JPH10160951A - Optical multiplexing and demultiplexing circuit - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a small-sized optical multiplexing and damutiplexing circuit which multiplexes or demultiplexes a light wave propagated in a light guide with low loss. SOLUTION: The optical multiplexing and demultiplexing circuit is composed of at least light input/output parts 13 and 14 or a multi-mode interference waveguide(MMI) 15 with a light output part so that the intervals d1 and d3 from the input/output waveguide 16 which is arranged as the outermost one to the edge along the same direction with the waveguide direction of the multi- mode interference waveguide are made narrower than the interval d2 of the light input/output waveguide.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to the structure of a small optical multiplexing / demultiplexing circuit for multiplexing or demultiplexing lightwaves transmitted through an optical waveguide with low loss.

[0002]

2. Description of the Related Art An optical multiplexing / demultiplexing circuit is used to optically couple light waves from an optical switch or a plurality of semiconductor laser diodes (LDs) to one or more waveguides or optical fibers. As the optical multiplexing / demultiplexing circuit, a directional coupler, an X-type multiplexing / demultiplexing device, a strong coupling directional coupler, and the like have been often used. These optical multiplexing / demultiplexing circuits use an interference effect between a fundamental mode (symmetric mode) and a first-order mode (antisymmetric mode) of the optical waveguide as an operation principle. For this reason, the device structure, material, and dimensions greatly depend on the device characteristics, and the required manufacturing tolerances thereof are extremely strict. Therefore, there have been serious problems such as a low product manufacturing yield.

On the other hand, as one of the optical multiplexing / demultiplexing circuits, a multi-mode waveguide through which higher-order modes can propagate is used, and the multiplexing / demultiplexing of light waves is performed by utilizing interference between modes in the waveguide. Multimode interference waveguides with functions (hereinafter referred to as “MM
I "(: Multi-Mode Interferometer). ).
The MMI optical multiplexing / demultiplexing circuit has recently been widely used because of its features such as low loss and ease of manufacture.

FIGS. 10 and 11 show examples of the basic configuration of an MMI using a conventional semiconductor. In the case of this prior art, 2
FIG. 10 shows a perspective view and FIG.
1 is a top view thereof. In these drawings, reference numeral 01 denotes a semiconductor substrate, 02 denotes a waveguide core layer, 07 denotes a cladding layer, and 0 denotes a cladding layer.
3 is an input waveguide, 04 is an output waveguide 05 is an MMI area, and 06 is an input / output waveguide. In this configuration,
The waveguide has a ridge structure. In addition to this, in order to protect the waveguide side view, the waveguide is configured to cover a dielectric film, or a semiconductor layer (cladding layer) is buried in the waveguide side by an epitaxial growth method. It is configured as follows. Here, the input / output waveguides 03 and 04 of the MMI are
A structure is adopted in which a spot size of the same size as the guided light of the functional device waveguide such as an optical switch connected to the MI region 05 by hybrid integration or at least monolithically integrated with the semiconductor substrate 01 is used. For example, in the case of a semiconductor MMI having a wavelength band of 1.55 μm,
For the semiconductor substrate 01 and the cladding layer 07, InP is used, and for the core layer 02, InGaAsP or a waveguide material / structure of a semiconductor functional device portion to be monolithically integrated is used.

[0005] The semiconductor feature spot size of the device portion guided light (radius) is generally to be about 0.1-2 .mu.m, input waveguide width _{(w i (= w 0)} ) = 0.5
３3 μm, MMI width (w _g ) = 5-20 μm, and core thickness (t _g ) = 0.1-1 μm. M
In order to reduce the loss and crosstalk of MI and to shorten the waveguide length L of the MMI region 05, in the case of the 2 × 2 configuration shown in FIGS.
It was set in the relationship of d ₁ = d ₂ = d ₃ = w _g / 3.

