoming end surface 11 a . In other words, distances X 1 and X 2 are each about a quarter of the waveguide width W 2 in FIG. 7 . Here, the distance X 1 is the distance from the upper end of the incoming end surface 11 a in FIG. 7 (i.e., from the end surface 11 c ) to the center in the width direction of the input port P 21 (the vertical direction in FIG. 7 ). Moreover, the distance X 2 is the distance from the upper end of the incoming end surface 11 a in FIG. 7 (i.e., from the end surface 11 d ) to the center in the width direction of the input port P 22 (the vertical direction in FIG. 7 ).

The unwanted-light waveguides 41 and 42 are connected to the outgoing end surface 11 b so as to sandwich the output waveguide 31 . In other words, the respective unwanted-light waveguides 41 and 42 are connected onto the outgoing end surface 11 b on one and the other sides of the output waveguide 31 . In other words, the output waveguide 31 is connected to the outgoing end surface 11 b , between the unwanted-light waveguides 41 and 42 . The unwanted-light ports P 41 and P 42 are preferably disposed on the incoming end surface 11 b so as to be almost symmetrical with respect to the center line CL.

The unwanted-light waveguides 41 and 42 have waveguide widths W 1 a and W 1 b narrower than the waveguide width W 2 of the multi-mode-interference part 11 . The unwanted-light waveguides 41 and 42 each satisfy a single-mode condition. Thus, light entering the unwanted-light waveguides 41 and 42 propagates through the unwanted-light waveguides 41 and 42 not in higher-order mode, but in fundamental mode. The unwanted-light waveguides 41 and 42 preferably, respectively extends at angles θa and θb greater than 0 degrees and smaller than 85 degrees with respect to the output waveguide 31 .

For a reason similar to that in the first embodiment, the unwanted-light port P 41 is preferably disposed on the outgoing end surface 11 b so as to be away from a point P ( FIG. 7 ) by about a quarter of the waveguide width W 2 ; moreover, the unwanted-light port P 41 is preferably disposed on the outgoing end surface 11 b so as to be away from the point P by about three-quarters of the waveguide width W 2 . Here the point P is the intersection of an extension line 11 c ′ extended from the end surface 11 c of the rectangular portion 11 n and an extension line 11 b ′ extended from the outgoing end surface 11 b of the rectangular portion 11 n . In other words, distances X 4 and X 5 are each preferably about a quarter of the waveguide width W 2 in FIG. 7 . Here, the distance X 4 is the distance from the end surface 11 c of the rectangular portion 11 n to the center in the width direction of the unwanted-light port P 41 (the vertical direction in FIG. 7 ). Moreover, the distance X 5 is the distance from the end surface 11 d of the rectangular portion 11 n to the center in the width direction of the unwanted-light port P 42 (the vertical direction in FIG. 7 ).

In the aforementioned arrangement, the width of the outgoing end surface 11 b is smaller than the width of the incoming end surface 11 a by {W 2 /2−W 1 a /(2·cos θa)−W 1 b /(2·cos θb}. The unwanted-light ports P 41 and P 42 in the present embodiment are thus disposed at both ends of the outgoing end surface 11 b.

In the optical multiplexer 101 like the optical multiplexer 100 ( FIG. 1 ), the waveguide width W 0 a may be, for instance, 1.4 μm; the waveguide width W 0 b, 1.4 μm; the waveguide width W 0 c, 1.4 μm; the waveguide width W 1 a, 2.1 μm; the waveguide width W 1 b, 2.1 μm; the waveguide width W 2 , 12.0 μm; the distances X 1 and X 2 , 3.0 μm; the distance X 3 , 6.0 μm; the distances X 4 and X 5 , 3.0 μm; and a length L, 179 μm. Here, the length L is a size in a direction along the center line CL. Further, the length of the tapered portion TP 1 , which is included in the length L, can be 20 μm.

The configuration of the optical multiplexer 101 in FIG. 7 is one example. Any change may be made to, for instance, the sizes of the multi-mode-interference part, the input waveguides, the output waveguide, and the unwanted-light waveguides, and the positions of the ports.

According to the present embodiment, the multi-mode-interference part 11 includes the tapered portion TP 1 having a shape tapering down toward the outgoing end surface 11 b . Accordingly, the outgoing end surface 11 b has portions with a small size where the waveguides are not connected. This reduces light reflection on the outgoing end surface 11 b.

