JPH02179621A - Optical matrix switch - Google Patents

Abstract

PURPOSE:To improve the controllability by providing a cut part, where a part of a waveguide is removed until the surface of an optical guide layer is exposed, between each first directional coupler of an input waveguide and each second directional coupler of an output waveguide. CONSTITUTION:Three input waveguides 17 constituted by connecting first directional couplers 15 having first waveguides 11 and second waveguides 13 in three stages and three output waveguides 27 constituted by connecting second directional couplers 25 having third waveguides 21 and fourth waveguides 23 in three stages are provided, and first waveguides 11 and fourth waveguides 23 are connected with total reflection corners 31 between them and second waveguides 13 and third waveguides 21 intersect to arrange input and output waveguides 17 and 27 in a matrix. Cut parts 61 are provided where parts of waveguides are removed in such degree to expose surfaces of optical guide layers that directional couplers 15 and 25 are electrically separated but optical waveguide is secured. Consequently, directional couplers are electrically separated by cut parts but optical waveguide is secured, and the light loss does not matter.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to an optical matrix switch in an optical exchanger.

(Prior Art) Optical matrix switches are important basic elements of optical exchanges, and for this reason, research on them has been energetically conducted for a long time.

FIG.
552t) is a plan view schematically showing the optical matrix switch proposed in No. 1.

This optical matrix switch has 178 m input waveguides (in this example, 178 m in number) which are configured by connecting first directional couplers 15 having a first waveguide 11 and a second waveguide 13 in n stages (three stages in this example). a second directional coupler 25t having a third waveguide 21 and a fourth waveguide 23; and an output waveguide 27'a configured by connecting m stages (three stages in this example).
n pieces (three in this example), the first waveguide 11 and the fourth waveguide 23 are connected via a total reflection corner 31, and the second waveguide 13 and the third waveguide 21 are provided.
The input and output waveguides 17 and 27 were arranged in a matrix by crossing each other. According to this optical matrix switch, when using any of the many propagation paths for light formed between the input boats 17a, 17b, 17c and the output boats 27a, 27b, 27c, the optical signal passes through a total reflection corner. Since the structure is such that it only needs to pass through once, the loss during propagation of the optical signal can be reduced compared to the conventional structure even if each component is made of known components.

By the way, each waveguide of the optical matrix switch shown in FIG. 3 is made of a compound semiconductor material such as GaAs/A
If an attempt is made to construct a strip-loaded waveguide using a uGaAs-based material, the structure will be as described below, for example. 4 and 5 are diagrams for explaining the same, and FIG. 4 is a cross-sectional view at a depth equivalent to the line I-I in FIG. 3, and FIG. 5 is an enlarged view of the vicinity of the total reflection corner 31. FIG. However, in FIG. 4, hatching indicating the cross section is omitted to avoid complicating the drawing.

In Figure 4, 41 is the first conductivity type (n type in this example)
This is a GaAs substrate. On this n-type GaAs substrate 41, an n-type/1lGaAs lower rad layer 43 and a 1-type GaAs substrate 41 are formed.
An As light guide layer 45 is provided in this order, and a p-type AiLGaAs upper rat layer 47 and a p-type A GaAs cap layer 49 is provided in this order. Further, an n-side electrode 51 is formed on the region of the p-type GaAs cap layer 49 corresponding to the directional coupler.
An n-side electrode 53 is provided on the lower surface of the As substrate 41.

In this structure, light is transmitted through two laminates 55a and 55b (hereinafter referred to as first ribs 55a and second ribs 55b), which are composed of the upper cladding layer 47, the cap layer 49, and the n-side electrode 51. is confined within the lower light guide layer portion of the.

Also, from now on, the total reflection corner 31 of the matrix switch
For example, as shown in FIG.
1. p-type AQGaAs cladding layer 49, i-type GaAs optical guide layer 45, and n-type A9GaAs lower cladding layer 4
3 can be constructed by removing a portion of each of the substrates 41 perpendicularly to the main surface of the substrate 41.