A multiplexing / demultiplexing circuit using such an MMI is as follows.
Compared to a multiplexing / demultiplexing circuit using a directional coupler, it has been clarified that characteristic variations due to deviations of waveguide dimensions and refractive indexes are small and especially deterioration of crosstalk is small during device manufacturing. It is used in many semiconductor devices. The MMI optical multiplexing / demultiplexing circuit having such features is expected to be applied to an optical device using a dielectric such as LiNbO ₃ or LiTaO ₃ or glass or an organic material.

[0007]

However, since the refractive index Δn between the core and the clad of these non-semiconductor-based optical waveguides is about 1% or less, the size of the guided light spot of the semiconductor (: Δn = several% or more), which is several times the spot size. From this, MMI
Due to the low loss and low crosstalk characteristics described above, the MMI waveguide width w _g also needs to be increased.

However, the optimum MMI device length L
_{Has a} relationship as shown in the following equation (1) with respect to the core width w _g in principle. Therefore, as the core width w _g is increased, L becomes significantly longer, and the MMI optical multiplexing / demultiplexing circuit becomes large. There is a problem that it takes shape.

## EQU1 ## L to C.w _g ² (C: constant) (1)

Similarly, in the conventional optical multiplexing / demultiplexing circuit in which the number of branches of the input / output waveguide is M × N, the input / output waveguide width w _i is limited due to the limitations on characteristics such as waveguide processing or crosstalk during device fabrication. , W ₀ , and the waveguide distance d cannot be reduced without limit. Therefore, as the number of branches M and N increases, MM
There is a problem that it is necessary to increase the I waveguide width w _g , and accordingly, the device length L also increases.

Further, also in the conventional semiconductor MMI device, the relationship between the waveguide spacing in the input / output waveguide and the MMI waveguide spacing is set by the relationship d ₁ = d ₂ = d ₃ = w _g / 3. Therefore, there is a problem that the degree of freedom in design is restricted in relation to the structure of the functional device which is connected or monolithically integrated, which hinders the optimal structure design of the entire device. An object of the present invention is to provide a low-loss optical multiplexing / demultiplexing circuit that can be configured in a small size.

[0011]

An optical multiplexing / demultiplexing circuit according to the present invention for solving the above-mentioned problems comprises an optical multiplexing / demultiplexing circuit constituted by a multimode interference waveguide having at least one or a plurality of optical input sections or optical output sections. In the circuit, the interval from the outermost optical input / output waveguide to the edge along the same direction as the waveguide direction of the multimode interference waveguide is configured to be smaller than the interval between the optical input / output waveguides. It is characterized by the following.

In the above-mentioned optical multiplexing / demultiplexing circuit, the edge of the optical input / output waveguide disposed at the outermost position is substantially matched with the edge of the multimode interference waveguide.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the principle and effects of an embodiment of the present invention will be described in detail with reference to the drawings.

[First Embodiment] A first embodiment according to the present invention will be described with reference to FIGS. FIGS. 1 and 2 show that LiNbO as a substrate is a strong derivative material.
₃ (hereinafter referred to as “LN”) is an embodiment of a 2 × 2 optical multiplexing / demultiplexing circuit according to the present invention. FIG. 1 is a perspective view of the MMI unit, FIG.
(B) is a sectional view taken along the line II-II in FIG. In these drawings, reference numeral 11 denotes an LN substrate, which is a clad portion of an optical waveguide,
Reference numeral 12 denotes a core layer formed by diffusion of an impurity such as titanium (Ti) or proton, 13 denotes an input waveguide, 14 denotes an output waveguide, 15 denotes an MMI region (effective core region), and 16 denotes an input / output waveguide. .

