JPH06194536A - Optical coupling device - Google Patents

Abstract

PURPOSE:To provide an optical coupling device whose productivity is excellent and capable of accomplishing optical coupling between two different optical functional elements, especially between the optical functional elements which are obtained by integrating plural devices at a small loss. CONSTITUTION:As to the optical coupling device constituted of at least a semiconductor substrate 201 and optical waveguide layers 202 and 203 which are formed on the semiconductor substrate 201 and whose size and refractive index are gradually varied in a light transmission direction, the optical waveguide layers 202 and 203 are branched near a light incident end, or a light emitting end and arranged. Thus, the optical coupling is accomplished at a small loss.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an optical coupling device for converting the spot size of a light wave propagating through an optical waveguide of an optical functional device to another optical functional device with low loss.

[0002]

2. Description of the Related Art In the case of optically coupling a semiconductor laser diode (LD) and a single mode fiber, if the LD element end face and the fiber are directly butt-coupled with each other, the optical waveguide light wave spot sizes are different from each other. Therefore, the coupling loss of the direct butting portion becomes a problem. Usually, the LD light wave spot size (mode radius: W) is 1
The spot size of the fiber is about 5 μm.
Therefore, the coupling loss is about 10 dB. Therefore, a method of reducing the coupling loss by converting the spot size with a lens is generally used.

FIG. 5 shows an example of a conventional configuration in the case of optically coupling an optical fiber having a plurality of laser diodes (LDs) and an array fiber with a single lens. In FIG. 5, 501 is a semiconductor substrate, 502 is an active region (optical waveguide portion), 503 is a fiber, 504 is a V-groove array for fixing the fibers at regular intervals, and 505 is a lens. In such a configuration, as the integration scale of LDs increases, the coupling loss increases due to the influence of lens aberration and the like, and thus the number of LDs that can be integrated on one semiconductor substrate is limited.

There is a method of optically coupling the LD and the fiber with low loss by using an optical coupling device for converting the spot size of light by a tapered optical waveguide as shown in FIG. 6 instead of the lens. FIG. 6 (a) is a top view of a conventional optical coupling device, FIG. 6 (b) is a sectional view, and FIG. 7 is a diagram for explaining the operation principle. That is, as can be seen from FIG. 7, the refractive index difference Δn of the core layer 602 of the optical waveguide is
[= (N1-n2) / n1, n1: clad layers 601, 603
, N2: refractive index of the core layer 602] is fixed to a constant value, and when the thickness t and width w of the core layer 602 are gradually increased from 0, guided light (fundamental mode light) is obtained. The spot size W is gradually decreased from an infinite size, has a minimum value, and then increases again. Here, if t and w become too large, a multimode waveguide is formed, and the loss due to the higher-order mode conversion becomes large, so that the size of this region is not usually used. Utilizing this relationship, in designing the sizes t and w of the core layer 602 of the optical coupling device, L is set on the light incident end side (coupling side to the LD).
Dimensions w i and t i (= several 100 nm to several μ) that give a spot size Wi approximately equal to the spot size of D light (about 1 μm)
m), on the light emitting end side, the dimensions to and wo (= several tens to several hundreds of nm) are set so as to give a spot size Wo of the same size as the fiber spot size (about 5 μm) (specific design example). As for
See C-201, 1991.). Further, the length L of the region where the size of the core layer 602 is tapered is set to a length of -100 μm to several mm or more in order to reduce the loss due to radiation.

[0005]

In such a structure, if the dimensions to and wo on the light emitting end side are made small and the spot size Wo is made large, the light confinement becomes weak, so that the light intensity distribution of the waveguide is reduced. Has an exponential shape. On the other hand, since the guided light intensity distribution of the optical fiber has a substantially Gaussian distribution shape, coupling loss due to shape mismatch occurs in principle. Also, the optimum t for obtaining low coupling loss
Since the allowable dimensional variation of o and wo is relatively small, there is a manufacturing difficulty when using the normal method.

On the other hand, the spot size can be enlarged by gradually reducing the magnitude of Δn along the light traveling direction and gradually increasing t and w accordingly. However, in this case, it is difficult to change the refractive index into a taper shape and to make it sufficiently small at the light emitting end, and there is a drawback that low loss characteristics cannot be realized. It is an object of the present invention to provide a manufacturable optical coupling device capable of performing optical coupling with low loss between two different optical functional elements, particularly an optical functional element integrating a plurality of devices. .

[0007]

In order to solve the above-mentioned problems, the present invention provides a semiconductor substrate and a semiconductor substrate formed on the semiconductor substrate, wherein the size or the refractive index is gradually changed along the light propagation direction. In the optical coupling device including at least the optical waveguide layer, the plurality of optical waveguide layers are arranged in the vicinity of the light incident end or the light emitting end. According to a second aspect, in the first aspect, the width of at least a plurality of the optical waveguide layers or the intervals between the optical waveguide layers are spatially distributed. According to a third aspect of the present invention, in the first or second aspect, the thickness of the optical waveguide layer is constant in the light traveling direction.