When operating the optical matrix switch as explained using FIGS. 4 and 5, the n-side electrode 51 of the first rib 55a of each directional coupler and the n between the side electrode 53 and the n-side electrode 51 of the second rib 55b of each directional coupler and the first rib 5
A voltage VSwlF is applied between the n-side electrode 51 of 5a and
r will be applied.

(Problems to be Solved by the Invention) However, the optical matrix switch described using FIG. 3 has a strip-loaded waveguide structure using a compound semiconductor material as described using FIGS. 4 and 5. , each directional coupler will be connected to each other by the cap layer 49 and the upper cladding layer 47 at the ribs 55a, 55b, therefore, from now on, the voltage will be applied to the matrix switch as shown in FIG. If the voltage is applied and operated, the same voltage will be applied to all the directional couplers, which poses a problem in that each directional coupler cannot be individually controlled and operated.

The present invention has been made in view of the above, and an object of the present invention is to provide an optical matrix switch that is made of a compound semiconductor material and uses a strip-loaded waveguide, and that has excellent controllability. It is about offering sacrifices.

(Means for Solving the Problem) In order to achieve this object, according to the present invention, a first directional coupler having a first waveguide and a second waveguide is connected in zero stages. In addition to having m input waveguides and n output waveguides configured by connecting m stages of second directional couplers each having a third waveguide and a fourth waveguide, An optical matrix switch in which a waveguide and the aforementioned fourth waveguide are connected through a total reflection corner, and the aforementioned second waveguide and the aforementioned third waveguide are crossed to form a matrix of the aforementioned input/output waveguides. A lower cladding layer and a light guide layer in which each of the above-mentioned image waveguides to the fourth waveguide are sequentially provided on the substrate, and a region of the light guide layer that becomes the first to fourth waveguides. In an optical matrix switch made of a compound semiconductor and configured with a strip-loaded waveguide having an upper cladding layer provided thereon, between each first directional coupler of each of the m input waveguides and the n input waveguides, A portion of the waveguide is connected to the surface of the light guide layer to the extent that each directional coupler is electrically separated from the output waveguides of the second directional couplers and the optical waveguide is secured. It is characterized by having a cut-out portion, which is removed until it is exposed.

Furthermore, in carrying out the present invention, one of the electrodes provided in the two waveguides of each of the first directional couplers and the second directional couplers electrically separated by the above-mentioned cutting portions. It is preferable to have a configuration including an inter-electrode connection portion that connects the electrodes that are used as a common electrode.

(Function) According to this configuration, the cutout part has a structure in which there is no upper layer including the upper cladding layer (for example, the cap layer and the p-side electrode), so each directional coupler of the optical matrix switch Each of them is electrically charged by minutes Mc5. However, the discontinuity of the upper cladding layer at the cutout is small enough to ensure optical waveguide, and the light guide layer remains even at the cutout, so optical loss is practically no problem. Therefore, as described with reference to FIG. '&
Since the cross state or the bar state is achieved depending on whether the voltage is applied or not, each directional coupler can be operated individually.

Furthermore, since the electrodes that are at the common potential of each directional coupler are sequentially connected by the inter-electrode connection portion, there is no need to provide common electrode wiring for each directional coupler.

(Embodiments) Hereinafter, embodiments of the optical matrix switch of the present invention will be described with reference to the drawings. However, the drawings used in the following explanation are only schematic illustrations to the extent that the present invention can be understood, and therefore, the dimensions, shapes, and arrangements of each component may vary. It should be understood that the ll1ffi coefficient and the dimensional ratio between each component are schematic, and the present invention is not limited to the illustrated example.In addition, in the following example, the number of input waveguides is rr + 7a3 and the output The number of waveguides is n7a3, 3
An example in which the present invention is applied to a ×3 optical matrix switch will be explained. Further, as the material for manufacturing the optical matrix switch, an AQGaAs/GaAs-based compound semiconductor was used as in the past.

1 and 2 are diagrams used to explain the optical matrix switch of the embodiment. FIG. 1 is a plan view schematically showing the overall configuration, and FIG. 2 is a diagram shown by P in FIG. 1. It is a perspective view which expanded and showed a part. In each figure, components similar to conventional components are designated by the same reference numerals.

