JPH07174927A - Optical modulator - Google Patents

Abstract

PURPOSE:To reduce the crosstalk of emitted light in a laminated directional coupler type optical modulator. CONSTITUTION:A lower waveguide layer 44, a separation layer 46, an upper waveguide layer 48, upper clad layers 50 and 52 are successively provided on a substrate main face 42a. Each layer from the lower waveguide layer 44 to the upper clad layer 50 is formed into a stripe state, and buried layers 56 and 58 are buried in both end parts of each layer. And a notched end face 54 is provided on the emitting end face 48a of the upper waveguide layer. The notched end face 54 is made plane so as to be parallel in the normal direction H of the substrate main face 42a and inclined at an angle alpha deg. against the emitting end face 44a of the lower waveguide layer. Thus, the light emitting directions of the upper and lower emitting end faces 48a and 44a shift each other. Thus, the light emitted from the emitting end face 48a of the upper waveguide layer is prevented from being made incident on the emitting end face 44a of the lower waveguide layer in the case where an optical fiber is connected to the emitting end face 44a of the lower waveguide layer 44a.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION The present invention relates to, for example, light intensity modulation,
The present invention relates to a laminated directional coupler type optical modulator used for optical switching control or optical signal generation.

[0002]

2. Description of the Related Art Hitherto, a laminated directional coupler type optical modulator having two coupling waveguides laminated in a direction perpendicular to a main surface of a substrate has been proposed. FIG. 16 shows Document 1: IEEE P
HOTONICS TECHNOLOGY LETTERS, VOL.4, NO.7, p43〜45, JUL
Y 1992 `` Vertical Multiple-Quantum-Well Directional
-Cupler Switch with Low Switcing Voltage "is a schematic diagram showing the configuration of a laminated directional coupler type optical modulator.

In the optical modulator shown in FIG. 1, n-GaA is used.
On the one main surface 10a of the substrate 10, n-Al is sequentially formed.
GaAs / GaAs clad layer 12, n-AlGaAs
/ GaAs waveguide layer 14, n-AlGaAs / GaAs separation layer 16a, i-AlGaAs / GaAs separation layer 16
b, p-AlGaAs / GaAs separation layer 16c, p-A
lGaAs / GaAs waveguide layer 18 and p-AlGaAs
/ GaAs clad layer 20 is provided. Further, on the clad layer 20, p-GaAs contact layers 22 and p are sequentially formed.
The side electrode 24 is provided, and n is provided on the other main surface 10b of the substrate 10.
The side electrode 26 is provided. The p-side electrode 24 is an Au / Zn electrode,
The n-side electrode 26 is an Au / Ge electrode. Lower waveguide layer 1
4 and the upper waveguide layer 18 form a directional coupler in close proximity so as to cause an interaction of light, and when light is incident on one of the waveguide layers 14 and 18, a light intensity is generated between these waveguide layers. The periodic transition of occurs.

The conductivity type of the lower clad layer 12, the waveguide layer 14 and the separation layer 16a is n-type, and the upper clad layer 2 is
0, the conductivity type of the waveguide layer 18 and the separation layer 16c is p-type,
Since the separation layer 16b in the central portion is non-doped, when a voltage is applied to the pin junction formed by them,
The refractive index change can be caused only in the central separation layer 16b. By variably adjusting the voltage applied to the pin junction, the ratio between the optical output intensity of the lower waveguide layer 14 and the optical output intensity of the upper waveguide layer 18 changes, so that the intensity of light is modulated. Can control the switching of light.

Further, in this conventional example, the cladding layers 12 and 2 are
0, the waveguiding layers 14 and 18, and each of the separation layers 16a to 16c are formed of MQW (Multi Quantum Well), so that the voltage required to cause a change in the refractive index of the separation layer 16b in the central portion can be reduced.

