JPH03252619A - Light modulator - Google Patents

Abstract

PURPOSE:To realize an optimum light confining condition in each part and to shorten the whole length of an element by changing the impurity composition of an optical waveguide at an electrode part and a combined part. CONSTITUTION:The optical waveguide 12 consists of an incident optical waveguide 12a, a branched optical waveguide 12b, arm parts 12c, a combined optical waveguide 12d, and a projecting optical waveguide 12e. Titanium Ti to be an impurity for increasing a refractive index is diffused in air atmosphere and MgO to be an impurity for reducing the refractive index is diffused in oxygen atmosphere existing in the waveguide 12e following the combined part. Then a SiO2 film 14 is formed as a buffer layer and a control electrode 13 is formed on the surface of the film 14. Consequently, the light confinement strength values of the electrode part and the combined part of an MZ modulator can be respectively independent and optimum values, the whole length of the element can be shortened and the compact light modulator can be obtained.

Description

[Detailed description of the invention] [Industrial application field] The present invention relates to an optical modulator, and particularly to an optical waveguide type optical modulator.

[Conventional technology]

Modulators are important key devices in optical communications. In particular, waveguide external modulators have the excellent feature of being capable of high-speed modulation without chirping. Matsuhatsu Ender uses optical branching interference as a method of this waveguide external modulator.
type modulators (hereinafter referred to as MZ modulators) are known. T
The operating principle of an MZ modulator using an i-diffusion LiNbO3 waveguide will be explained as an example.

FIG. 4 is a perspective view showing this MZ modulator.

A single mode optical waveguide 12 is formed by diffusing Ti on a LiNb09 substrate 11.

This optical waveguide 12 is divided into two arms by seven branches, and after passing through the control electrode 13, the two arms are merged again by seven branches.

Further, an Si 02 film 14 is formed between the optical waveguide 12 and the control electrode 13 as a buffer layer to prevent absorption of light by the electrode.

FIG. 5 is a plan view of the MZ modulator. When no voltage is applied between the control electrodes 13 as shown in FIG. 5(A),
The zero-order mode light 61 incident on the optical waveguide 12 is divided into two beams 62 and 63 that have the same phase, that is, in-phase, at the Y branch part.
The light beams are branched and merged while remaining in the same phase to become output light 64 in the zero-order mode.

When a voltage equivalent to a half-wavelength voltage is applied between the control electrodes 13 as shown in FIG. 5(B), the two branches of in-phase light have phases reversed after passing through the electrode part due to the electro-optic effect of LiNbO3. The light beams 65 and 66 have opposite phases, and a first-order mode light 67 is generated at the merged portion. In the single mode optical waveguide 12, the first mode light 67 is cut off and cannot be guided, so it is radiated into the substrate, so no output light appears. Therefore, by applying a modulation signal voltage between the control electrodes 13, light can be modulated.

The operating principle of the MZ modulator was explained above using a Ti-diffused LiNbo3 optical waveguide as an example, but the same applies to optical waveguides with electro-optic effects such as semiconductors, and furthermore, it can be guided by means of stress application, current injection, etc. If the effective optical path length between the two arms can be changed by changing the waveguide refractive index, MZ
A modulator can be configured.

[Problem to be solved by the invention]

In order to miniaturize the waveguide external modulator, the modulation efficiency is increased by increasing the overlap between the electric field distribution by the control electrode and the field distribution of the light propagating through the optical waveguide.
An effective method is to shorten the electrode length. In other words, it is important to keep the optical field distribution of the guided light small and to confine it so that it is distributed near the control electrode. However, in this case, the confinement of light becomes strong even at the junction of the branches, so 11
The distance required for radiation of the next mode becomes relatively long. Therefore, there is a problem that shortening the electrode length due to improved modulation efficiency does not necessarily lead to shortening the overall length of the device.

[Means to solve the problem]

In the present invention, an optical waveguide formed by diffusing an impurity having an effect of increasing the refractive index on a substrate is composed of an input optical waveguide, a branch optical waveguide, an arm part, a merging optical waveguide, and an output optical waveguide, and the arm part In the optical modulator including a control electrode, the structure is characterized in that another impurity having an effect of reducing the refractive index is introduced into at least the output optical waveguide.

[Effect]

In the present invention, the optical waveguide of the electrode part of the MZ modulator is strongly confined near the electrode to improve the modulation efficiency and shorten the electrode length, and furthermore, the optical waveguide of the converging part is provided with another layer having the effect of reducing the refractive index. By introducing impurities, the confinement of light is partially weakened and the distance required for emission of first-order mode light is shortened. In other words, by changing the impurity composition of the optical waveguide at the electrode section and the confluence section, the optimal light confinement conditions are achieved at each section, thereby shortening the overall length of the device.