Further, in order to avoid complicating the drawings, some numbering for similar components in the drawings is omitted.

The optical matrix switch of this embodiment is constructed by connecting three stages of first directional couplers 15 having a first waveguide 11 and a second waveguide 13, as shown in the plan view of FIG. It includes three input waveguides 17 and three output waveguides 27 configured by connecting three stages of second directional couplers 25 having a third waveguide 21 and a fourth waveguide 23. , the first waveguide 11 and the fourth waveguide 23 are connected through the total reflection corner 31, and the third waveguide 13 and the third waveguide 23 are crossed to form a matrix of input and output waveguides +7.27. be. As already explained using FIG. 4, the n-type AlGaAs lower cladding layer 43 and the i-type GaAs lower cladding layer 43 and i-type GaAs are sequentially provided on the n-type GaAs substrate 41 to form each of the first to fourth waveguides. A light guide layer 45 and this light guide layer 45
The strip-loaded waveguide is composed of a p-type GaAs upper cladding layer 47 and a p-type GaAs cap layer 49, which are sequentially provided on the first to fourth waveguide regions.

Further, an n-side electrode 51 is provided on the p-type GaAs cap layer 49 of the two waveguides of each directional coupler, and an n-side electrode 53 is provided on the lower surface of the n-type GaAs substrate 41. .

Furthermore, in the future matrix switch, as shown in FIG. 1 and FIG. A cutout portion 61 is provided between each coupler, in which a portion of the waveguide is removed until the surface of the optical guide layer is exposed to the extent that each directional coupler is electrically isolated and the optical waveguide is ensured. It is a provision. On the i-type GaAs light guide layer 45 of each waveguide in this embodiment, there is a p-type J (lGaAs upper cladding layer 47, a p-type GaAs cap 1949, and an n-side electrode 5).
1 is a stacked layer, the cutout portion 61 includes the n-side electrode 51 of the waveguide, the p-type GaAs cap layer 49, and the p-type AuGa
It is formed by removing a portion of each of the As upper rad layers 47 until the surface of the light guide layer 45 is exposed.

Here, the cutout portion 61 is large enough to electrically isolate each directional coupler and ensure optical waveguide, and the size of the cutout portion is l.
(see FIG. 2) is the width W of the waveguide (see FIG. 2), that is, the rib 5 shown in FIG.
The specific value of the 0 dimension β, which is obtained by making the value smaller than the widths of the optical matrix switches 5a and 55b, is determined depending on the design of the optical matrix switch.

Furthermore, in the optical matrix switch of this embodiment,
The substrate 41 of the n-side electrodes 51 provided on the two waveguides of each of the first directional couplers 15 and the second directional couplers 25 that are electrically separated by the cutout 61 n-side electrode 51 connected to and used as a common electrode
(See FIG. 4) Specifically, with reference to FIG. 2, the n-side electrode 51X of the third waveguide 21 of the second directional coupler 25 and the first directional coupler 1
5 and the n-side electrode 51y of the second waveguide 13 in FIGS.
It is connected by 3. And this inter-electrode connection part 6
3 is finally connected to n by the wiring electrode 65 (see Figure 1).
It is connected to the side electrode and serves as a common electrode. Therefore, the inter-electrode connection portion 63 and the p of the common electrode of each directional coupler
Since the electrodes that are at the common potential of each directional coupler are successively connected to each other by the side electrodes, it is no longer necessary to provide common electrode wiring for each directional coupler.

Note that the present invention is not limited to the above-mentioned embodiments, and various modifications as described below can be made.

For example, in the optical matrix switch of the embodiment, the waveguide is described as having a cap layer 49. However, a strip-loaded waveguide constructed by removing the cap layer 49 can also provide the same effects as in the embodiment. In this case, the cutout portion 61 is formed by removing a portion of the upper cladding layer 47 until the surface of the light guide layer 45 is exposed.