Third Embodiment

FIG. 8 is a schematic plan view of the configuration of an optical multiplexer 102 according to a third embodiment of the present invention. The unwanted-light waveguides 41 and 42 each have one end (the left end in the drawing) connected to the outgoing end surface 10 b of the multi-mode-interference part 10 , and the other end (the right end in the drawing) opposite the one end. The optical multiplexer 102 has absorption layers 61 and 62 that are connected to the respective other ends of the unwanted-light waveguide 41 and 42 , and absorb respective light beams from the unwanted-light waveguides 41 and 42 . In manufacturing an apparatus that includes the optical multiplexer 102 integrated with a laser, the absorption layers 61 and 62 can be formed by, for instance, leaving an active layer of the laser.

It is noted that configuration other than that described above is almost the same as the configuration of the optical multiplexer 100 ( FIG. 1 : the first embodiment); thus identical or corresponding components are denoted by the same signs and will not be elaborated upon. It is also noted that the absorption layers 61 and 62 may be used in the optical multiplexer 101 ( FIG. 7 : the second embodiment).

According to the present embodiment, the absorption layers 61 and 62 are disposed at the other ends of the unwanted-light waveguide 41 and 42 . This prevents light entering the unwanted-light waveguides 41 and 42 from returning to the other-mode interferometer due to reflection at the other ends.

Fourth Embodiment

FIG. 9 is a schematic plan view of the configuration of an optical multiplexer 103 according to a fourth embodiment of the present invention. The optical multiplexer 103 has input waveguides 21 A and 22 A instead of the input waveguides 21 and 22 . The optical multiplexer 103 also has an output waveguide 31 A instead of the output waveguide 31 . The input waveguides 21 A and 22 A are made of a semiconductor material similar to that of the input waveguides 21 and 22 ( FIG. 1 : the first embodiment). The output waveguide 31 A is made of a semiconductor material similar to that of the output waveguide 31 ( FIG. 1 : the first embodiment).

The input waveguide 21 A has a waveguide portion 21 n and a tapered portion TP 2 . The tapered portion TP 2 has a shape tapering down from the incoming end surface 10 a . Here, the “shape tapering down from the incoming end surface 10 a ” is a shape such that a width size (the size in the vertical direction in FIG. 9 ) decreases along with distance from the incoming end surface 10 a . The tapered portion TP 2 extends from the input port P 21 of the incoming end surface 10 a , and is connected to the waveguide portion 21 n . The waveguide portion 21 n is thus connected to an end whose width is smallest in the tapered portion TP 2 . The waveguide portion 21 n has a uniform width W 0 a.

Likewise, the input waveguide 22 A has a waveguide portion 22 n and a tapered portion TP 3 . The tapered portion TP 3 has a shape tapering down from the incoming end surface 10 a . The tapered portion TP 3 extends from the input port P 22 of the incoming end surface 10 a , and is connected to the waveguide portion 22 n . The waveguide portion 22 n is thus connected to an end whose width is smallest in the tapered portion TP 3 . The waveguide portion 22 n has a uniform width W 0 b.

The output waveguide 31 A has a waveguide portion 31 n and a tapered portion TP 4 . The tapered portion TP 4 has a shape tapering down from the outgoing end surface 10 b . Here, the “shape tapering down from the outgoing end surface 10 b ” is a shape such that a width size (the size in the vertical direction in FIG. 9 ) decreases along with distance from the outgoing end surface 10 b . The tapered portion TP 4 extends from the output port P 31 of the outgoing end surface 10 b , and is connected to the waveguide portion 31 n . The waveguide portion 31 n is thus connected to an end whose width is smallest in the tapered portion TP 4 . The waveguide portion 31 n has a uniform width W 0 c.