〔Example〕

Next, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a perspective view showing a first embodiment of the present invention.

An optical waveguide 12 is formed on a Z-Cut LiNbO5 substrate 11 under the following conditions. This optical waveguide 12 includes an input optical waveguide 12a, a branch optical waveguide 12b, an arm portion 12c,
It consists of a merging optical waveguide 12d and an output optical waveguide 12e. First, Ti was added as an impurity to increase the refractive index, and the optical waveguide width W was 8 μm, and the Ti film thickness before diffusion was dl-800.
Diffusion is performed as a human in an air atmosphere at 1050°C for 8 hours. Furthermore, MgO is added as an impurity to reduce the refractive index and is diffused in an oxygen atmosphere at 950° C. for 4 hours with a width W of 8 μm and an MgO film thickness of d2-450 mm in the output optical waveguide 1 after merging.

After that, a 5i02 film 14 with a thickness of 3000 was added as a buffer layer.
A film is formed manually, and a control electrode 1.3 is formed on the film.

The Ti film thickness d0 of the arm portion 12c before diffusion is set so that the propagation mode of the optical waveguide after diffusion has a confinement strength slightly weaker than the primary mode cutoff. Furthermore, the output optical waveguide 12e after merging is made of Ti, which increases the refractive index.
By doubly diffusing MgO, which reduces the refractive index, the confinement strength is set to be sufficiently weaker than the first-order mode cutoff. As a result, the arm portion 12
In C, since the overlap between the electric field distribution by the control electrode 13 and the field distribution of the propagating light is maximized, the modulation efficiency is improved and the electrode length can be made shorter than before.
In addition, in the output optical waveguide 12e after merging, the 0th mode light propagates because the confinement of light is weak, but the 1st mode light generated when a voltage is applied can be radiated out of the optical waveguide over a sufficiently short distance. .

By selectively double-diffusing Ti and MgO in this way, the light confinement strength at the electrode part and the convergence part can be optimized.
The total element length of the Z modulator can be shortened.

FIG. 2 is a perspective view showing a second embodiment of the invention.

In this embodiment, in order to form the optical waveguide 12, first T
i is the optical waveguide width W = 8 μm, the Ti film thickness before diffusion is dl, and then MgO is diffused into the straight portion after merging and before branching, that is, the input optical waveguide 12a and the output optical waveguide 12e.
The width W is 8 μm, and the MgO film thickness before diffusion is d2. After that, a 5i02 film 14 with a thickness of 3 is used as a buffer layer.
The control electrode 13 is formed on top of the film. As a result, the modulator has a symmetrical structure on both the input side and the output side, and there is no need to use each side separately.

FIG. 3 is a perspective view showing a third embodiment of the present invention.

In this embodiment, in order to form the optical waveguide 12, Ti is diffused to have a film thickness d1 before diffusion, and then the MgO film thickness d2 of the output optical waveguide 12e after merging and the input optical waveguide 12a before branching, and the MgO film thickness d2 of the optical waveguide 12 before being split. By providing the MgO film thickness taper portion 37 in the converging optical waveguide 12d and the branching optical waveguide 12b and diffusing the MgO film by gradually changing the MgO film thickness before diffusion, it is possible to reduce mode conversion loss due to sudden changes in optical confinement strength. I'm trying.

The above description has been given using an example in which an optical waveguide is formed using Ti diffusion in a Z-Cut LiNbO3 substrate, but it can also be formed in other substrate orientations, or in semiconductors such as GaAs, InP, or other substrate materials using impurity diffusion in MZ. The fact that the method according to the present invention is effective even when a modulator is configured is that MZ
This is obvious considering the principle of the modulator.

〔Effect of the invention〕

As explained above, according to the present invention, it is possible to control the light confinement strength of the electrode part and the confluence part of the MZ modulator to the optimal strength independently, thereby shortening the overall length of the element and realizing compact optical modulation. You can get the equipment.

61... 0th-order mode incident light, 62.63... Mutually in-phase guided light, 64... 0th-order mode output light, 65°6
6... Guided light with opposite phases to each other, 67... First-order mode light.

Claims (1)

[Claims] The optical waveguide is formed by diffusing impurities that increase the refractive index on the substrate, and consists of an input optical waveguide, a branch optical waveguide, an arm section, a merging optical waveguide, and an output optical waveguide, and a control electrode is attached to the arm section. An optical modulator comprising: an optical modulator, characterized in that another impurity having an effect of reducing the refractive index is introduced into at least the output optical waveguide.