FIG. 17 is a diagram showing an example of usage of the optical modulator. In the example shown in the figure, the light emitted from the semiconductor laser 28 as the light source is incident on the optical modulator 32 through the optical fiber 30. Then, the light emitted from the light modulator 32 is incident on the optical device 36 at the next stage through the condenser lens 34. When the device of FIG. 16 is used as the optical modulator 32, the optical fiber 30 is coupled to the incident end of the upper waveguide layer 18, and the emitting end of the lower waveguide layer 14 is connected to the optical device 3 of the next stage.
Bind to 6. Reference numeral 38 denotes a driving power source for the semiconductor laser 28.

By the way, even if the outgoing beam of the optical fiber 30 is focused and incident on the incident end of the waveguide layer 18, its beam diameter is generally about 2 μm. On the other hand, in the optical modulator of FIG. 16, the waveguide layers 14 and 18 both have a thickness of approximately 0.5 μm, and the separation layers 16a, 16b and 16c have a thickness of approximately 0.3, 0.6 and 0, respectively. ．
It becomes 3 μm. Therefore, with the configuration shown in FIG. 16, the outgoing beam from the optical fiber 30 is incident not only on the upper waveguide layer 18 but also on the lower waveguide layer 14.

To prevent this, a structure has been proposed in which the incident end side of the upper waveguide layer 18 is removed by a predetermined length in the optical modulator of FIG. This structure is shown in FIG.
Shown in. In this structure, since the incident end position of the upper waveguide layer 18 and the incident end position of the lower waveguide layer 14 are displaced in the optical axis direction of the upper waveguide layer 18, the outgoing beam from the optical fiber 30 is guided to the lower side. It can be prevented from entering the wave layer 14. still,
Upper waveguide layer 18 after removing the incident end side by a predetermined length
And the incident end surface of the lower waveguide layer 14 are substantially parallel to each other.

[0009]

However, in the above-described conventional optical modulator, since the lower waveguide layer 14 and the upper waveguide layer 18 are extremely close to each other, the light emitted from each of these waveguide layers is separated. Is difficult, resulting in a poor extinction ratio. The extinction ratio also deteriorates when considering the coupling of the light emitted from the lower waveguide layer 14 with the optical fiber. Since the core diameter of a standard optical fiber is about 10 μm, the distance between the waveguide layers 14 and 18 is very small compared to the core diameter. For this reason, the light emitted from both the waveguide layers 14 and 18 enters the core of the optical fiber, and the extinction ratio deteriorates.

Further, even if the output beam of the optical fiber 30 is focused on the input end of the upper waveguide layer 18, the output beam diameter at the input end of the upper waveguide layer is generally 2
It is about μm. Therefore, the exit beam diameter is
8 is larger than the layer thickness of 8 and is substantially equal to or larger than the separation distance between the upper waveguide layer 18 and the lower waveguide layer 14.
As a result, it becomes difficult to couple the optical fiber 30 and the upper waveguide layer 18 so that the outgoing beam does not enter the lower waveguide layer 14 but only the upper waveguide layer 18. Also, for such coupling, the upper waveguide layer 1
Insertion loss into 8 tends to increase.

In order to reduce the size of the optical modulator, it is sufficient to shorten the distance between the lower waveguide layer 14 and the upper waveguide layer 18, thereby shortening the element length of the optical modulator. The above-mentioned problems are aggravated as the separation distance of the wave layers becomes shorter.

An object of the present invention is to solve the above-mentioned conventional problems and to provide an optical modulator capable of obtaining a better extinction ratio.

More preferably, it is to provide an optical modulator capable of reducing insertion loss.

[0014]

In order to achieve this object, an optical modulator according to the present invention is provided on a main surface of a semiconductor substrate.
In a laminated directional coupler type optical modulator including a lower waveguide layer, a separation layer, an upper waveguide layer and an upper clad layer, which are sequentially provided, a cutout end face for deflecting emitted light is provided at an emission end of the upper waveguide layer. Is provided, and the notched end face is a plane parallel to the normal line direction of the main surface of the substrate and inclined by a predetermined angle α ° with respect to the emission end face of the lower waveguide layer.