The following describes an example of the detailed configurations of the input waveguides 21 A and 22 A. The waveguide widths W 0 a and W 0 b of the waveguide portions 21 n and 22 n are 1.4 μmm for instance. Further, the tapered portions TP 2 and TP 3 are each 20 μm long in a light propagation direction. The tapered portions TP 2 and TP 3 are trapezoidal. The tapered portions TP 2 and TP 3 are each 3.0 μm wide in the incoming end surface 10 a . The widths of the tapered portions TP 2 and TP 3 decrease from 3.0 μm to a minimum of 1.4 μm along with distance from the respective input ports P 21 and P 22 . It is noted that the shapes of the tapered portions TP 2 and TP 3 are not limited to a trapezoid having straight, oblique lines. For instance, the oblique lines may be not straight, but curved. In this case, respective portions whose widths are smallest in the tapered portions TP 2 and TP 3 can be gently connected to the waveguide portions 21 n and 22 n by the curved lines.

The following describes an example of the detailed configuration of the output waveguide 31 A. For instance, the waveguide width W 0 c of the waveguide portion 31 n is 1.4 μm. Further, the tapered portion TP 4 is 20 μm long in a light propagation direction. The tapered portion TP 4 is trapezoidal. The tapered portion TP 4 is 3.0 μm wide in the outgoing end surface 10 b . The width of the tapered portion TP 4 decreases from 3.0 μm to a minimum of 1.4 μm along with distance from the output port P 31 . It is noted that the shape of the tapered portion TP 4 is not limited to a trapezoid having straight, oblique lines. For instance, the oblique lines may be not straight, but curved. In this case, a portion whose width is smallest in the tapered portion TP 4 can be gently connected to the waveguide portion 31 n by the curved lines.

The configuration of the optical multiplexer 103 in FIG. 9 is one example. Any change may be made to, for instance, the sizes of the multi-mode-interference part, the input waveguides, the output waveguide, and the unwanted-light waveguides, and the positions of the ports. Further, configuration other than that described above is almost the same as the configuration of the optical multiplexer 100 ( FIG. 1 : the first embodiment); thus identical or corresponding components are denoted by the same signs and will not be elaborated upon. Still further, the input waveguides 21 A and 22 A and the output waveguide 31 A may be used in the optical multiplexer 101 ( FIG. 7 : the second embodiment) or the optical multiplexer 102 ( FIG. 8 : the third embodiment).

According to the present embodiment, even if the shape of the optical multiplexer 103 is depart from design in some degree due to processing error, loss and reflection do not greatly increase.

Fifth Embodiment

FIG. 10 is a schematic plan view of the configuration of an optical multiplexer 104 according to a fifth embodiment of the present invention. The optical multiplexer 104 has unwanted-light waveguides 41 A and 42 A instead of the unwanted-light waveguides 41 and 42 . The unwanted-light waveguides 41 A and 42 A are made of a semiconductor material similar to that of the unwanted-light waveguides 41 and 42 ( FIG. 1 : the first embodiment).

The unwanted-light waveguide 41 A has a waveguide portion 41 n and a tapered portion TP 5 . The tapered portion TP 5 has a shape tapering down from the outgoing end surface 10 b . Here, the “shape tapering down from the outgoing end surface 10 b ” is a shape such that a width size (the size in the vertical direction in FIG. 10 ) decreases along with distance from the outgoing end surface 10 b . The tapered portion TP 5 extends from the unwanted-light port P 41 of the outgoing end surface 10 b , and is connected to the waveguide portion 41 n . The waveguide portion 41 n is thus connected to an end whose width is smallest in the tapered portion TP 5 . The waveguide portion 41 n has a uniform width W 1 a.

Likewise, the unwanted-light waveguide 42 A has a waveguide portion 42 n and a tapered portion TP 6 . The tapered portion TP 6 has a shape tapering down from the outgoing end surface 10 b . The tapered portion TP 6 extends from the unwanted-light port P 42 of the outgoing end surface 10 b , and is connected to the waveguide portion 42 n . The waveguide portion 42 n is thus connected to an end whose width is smallest in the tapered portion TP 6 . The waveguide portion 42 n has a uniform width W 1 b.

The following describes an example of the detailed configurations of the input waveguides 41 A and 42 A. The waveguide widths W 1 a and W 1 b of the waveguide portions 41 n and 42 n are 2.1 μm for instance. Further, the tapered portions TP 5 and TP 6 are each 20 μm long in a light propagation direction. The tapered portions TP 5 and TP 6 are trapezoidal. The tapered portions TP 5 and TP 6 are each 3.0 μm wide in the outgoing end surface 10 b . The widths of the tapered portions TP 5 and TP 6 decrease from 3.0 μm to a minimum of 2.1 μm along with distance from the respective unwanted-light ports P 41 and P 42 . It is noted that the shapes of the tapered portions TP 5 and TP 6 are not limited to a trapezoid having straight, oblique lines. For instance, the oblique lines may be not straight, but curved. In this case, respective portions whose widths are smallest in the tapered portions TP 5 and TP 6 can be gently connected to the waveguide portions 41 n and 42 n by the curved lines.