[0008]

In the present invention, a low-loss optical coupling is realized by arranging a plurality of optical waveguide layers in a branched manner.

[0009]

Embodiments of the present invention will now be described in detail with reference to the drawings. 1 and 2 are views showing an embodiment of an optical coupling device according to the present invention, and FIG. 3 is a view for explaining the principle of the present invention.
FIG. 4 shows another embodiment of the present invention.

FIGS. 1 (a) and 1 (b) show an embodiment of an optical coupling device according to the present invention, in which the optical coupling device of the present invention is inserted between an array LD element and a fiber to reduce loss. It is a block diagram in the case of taking optical coupling in. 1A is a top view and FIG. 1B is a cross-sectional view. 101 is a semiconductor substrate of an optical coupling device according to the present invention, 102 is a core layer, 103 is a cladding layer, and 104 is an antireflection film. In addition, 105 is an optical fiber and 106 is a V groove array. In this configuration, the size of the light wave spot of the LD is gradually converted by the core layer 102 of the optical coupling device, and the size is converted to an appropriate size at the light emitting portion.

FIG. 2A is a top view of one embodiment of the present invention for optical coupling between one LD element and one fiber, and FIG. 2B is a sectional view of the central portion. 201 is a semiconductor substrate made of, for example, InP (refractive index n1), 202 is a spot size conversion waveguide core layer (for example, InGaAsP, refractive index n2), and 203 is a semiconductor cladding layer (for example, InP, refractive index n3). is there. The magnitude of these refractive indices is n1, n
There is a relationship of 3 <n2. 204 is the incident light from the LD, 205
Is the expanded outgoing light on the single mode optical fiber side.
The optical waveguide portion in the region shown in the figure has a normal waveguide structure having a spot size substantially the same as the light wave spot size of the LD. In the region, the core layer 202 is divided into a plurality of branches, and the structure and the dimension w of the waveguide are designed so that the spot size and shape are expanded to obtain the low loss coupling characteristics when the optical coupling is performed directly with the fiber. , T and the refractive index n are set. In the region, the effective waveguide width w is formed in a taper shape so that the spot size is gradually converted, and the taper length is L.

The principle of the present invention will be described below. Figure 6 (a),
In the conventional spot size conversion taper waveguide shown in (b), the mode field shape is exponential because the size of the core is extremely small in order to enlarge the spot size. However, if the refractive index difference Δn is reduced, as can be seen from FIG. 7, the spot size can be increased without reducing w and t. At this time, Δ
By setting w and t to appropriate values as n is made small, the confinement coefficient of the optical waveguide is kept large, so that the mode field shape becomes Gaussian like an ordinary single mode fiber. . In the present invention, as shown in FIG. 2, in the region, the total waveguide width w3 is set to the same size as the core diameter of the fiber, and the width w31 of the core layer 202 and the interval w32 between the core layers are set to appropriate dimensions. By doing so, the effective Δn of the optical waveguide is reduced without changing the size of n2. As a result, the spot size in the x direction of the surface of the semiconductor substrate 201 is enlarged, so that the coupling loss with the fiber can be reduced.

FIG. 3 shows the approximate value of the relationship between the waveguide structure in the region and the mode coupling efficiency ηx between the waveguide and the fiber in the x direction. Here, the wavelength of light is λ = 1.55 μ
m, the refractive index of the core and clad layers is n2 = 3.3, n1 =
With n3 = 3.17 constant, the mode radius of the fiber is wf = 4.5
μm, and the number of branches of the core layer 202 is 5 and w31 + w32
= 2 μm constant. As can be seen from this figure, the core width w
It can be seen that the fiber coupling loss can be made extremely small (0.1 dB or less) when 31 (dx) and the core thickness t3 (= dy) are set to predetermined values. At this time, the direction (y
The mode field shape in the (direction) is an exponential function, and the coupling loss in the y direction is as large as that of the conventional example (0.2 dB). Therefore, the total coupling loss with the fiber is
It can be less than 0.3dB. Further, by reducing the core dimensions w31 and w32, increasing the number of branches, and setting the overall width w3 to an appropriate size to match the mode field shape of the fiber, the coupling loss can be further reduced. These structures can be accurately designed by analysis such as the finite element method. In the present invention, the same characteristics can be realized by removing at least the region or the region waveguide and forming only the region waveguide.

When manufacturing the optical waveguide according to the present invention, the sizes of n1, n2 and n3 can be arbitrarily set by selecting a semiconductor material. For example, InP
Is used, n = for light in the wavelength λ = 1.55 μm band
3.166. Further, the refractive index of InGaAsP can be set to any value from about 3.2 to 3.5 depending on its composition. Further, a multiple quantum well layer is used as the core layer, and the refractive index can be arbitrarily set by selecting the material and thickness of the well layer and the barrier layer. In addition, the refractive index n2 of the core layer 202 is set by using, for example, a selective growth mask, an epitaxial selective growth technique, or a photolithography technique.
It is also possible to set and manufacture the waveguide dimensions w and t in a spatially tapered manner.