In this embodiment, in this case, the structure of the input / output waveguide 16 basically has the same structure as an optical device using a conventional LN substrate. For example, an optical function processing device unit connected to an input or output waveguide by monolithic integration or hybrid integration (for example, a refractive index modulation unit in which a modulation electrode using the EO effect is arranged, or a nonlinear function such as second harmonic generation) effect light wavelength controller, etc.) and using, takes a structure that gives it connected thereto size spot size of guided light as much, such as an optical fiber, the input side of the waveguide 16 width (w _i), the output side Waveguide 16 width (w
₀ ), the size of the core 12 thickness (t _g ) (diffusion length of impurities) and the diffusion concentration of those impurities are set.

Here, as shown in FIG. 2A, an input / output waveguide interval d ₂ (here, each d ₂ is a distance between the axes of the waveguides 16, 16) is connected. This is set in consideration of the structure of the optical function processing unit and the like. In the present embodiment, the effective core width (w _g ) = d ₁ + d ₂ + d ₃ . Here, d ₁ and d ₃ refers to the distance from the most arranged optical input and output waveguides on the outside, to the edge along the waveguide in the same direction as the direction of the multi-mode interference waveguide. Therefore, in the present embodiment, the dimensions d ₁ and d ₃ from the axis of the input / output waveguide 16 to the edge of the MMI region 15 are d ₁ and d ₃ <d, respectively.
It is set by the relationship of ₂ . Note that d ₁ and d ₃ may have the same size or different sizes.

[Second Embodiment] FIG. 3 shows another embodiment. FIG. 3 shows a structure in which the outer edge of the input / output waveguide 16 substantially coincides with the edge of the MMI region 15 in the embodiment of FIGS. 1 and 2, and d ₄ =＝ 0 ( d
₁ = d ₃ = showing an example of an optical multiplexing and demultiplexing circuit set in relation ~w _i / 2). That is, in the present embodiment, there is a relationship of effective core width (w _g ) = d ₂ + d ₄ + d ₄ .

[Principle and Effect of the Embodiment] FIGS. 4 to 7 are diagrams for explaining the principle and effect of the first and second embodiments, and show a 1.55 μm band 2 × 2 MMI optical system. The calculation results using the eigenmode expansion method for the demultiplexer are shown. Here, in the embodiment of FIG. 1, FIG. 2, and FIG. 3, the case where the optical waveguide is formed by diffusing Ti into the LN substrate by the ordinary thermal diffusion method is analyzed.
In order to simplify the calculation, the slab waveguide model analysis by the equivalent refractive index method is performed, assuming that the core layer and the cladding part have a uniform refractive index for the diffusion waveguide. The input / output waveguide width was calculated using w _i = 6 μm as an example.

[0019] Figure 4, the input and output waveguides distance as d ₂ = 33 .mu.m constant in FIGS. 1 and 2, when the d ₁ = d _3, of the 2 × 2 optical demultiplexing circuit for MMI waveguide width w _g The relationship between the length L and the excess loss obtained at that time is shown. In this figure, the magnitudes of w _g (μm) and L (mm) are displayed on a logarithmic scale.

In FIG. 4, when w _g = 99 μm, that is, when d ₁ = d ₂ = d ₃ = 33 μm, a wavy line (black circle)
It can be seen that the configuration of the device length (: L is about 10 mm) shown in FIG.
However, w _g deviates from 99 μm (d ₁ and d ₃ are 33
It becomes smaller or larger than μm. ), The loss increases sharply, the allowable range of the MMI waveguide width (effective core width) w _g is relatively small, and the degree of freedom in structural design is small. This is due to the fact that the lightwave mode excited in the MMI waveguide by the input light from the input waveguide arranged in the conventional example has a relatively low order and uses a small number of modes.

On the other hand, in FIG.
As the size of the L capable of obtaining a low-loss characteristics for the waveguide width (effective core width) w _g, at the indicated lengths solid line (open circles) other than that shown by the broken line, or the integral multiple thereof magnitude It can be seen that there is a device structure given by: When configuring a device with a length indicated by the solid line, even when narrower than 99μm size of the MMI waveguide width (effective core width) w _g, set an appropriate device length L to fit the width w _g Then, a relatively low loss characteristic shown by the solid line is obtained.