[0015]

According to this structure, since the notched end face is a plane inclined by a predetermined angle α ° with respect to the emission end face of the lower waveguide layer, the light emission direction of the upper waveguide layer and the lower waveguide layer are reduced. It is possible to deviate from the light emission direction of. Moreover, since the notch end face is parallel to the normal direction of the main surface of the substrate, even if the light guided through the upper waveguide layer is reflected by the notch end face and becomes return light,
The reflected light can be reflected in the direction along the substrate surface, that is, in the direction not toward the lower waveguide layer.

[0016]

Embodiments of the present invention will be described below with reference to the drawings. It should be noted that the drawings are merely schematic representations so that the invention can be understood, and therefore the invention is not limited to the illustrated examples.

FIG. 1 is an overall perspective view schematically showing the structure of the apparatus of the embodiment. The laminated directional coupler-type optical modulator 40 shown in the figure has a lower waveguide layer 44, a separation layer 46, an upper waveguide layer 48, and upper cladding layers 50, 52 on a main surface 42 a of a semiconductor substrate 42. Are sequentially provided. A cutout end face 54 for deflecting emitted light is provided at the output end 48a of the upper waveguide layer 48, and the cutout end face 54 is parallel to the normal line direction H of the substrate main surface 42a and emitted from the lower waveguide layer 44. It is a plane inclined by a predetermined angle α ° with respect to the end face 44a.

2 to 4 are sectional views schematically showing the construction of the apparatus of the embodiment. FIG. 2 is a cross section of the example device taken along a plane parallel to the normal direction H and passing through the waveguide layers 42 and 46,
3 and 4 show cross sections corresponding to the III-III cross section and the IV-IV cross section of FIG.

In this embodiment, the light modulation device 40 is shown in FIG.
As also shown, the optical modulator 40a and the semiconductor laser 40b are provided on the same substrate 42. Semiconductor laser 40
b comprises an active layer 68 and upper clad layers 70 and 72 sequentially provided on the main surface 42a of the substrate, and the emitting end of the active layer 68 is coupled to the incident end of the lower waveguide layer 44 included in the optical modulator 40a. Let

As shown in FIGS. 2 and 3, the optical modulator 40a includes a lower waveguide layer 44, a separation layer 46, and an upper waveguide layer 48 which are sequentially provided on the substrate main surface 42a of the modulator region A. And the upper clad layers 50 and 52, the lower waveguide layer 44, and the separation layer 46.
And the buried layer 5 filling both sides of each layer of the upper waveguide layer 48.
6, 58. Although a clad layer may be separately provided between the substrate 42 and the lower waveguide layer 44, the substrate 42 is used here.
Also serves as the lower clad layer of the optical modulator 40a.

Then, a contact layer 60, an ohmic layer 62 and one control electrode 64 are sequentially provided on the upper clad layer 52. Further, the other control electrode 66 is provided on the other substrate main surface 42b. In order to electrically control the refractive index of the separation layer 46, the lower clad (the substrate 42 in this example) and the lower waveguide layer 44 are the first conductivity type semiconductor layer, and the separation layer 4
6 is a non-doped semiconductor layer (i layer), the upper waveguide layer 48 and the upper clad layers 50 and 52 are second conductivity type semiconductor layers, and a pin junction is formed by these layers.
By applying a voltage to this pin junction via the electrodes 64 and 66, the refractive index of each of the first and second conductivity type semiconductor layers is not substantially changed, while the separation layer 46, which is the i layer, is not changed. The refractive index can be changed. The ratio between the optical output intensity of the lower waveguide layer 44 and the optical output intensity of the upper waveguide layer 48 changes according to the amount of change in the refractive index of the separation layer 46.

The buried layers 56 and 58 are for confining light in the direction parallel to the main surface 42a of the substrate and orthogonal to the waveguide layers 44 and 48, and therefore, the buried layers 56 and 58. At least the lower waveguide layer 44 and the separation layer 4
6 and the both side portions of the upper waveguide layer 48 may be embedded. Preferably, the buried layers 56 and / or 58 are formed of a high resistance material, which reduces the capacitance between the control electrodes 64, 66. The upper clad layer 50 also serves as a protective layer for preventing the upper waveguide layer 48 from being contaminated during the manufacturing process of the optical modulator 40a.
0 does not always have to be provided. The optical modulator 40a may have the same configuration as the conventional one.