The configuration of the optical multiplexer 104 in FIG. 10 is one example. Any change may be made to, for instance, the sizes of the multi-mode-interference part, the input waveguides, the output waveguide, and the unwanted-light waveguides, and the positions of the ports. Further, configuration other than that described above is almost the same as the configuration of the optical multiplexer 100 ( FIG. 1 : the first embodiment); thus identical or corresponding components are denoted by the same signs and will not be elaborated upon. Still further, the unwanted-light waveguides 41 A and 42 A may be used in the optical multiplexer 101 ( FIG. 7 : the second embodiment), the optical multiplexer 102 ( FIG. 8 : the third embodiment), or the optical multiplexer 103 ( FIG. 9 : the fourth embodiment).

According to the present embodiment, unwanted light beams from the multi-mode-interference part 10 efficiently join to the unwanted-light waveguide 41 and 42 .

Sixth Embodiment

FIG. 11 is a schematic plan view of the configuration of an optical multiplexer 105 according to a sixth embodiment of the present invention. The unwanted-light waveguide 41 has one end (the left end in the drawing) connected to the outgoing end surface 10 b of the multi-mode-interference part 10 , and the other end (the right end in the drawing) opposite the one end. The other end includes a terminal surface that is flat. Likewise, the unwanted-light waveguide 42 has one end (the left end in the drawing) connected to the outgoing end surface 10 b of the multi-mode-interference part 10 , and the other end (the right end in the drawing) opposite the one end. The other end includes a terminal surface E 42 that is flat. The terminal surfaces E 41 and E 42 respectively form angles φa and φb with respect to the output waveguide 31 . The terminal surfaces E 41 and E 42 face medium 70 ; and light beams are emitted from the terminal surfaces E 41 and E 42 to the medium 70 . The angles φa and φb satisfy the following expressions in radians, where λ 0 denotes the wavelength of light in a vacuum, where n denotes the refractive index of each medium 70 :

φ a>λ 0 (π· W 1 a·n )+π/2; and

φ b>λ 0 (π· W 1 b·n )+π/2.

It is noted that configuration other than that described above is almost the same as the configuration of the optical multiplexer 100 ( FIG. 1 : the first embodiment); thus identical or corresponding components are denoted by the same signs and will not be elaborated upon. It is also noted that the terminal surfaces E 41 and E 42 may be used in the optical multiplexer 101 ( FIG. 7 : the second embodiment), the optical multiplexer 102 ( FIG. 8 : the third embodiment), the optical multiplexer 103 ( FIG. 9 : the fourth embodiment), or the optical multiplexer 104 ( FIG. 10 : the fifth embodiment).

According to the present embodiment, light beams emitted from the other ends of the unwanted-light waveguides 41 and 42 do not join to a light beam propagating through the output waveguide 31 . This prevents deformation of a beam emitted from the outgoing end surface 10 b.

It is noted that in the present invention, the individual embodiments can be freely combined, or can be modified and omitted as appropriate, within the scope of the invention. While the invention has been shown and described in detail, the foregoing description is in all aspects illustrative and not restrictive. It is therefore understood that numerous modifications and variations can be devised without departing from the scope of the invention.

P 21 , P 22 input port, P 31 output port, P 41 , P 42 unwanted-light port, CL center line, TP 1 to TP 6 tapered portion, 10 , 11 multi-mode-interference part, 10 a , 11 a incoming end surface, 10 b , 11 b outgoing end surface, 10 c , 10 d , 11 c , 11 d end surface, 11 n rectangular portion, 21 , 21 A, 22 , 22 A input waveguide, 21 n , 22 n , 31 n , 41 n , 42 n waveguide portion, 31 31 A output waveguide, 41 , 41 A, 42 , 42 A unwanted-light waveguide, 50 substrate, 51 core layer, 52 , 53 cladding layer, 61 , 62 absorption layer, 100 to 104 optical multiplexer.