[0022] Moreover, as the MMI waveguide width w _g is reduced, the device length L is shortened, when less than about 60μm The MMI waveguide width w _g L can the following dimensions 10 mm, smaller than the conventional example Can be.

In this case, compared to the above embodiment,
In the MMI waveguide, a higher order and a large number of modes are excited, and a low loss characteristic is obtained by using these modes. The present invention basically uses this principle and effect.

When w _{g in} FIG. 2 is 39 μm,
1, 2 and 3, d ₁ = d ₃ = 3 μ
This corresponds to the case where m and d ₄ = 0. However, when the MMI waveguide width w _g is smaller than 39 μm, the input / output waveguide and M
The coupling loss with the MI waveguide becomes remarkable, and the loss due to radiation increases rapidly.

Further, as in the embodiment of FIG. 3, d ₄ is made slightly larger than 0 (in this case, the MMI waveguide width w _g = 4).
3 μm (d ₄ = 2 μm, d ₁ = d ₃ = 5 μm), and the loss can be extremely reduced as shown in FIG. This is due to the fact that the optical confinement effect of the input / output waveguide is relatively weaker than that of the MMI part when a normal device configuration is adopted in which the thickness of Ti is made constant throughout the device and thermal diffusion is performed. I do. For this reason, by configuring the edge of the MMI to be slightly outward with respect to the edge of the outermost input / output waveguide, the guided light of the input / output waveguide and the MMI Optical coupling with guided light can be performed efficiently.

FIG. 5 shows the MMI waveguide with respect to the input / output waveguide spacing d ₂ when w _i = 6 μm, d ₁ = d ₃ = 3 μm, and d ₄ = 0 in the first and second embodiments. The relationship between the length L and excess loss and crosstalk at that time is shown.
The magnitude of L has a relationship proportional to the square of d ₂ .

FIGS. 6 and 7 correspond to FIGS. 1, 2 and 3 respectively.
5 is an example showing the manufacturing tolerance of the MMI waveguide in the embodiment of the present invention. FIG. 6 shows that w _i = 6 μm, w
_{When g} = 39 μm and d ₂ = 33 μm, the MMI
The relationship between the excess loss and the crosstalk due to the deviation of the waveguide length L from the optimum length is shown. L is ± 1.
It can be seen that even if the distance is shifted by 5 mm, the increase in loss is suppressed to 1 dB or less, and the crosstalk does not deteriorate much.

FIG. 7 shows that when w _i = 6 μm, the length of the MMI waveguide L = 5.3 mm, and d ₂ = 33 μm, M
Shows the relationship between the excess loss and crosstalk caused by deviation Δw of MI waveguide w _g. Even if Δw is about ± 0.5 μm, the increase in loss can be suppressed to 1 dB or less, and a device can be manufactured without any problem if a waveguide is formed using ordinary photolithography technology.

Embodiments 3 and 4 FIG. 8 and FIG.
FIG. 4 is a top view of another embodiment of the present invention, showing a 1 × 2
The configuration of an optical demultiplexing circuit (Embodiment 3) or the configuration of a 1 × 3 optical demultiplexing circuit (Embodiment 4) is shown. FIG. 8 shows the relationship between the input / output waveguide structure w _i and the waveguide spacing d ₂ .
Using the principle of the present invention as in the above embodiment, d ₁ , d
_If the size such as ₃ is set to an appropriate size and the MMI waveguide width w _g is reduced, the waveguide length L can be reduced. However, as shown in FIG. 4, if w _g is made too narrow as necessary, the loss increases. Therefore, the structure may be determined in consideration of the increase in the loss.