Although not shown, the emission end 44 of the lower waveguide layer 44
An antireflection film (non-reflection coating) is provided on the a and the emitting end 48a of the upper waveguide layer 48.

As shown in FIGS. 2 and 4, the semiconductor laser 40b emits light to the active layer 68 and the upper clad layers 70 and 72, which are sequentially provided on the main surface 42a of the substrate in the laser region B. A diffraction grating 74 for confinement and buried layers 76 and 78 provided on both sides of the active layer 68 are provided. Although a clad layer may be separately provided between the substrate 42 and the active layer 68, the substrate 42 also serves as the lower clad layer of the semiconductor laser 40b in this case.

Then, a contact layer 80, an ohmic layer 82 and one control electrode 84 are sequentially provided on the upper clad layer 72. The other control electrode 66 is connected to the semiconductor laser 40.
It is used as an electrode common to b and the optical modulator 40a. The one control electrode 84 and the ohmic layer 62 of the semiconductor laser 40b are electrically separated from the one control electrode 84 and the ohmic layer 82 of the optical modulator 40a. Therefore, for example, the control electrode 84 and the ohmic layer 62 are separated from the control electrode 84 and the ohmic layer 82 by the groove 88.

The buried layers 76 and 78 function as a current confinement layer, and the structure of the buried layers 76 and 78 is any suitable structure capable of preventing current injection into a region where the active layer 68 is not provided. Can be Here, the buried layer 76
Is formed of a high resistance material, and a pn junction is formed by the buried layer 78 and the upper clad layer 72. By providing these buried layers 76 and 78 in the region where the active layer 68 is not provided, current injection is blocked in the region where the active layer 68 is not provided, and the current is concentrated in the region where the active layer 68 is provided. inject. The upper clad layer 72 constitutes a current confinement layer together with the buried layer 78. The buried layers 76 and 78 are the main surface 4 of the substrate.
It also has a function of confining light in a direction parallel to 2a and orthogonal to the active layer 68.

Further, preferably, the semiconductor laser 40b is D
FB laser (Distributed Feedback Laser) or DBR
It is a laser (Distributed Bragg-reflector Laser). Therefore, the arrangement position of the diffraction grating 74 can be set to any suitable position where light can be confined in the active layer 68 and laser oscillation can be performed. Here, the optical waveguide layer 80 is provided between the active layer 68 and the upper clad layer 70, and the diffraction grating 74 is provided at the interface between the upper clad layer 70 and the optical waveguide layer 80.

In this embodiment, the entrance end of the lower waveguide layer 44 included in the optical modulator 40a and the exit end of the active layer 68 included in the semiconductor laser 40b are coupled to each other, and these modulators 40 are combined.
Since a and 40b are provided on the same substrate 42, it is possible to reduce optical insertion loss when light is incident from the semiconductor laser 40b as a light source to the optical modulator 40b.

Next, the manufacturing process of the optical modulator of this embodiment will be described with reference to an example. 5 to 12 are perspective views schematically showing the manufacturing process of the embodiment.

First, the substrate 42 has a carrier concentration of 1 × 1.
An n-InP substrate of about 0 ¹⁸ / cm ³ is prepared. The (100) substrate surface is defined as the substrate main surface 42a, and the main surface 42a
A non-doped InGaAsP active layer 68 and a p-InGaAsP optical waveguide layer 86 are sequentially epitaxially grown on a. Next, the diffraction grating 74 is formed on the optical waveguide layer 86 by the photolithography and wet etching techniques using the two-beam interference exposure apparatus. The period (pitch) of the diffraction grating 74 is set to about 2340 A ° (A ° represents angstrom). Next, in order to prevent the diffraction grating 74 from being contaminated, the p-InP upper cladding layer 70 is formed on the diffraction grating 74.
Are epitaxially grown (FIG. 5). Board end face 42c
And 42d are (011) planes, and the substrate end surfaces 42c, 42
A normal direction of d is a stripe direction S.