In FIG. 2, in order to realize the optimum core size for obtaining the best coupling loss, the core thickness t3 can be accurately manufactured by, for example, a crystal epitaxial growth technique. Since the core layer width w31 has a relatively large allowable fluctuation amount, it can be seen that the core layer width w31 can be easily manufactured by an ordinary photolithography technique. Further, the core layer thickness t is kept constant (t1 = t3) over the entire area,
It is also possible to have a structure in which only the core layer width w is converted, and in this case, the manufacturing process is simplified.

FIG. 4 is a top view of another embodiment of the present invention. By giving a distribution to the size of the core width w31 of the region, the coupling loss in which the mode field shape is more matched with that of the fiber is reduced. That is the case. Also, it is obvious that the same effect can be obtained by giving the size of the core interval w32 a spatial distribution.

As the structure of the tapered portion of the region, the tapered shape may be linear as shown in FIG. 2, or the curved shape such as an exponential function shape or a parabolic shape may be used to radiate when L has the same length. The loss can be made relatively small. Also,
The branching method of the core layer may be a branched shape as shown in FIG. 2 or a combined branched structure as shown in FIG.

In the above, the total waveguide width w3 in the region is as large as the core diameter of the fiber (usually w1 <w3).
However, by setting the size of w3 to be approximately the same as the waveguide width w1 of the region and setting w31 and w32 to appropriate sizes, the spot size of this region can be enlarged. However, in this case, although the mode field shape is an exponential function shape, it is obvious that the fiber optical coupling with low loss can be realized as in the conventional example.

In the above description, the case where the spot size conversion waveguide layer is formed on the InP substrate has been described. However, for an optical waveguide made of another semiconductor material such as GaAs or glass material such as SiO ₂ . It is obvious that the same effect can be obtained with. Also, in the above, the material of the clad layer and the material of the clad part of the optical waveguide are the same,
Although the case where the clad is composed of a single core layer has been described, the same principle as that of the present invention can be applied to a case where these clads are combined with different materials or a case where the clad is composed of a plurality of core layers. In the above, the case of connecting an optical fiber was explained, but in addition to this, the optical intensity of these waveguides is also applied to the connection parts with other optical waveguide components such as other semiconductor optical waveguide components or glass waveguide components. It is obvious that the characteristics of low coupling loss can be realized by appropriately setting the structure of the optical coupling device waveguide according to the present invention so as to match the distribution.

Since the present optical coupling device is made of a semiconductor material, for example, the present coupling device is monolithically integrated on the same substrate at the light incident end of an optical functional element such as a semiconductor laser, an LD amplifier or an optical switch. It is also possible to realize an optical device. In this case, when the optical functional element waveguide is formed on the semiconductor substrate, the optical coupling waveguide is formed at the same time, or after the optical functional element portion is formed, the optical waveguides are directly abutted to each other. A coupling tapered waveguide may be formed.

[0021]

As described above, according to the present invention, the size or the refractive index of the semiconductor layer which becomes the core portion of the optical waveguide is formed in a taper shape, and in the vicinity of the end face connected to another optical functional element, By arranging the core semiconductor layer in a plurality of branches, low-loss optical coupling can be realized.

[Brief description of drawings]

FIG. 1 is a top view and a cross-sectional view showing one configuration example of an optical coupling device according to the present invention.

FIG. 2 is a top view and a cross-sectional view showing one configuration example of the optical coupling device according to the present invention.

FIG. 3 is a diagram showing the operating principle of the present invention, showing the relationship between the waveguide structure and fiber coupling loss.

FIG. 4 is a diagram showing another configuration example of the optical coupling device according to the present invention.

FIG. 5 is a diagram showing a conventional optical coupling method.

FIG. 6 is a diagram showing a structure of a conventional optical coupling device.

7 is an explanatory diagram of an operation principle of the optical coupling device of FIG.

[Explanation of symbols]

101,201,401 ... Semiconductor substrate, 102,202,402 ... Core layer,
103,203,403… cladding layer, 204,404… incident wave, 205,40
5… Outgoing light.

 ─────────────────────────────────────────────────── ─── Continued Front Page (72) Inventor Masahiro Ikeda 1-1-6 Uchisaiwaicho, Chiyoda-ku, Tokyo Nihon Telegraph and Telephone Corporation

Claims (3)

[Claims] 1. An optical coupling device comprising at least a semiconductor substrate and an optical waveguide layer formed on the semiconductor substrate and having a size or a refractive index gradually changed along a light propagation direction. Alternatively, an optical coupling device characterized in that a plurality of the optical waveguide layers are branched and arranged in the vicinity of the light emitting end. 2. The optical coupling device according to claim 1, wherein widths of the plurality of optical waveguide layers arranged or intervals of the waveguide layers are spatially distributed. 3. The optical coupling device according to claim 1, wherein the optical waveguide layer has a constant thickness in the light traveling direction.