As described above, in the present embodiment, 2 × 2 or 1
Although a × 2, 1 × 3 optical multiplexing / demultiplexing circuit configuration is shown, other than this, an nxm optical multiplexing / demultiplexing circuit (n, m: any integer), an optical wavelength filter, an optical mode filter, or a different wavelength The present invention can be applied to any optical device using the characteristics of the MMI waveguide, such as an optical device that multiplexes and demultiplexes the light of the above.

In this embodiment, the operating wavelength is 1.
In 55μm band, the LN substrate, the case of using the T _i diffused waveguide core layer, and input and output waveguide structure, intervals,
According to the operating wavelength, the spot size of the optical functional device connected by monolithic integration, or the spot size of the optical fiber to be optically coupled, the material, material, structure,
It is obvious that the effects of the present invention can be similarly obtained by appropriately setting the dimensions.

It is obvious that the present invention can be applied to a semiconductor MMI device. Further, the waveguide core layer may have a multilayer structure. The periphery of the MMI waveguide may be covered with a dielectric or a semiconductor material, or may have a ridge structure or a buried structure. Further, the present invention is applicable to devices using any optical waveguide material such as a ferroelectric material such as LiTaO ₃ or PLZT, or a semiconductor material, an inorganic material such as glass or quartz, or an organic material such as polyimide as a waveguide material. Applicable.

In this embodiment, the case where the input / output waveguide is arranged at the input / output section of the MMI waveguide has been described.
When other optical waveguide devices are directly coupled to the I input / output section at their respective waveguide end faces or connected via lenses, the MMI is adjusted to match the spot size of the connected guided light. The effects of the present invention can be obtained by appropriately setting the structure, material, and dimensions of.

[0034]

As described above, the optical multiplexing / demultiplexing circuit according to the present invention has a small optical multiplexing / demultiplexing circuit by adopting a configuration in which at least the edges of the input and output waveguides and the edges of the MMI waveguide are substantially matched. A circuit can be realized.

[Brief description of the drawings]

FIG. 1 is a perspective view of an optical multiplexing / demultiplexing circuit according to a first embodiment of the present invention.

FIG. 2 is a top view (a) and a cross-sectional view (b) of the optical multiplexing / demultiplexing circuit according to the first embodiment of the present invention.

FIG. 3 is a perspective view of an optical multiplexing / demultiplexing circuit according to a second embodiment of the present invention.

FIG. 4 is a diagram for explaining the principle and effect of the present invention.

FIG. 5 is a diagram for explaining the principle and effect of the present invention.

FIG. 6 is a diagram for explaining the principle and effect of the present invention.

FIG. 7 is a diagram for explaining the principle and effect of the present invention.

FIG. 8 is a top view of an optical multiplexing / demultiplexing circuit according to a third embodiment of the present invention.

FIG. 9 is a top view of an optical multiplexing / demultiplexing circuit according to a fourth embodiment of the present invention.

FIG. 10 is a perspective view of a conventional optical multiplexing / demultiplexing circuit.

FIG. 11 is a top view of a conventional optical multi / demultiplexing circuit.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 11 Substrate 12 Waveguide core layer 13 Input waveguide 14 Output waveguide 15 MMI area (effective core) 16 Input / output waveguide 17 Cladding layer

Continuation of front page (72) Inventor Yasuo Shibata Nippon Telegraph and Telephone Corporation, 3-19-2 Nishishinjuku, Shinjuku-ku, Tokyo

Claims (2)

[Claims] 1. An optical multiplexing / demultiplexing circuit comprising a multi-mode interference waveguide having at least one or a plurality of optical input sections or optical output sections, wherein an optical input / output waveguide arranged at the outermost side is used for multi-mode interference. An optical multiplexing / demultiplexing circuit, wherein an interval to an edge of the interference waveguide along the same direction as the waveguide direction is narrower than an interval between the optical input / output waveguides. 2. The optical multiplexing / demultiplexing circuit according to claim 1, wherein an edge of the outermost optical input / output waveguide is made substantially coincident with an edge of the multimode interference waveguide. Optical multiplexing / demultiplexing circuit.