Next, the upper clad layer 70 in the laser region B is formed.
An island-shaped SiO ₂ etching mask 90 is formed thereon. The planar shape of the etching mask 90 has a length L and a width H.
Of rectangle. The length L is the length of the mask 90 in the stripe direction S, and this L is made equal to the optical resonator length of the semiconductor laser 40b. The width H is the width of the mask 90 in the direction parallel to the substrate main surface 42a and orthogonal to the stripe direction S. Next, each layer from the upper clad layer 70 to the active layer 68 is etched to form the main surface 42a of the area where the etching mask 90 is not formed, that is, the main surface 42a of the modulation area A and the main surface of the laser area B on both sides of the mask 90. 42a is exposed. At the same time, in the region where the etching mask 90 is formed, the upper clad layer 70,
The optical waveguide layer 86 and the active layer 68 are left in an island shape to obtain an island-shaped body 92 composed of these island-shaped layers 70, 86 and 68 (FIG. 6). The planar shape of the island-shaped body 92 is the etching mask 9
It is almost equal to the plane shape of 0.

Next, the n-InGaAsP lower waveguide layer 44 and i-In are sequentially formed on the substrate main surface 42a of the modulator region A.
P (non-doped InP) isolation layer 46, p-InGaAs
The P upper waveguide layer 48 and the p-InP upper cladding layer 50 are epitaxially grown (FIG. 7). During this growth, if the width H of the etching mask 90 is too wide, abnormal growth such as growth of a polycrystal (polycrystal) on the etching mask 90 occurs. Therefore, it is desirable that the width H of the etching mask 90 be narrow enough to prevent such abnormal growth. The etching mask 90 also functions as a crystal growth blocking layer, and thus the etching mask 9
No crystals grow on 0.

Next, the etching mask 90 is removed, and thereafter, a SiO ₂ stripe mask 94 is formed on the upper cladding layer 50 and the islands 92 (FIG. 8). The SiO ₂ stripe mask 94 parallel to the stripe direction S is extended from the start position X of the modulator region A to the end position Y of the laser region B. A stripe mask 94 in a direction parallel to the substrate main surface 42a and orthogonal to the stripe direction S
Width h (H> h) of the waveguide layer 4 of the optical modulator 40a and the transverse mode in the active layer 68 of the semiconductor laser 40b.
4, 48 and the lateral modes in the separation layer 46 are controlled.

Next, the upper clad layer 50 in the modulator area A is formed.
To the lower waveguide layer 44 and the layers from the upper cladding layer 70 to the active layer 68 in the laser region B are etched to expose the substrate main surface 42a on both sides of the stripe mask 94. Further, the stripe mask 9 in the modulator area A
4, the upper clad layer 50, the upper waveguide layer 48, the separation layer 46 and the lower waveguide layer 44 are left in a stripe shape, and these stripe layers 50, 48, 46 are formed.
And a stripe structure 96a of 44 are obtained. At the same time, the upper clad layer 70, the optical waveguide layer 86, and the active layer 68 are left in the stripe shape in the area of the laser area B where the stripe mask 94 is formed, and the laser stripe formed of these stripe layers 70, 86, and 68. Structure 9
6b is obtained. Etching is performed using, for example, Br ₂ and CH ₃ O.
Wet etching is performed by using a mixed aqueous solution of H (brom type etchant) as an etchant. These stripe structures 96a and 96b are integrated so that the lower active layer 44 of the modulator region A and the active layer 68 of the laser region B are coupled (FIG. 9). Stripe structure 96a
The planar shape of the stripe masks 94 and 96b is substantially equal to the planar shape of the stripe mask 94.