Furthermore, although the above embodiment describes an example of a 3x3 optical matrix switch, this is just an example, and the number of input/output waveguides is m, n! It is clear that the present invention can be applied even when the numbers of input and output waveguides are different, or when the number of input and output waveguides is changed to another number while keeping the same number.

Further, in the above-mentioned embodiment, an example using an n-type GaAs substrate is explained, but it is of course possible to use an IP-type substrate and each semiconductor layer to have a conductivity type opposite to that in the embodiment. The layers are not limited to i-type, but may be p-type or n-type.In general, the constituent materials of the optical matrix switch are
Other materials such as InGaAsP/InP may also be used.

(Effects of the Invention) As is clear from the above description, according to the optical matrix switch of the present invention, each directional coupler is electrically isolated by the cutting portion. In addition, the discontinuous portion of the upper cladding layer at the cutout is small enough to ensure optical waveguide, and the light guide layer remains even at the cutout, so optical loss is practically no problem. It won't happen. Therefore, it is made of compound semiconductor material and is strip-loaded.

Even in an optical matrix switch using a waveguide, each directional coupler can be individually driven and controlled.

In addition, since the electrodes that are at the common potential of each directional coupler are sequentially connected by the inter-electrode connection part and the p-side electrode on one waveguide of each directional coupler, the individual directional This provides the advantage that there is no need to provide common electrode wiring for each coupler.

[Brief explanation of the drawing]

Fig. 1 is a plan view for explaining the optical matrix switch of the embodiment, Fig. 2 is a perspective view showing an enlarged part of the optical matrix switch of the embodiment, and Fig. 3 is a conventional optical matrix switch.・A plan view for explaining the invention, Figure 4 is a diagram C for explaining the conventional and present invention
A partial cross-sectional view of the optical matrix switch shown in FIGS. 1 and 3 taken at a line equivalent to I in the same figure, and FIG. 5 is a perspective view for explaining the total internal reflection corner. 11...First waveguide, 13...Second waveguide 1
5... First directional coupler 17... Input waveguides 17a, I7b, I7c... Input boat 21...
Third waveguide 23-...Fourth waveguide 25...Second directional coupler 27...Output waveguides 27a, 27b, 27c...Output boat 3'1...
- Total reflection corner 4 +-... Substrate (n-type GaAs substrate) 43... Lower cladding layer (n-type A'1 GaAs layer) 45... Light guide layer (i-type GaAs layer) 47-... Upper side Cladding layer (p
Type A1GaAs layer) 49... Cap layer (p-type GaAs layer) 49... Cap layer (p-type GaAs layer)
s layer) 51.55x, 55y...p side electrode, 53...
・N-side electrode 55a...first rib, 55b--
・Second 1 knob 61...Waveguide cutting portion, 63...
- Inter-electrode connection part 65...wiring electrode.

Claims (2)

[Claims] (1) m input waveguides configured by connecting n stages of first directional couplers each having a first waveguide and a second waveguide, and a third waveguide and a fourth waveguide. The first waveguide and the fourth waveguide are connected through a total reflection corner, and the second waveguide is configured by connecting m stages of second directional couplers. An optical matrix switch in which each of the input and output waveguides is formed into a matrix by crossing a waveguide and the third waveguide, the lower side having each of the first to fourth waveguides sequentially provided on a substrate. An optical device made of a compound semiconductor, comprising a strip-loaded waveguide having a cladding layer, a light guide layer, and an upper cladding layer provided on the regions of the light guide layer that become the first to fourth waveguides. In the matrix switch, each directional coupler is electrically connected between each first directional coupler of each of the m input waveguides and between each second directional coupler of each of the n output waveguides. An optical matrix switch characterized in that a cutout portion is provided in which a portion of the waveguide is removed to the extent that the optical waveguide can be separated and the optical waveguide can be secured until the surface of the optical guide layer is exposed. (2) In the optical matrix switch according to claim 1, the electrodes provided in the two waveguides of each of the electrically isolated first directional couplers and second directional couplers each have a An optical matrix switch characterized by having an inter-electrode connection section that connects between electrodes that are considered as common electrodes.