Next, while leaving the stripe mask 94, the Fe-doped InP burying layer 98 and the n-InP burying layer 100 are sequentially epitaxially grown on the main surface 42a of the substrate, and these burying layers 98 and 100 are used. , Both sides of the stripe structure 96a and both sides of the stripe structure 96b are embedded. The stripe mask 94 also functions as a crystal growth blocking layer, so that no crystal grows on the stripe mask 94. Here, the buried layer 56 in the modulator region A,
58 and the buried layers 76 and 78 in the laser region B are formed of a common material, and the buried layers 98 and 100 in the modulator region A are formed.
Are used as the buried layers 56 and 58, and the buried layers 98 and 100 in the laser region B are used as the buried layers 76 and 78.
Preferably, the buried layer 98 has a high resistance, and the sidewall surfaces of the stripe structures 96a and 96b are covered with the buried layer 98 over substantially the entire surface thereof. Buried layer 9
After growing 8 and 100, the stripe mask 94 is removed (FIG. 10).

Next, the buried layer 100 and the stripe structure 96.
On the upper clad layer 50 of a and the upper clad layer 72 of the stripe structure 96b, the p-InP upper clad layer 102 and the p-InGaP contact layer 104 are sequentially epitaxially grown (FIG. 11). here,
The upper clad layer 52 in the modulator region A and the upper clad layer 72 in the laser region B are formed of a common material, and the upper clad layers 102 in the modulator region A and the laser region B are formed in the upper clad layers 52 and 72, respectively. Used as. Similarly, the contact layer 60 in the modulator region A and the contact layer 80 in the laser region B are formed of a common material, and the contact layers 104 in the modulator region A and the laser region B are used as the contact layers 62 and 80, respectively.

Next, the ohmic layer 106 is formed on the contact layer 104. The ohmic layer 106 is, for example,
Au film and A sequentially stacked from the contact layer 104 side
It consists of a uZn alloy film. After that, the ohmic layer 106 is sintered to contact the ohmic layer 1 and the contact layer 104.
Ohmic connection with 06. The ohmic layer 62 in the modulator region A and the ohmic layer 82 in the laser region B are formed of a common material, and the ohmic layers 106 in the modulator region A and the laser region B are formed in the ohmic layers 62 and 8, respectively.
Used as 2. Then the ohmic layer 1 in the modulator region A
The Au control electrode 64 and the Au control electrode 84 are formed on the ohmic layer 106 in the laser region B and the laser control region 06, respectively. After that, the groove 8 is formed by photolithography and etching technique.
8 are formed, and the ohmic layer 106 in the modulator region A and the laser region B is electrically separated by the groove 88. Further, the control electrode 66 is formed on the other substrate main surface 42b (FIG. 12).

Next, the emission end side of the upper waveguide layer 48 is etched by photolithography and etching techniques to form a flat cutout end face 54 on the emission end face 48a of the upper waveguide layer 48.
Are formed (FIG. 1). Although various methods can be used as the etching method, RIE (Reactive Ion E
tching) Other anisotropic dry etching is preferably used. Besides, chemical etching using a hydrochloric acid type etchant, a bromine type etchant, or another etchant may be used. However, when chemical etching is used, the cutout end face 54 is likely to be a tapered curved surface because the InP-based semiconductor is used as a forming material and the substrate end faces 42c and 42d are (011) faces. It is difficult and not preferable to form the end surface 54 in a flat shape.

After forming the notched end surface 54, the optical modulator 40
a is an antireflection film (antireflection coating) for example SiO _X film as an emitting facet 48a i.e. the notch end surfaces 54 of the light emitting face 44a and the upper waveguide layer 48 of the lower waveguide layer 44 covers each comprise. The antireflection film reduces the amount of light reflected by the emitting end face 48a of the upper waveguide layer and returning to the upper waveguide layer 48.

FIG. 13 is an enlarged schematic plan view of the structure of the main part of this embodiment. In the figure, an antireflection film 106 is provided on each of the lower waveguide layer emission end face 44a and the upper waveguide layer emission end face 48a of the optical modulator 40a, and the states of these emission end faces 44a and 48a and the vicinity thereof are indicated by normal lines. Shown from the direction H.

In this embodiment, the lower waveguide layer emission end face 44 is formed.
Let a be parallel to the substrate end face 42c. The upper waveguide layer emission end face 48a is parallel to the normal direction H of the substrate main surface 42a and is at an angle α ° with respect to the lower waveguide emission end face 44a.
Only the notched end surface 54 is inclined. Upper waveguide layer emission end face 4
The antireflection film 106 provided on the upper surface 8a
It is intended to reduce the reflectance of the light reflected by 8a. However, by setting the angle α ° to approximately 0 ° <α ≦ 12 °, the reflectance of the light can be reduced to a practically satisfactory level. it can.

The optical axis T1 of the lower waveguide layer 44 and the optical axis T2 of the upper waveguide layer 48 are parallel to the stripe direction S, and these optical axes T1 and T2 overlap in a plan view.

FIG. 14 is a diagram for explaining the relationship between the reflectance R of light on the end face 48a and the angle α ° when the antireflection film 106 is provided on the output end face 48a of the upper waveguide layer. In the figure, the vertical axis represents the reflectance R and the horizontal axis represents the antireflection film 10.
The film thickness dμm of 6 (see FIG. 13) is shown. Here, the oscillation wavelength of the semiconductor laser 40b is approximately 1.55 μm.
Therefore, the wavelength of light guided through the upper waveguide layer 48 is set to approximately 1.55 μm, and the antireflection film 106 is formed of a SiO _x film. Further, the refractive index (effective refractive index) felt when light propagates through the upper waveguide layer 48, the antireflection film 106 and the air is 3.2, 1.8 and 1, respectively, and the light is emitted from the upper waveguide layer emitting end face 48a. Consider a case where the light is emitted from the inside to the air through the antireflection film 106. And the angle α ° is 0 °, 5 °, 10
The value was changed stepwise to each of °, 16 ° and 17.2 °, and the relationship between the reflectance R and the film thickness d was calculated by simulation calculation at each step. Angle α ° is 0 °, 5 °, 1
The calculation results when 0 °, 16 ° and 17.2 ° are respectively 0 °, 5 °, 10 °, 16 ° and 17.2.
The curve with ° is shown. The angle α ° = 0 ° indicates that the upper emitting end face 48a and the lower emitting end face 44a are parallel to each other.

As can be understood from the figure, when the angle α ° is 0 ° and 5 °, the film thickness d of the antireflection film 106 is set to a value in the range of approximately 0.19 to 0.235 μm. As a result, the reflectance R can be set to 10 ⁻² or less, which is practically satisfactory. In either case, the film thickness d is approximately 0.2
The reflectance R can be minimized when the thickness is about 15 to 0.216 μm. When the angle α ° is 10 °, the thickness of the antireflection film 106 is approximately 0.20 to 0.24.
By setting the value in the range of 5 μm, the reflectance R can be set to 10 ⁻² or less, which is practically satisfactory. in this case,
The reflectance R can be minimized when the film thickness d is about 0.223 μm.

If the angle α ° is within the range of 0 ° <α ° ≦ 10 °, the reflectance R which can be practically satisfied can be obtained even if the angle α ° changes due to a production error. In this case, the film thickness d
Is preferably in the range of 0.20 μm <d <0.235 μm, and a standard value may be about d = 0.22 μm.

When the angle α ° is set to 16 °, a practically satisfactory reflectance R cannot be obtained. Angle α ° is 1
When the angle is 7.2 °, the reflectance R which is practically satisfactory can be obtained, but when the angle α ° deviates from 17.2 °, the reflectance R largely fluctuates and increases. Therefore, the angle α ° is accurately controlled. Otherwise, it is difficult to obtain a practically satisfactory reflectance R.

FIG. 15 is another diagram for explaining the relationship between the angle α ° and the reflectance R. In the figure, the vertical axis represents the reflectance R and the horizontal axis represents the angle α °. Here, the film thickness d =
The angle α ° is variously changed by using the 0.2156 μm SiO _x film as the antireflection film 106. The other conditions were the same as in FIG. 14, and the relationship between the reflectance R and the angle α ° was calculated by simulation calculation. Film thickness d = 0.2156
μm is the value of the film thickness at which the reflectance R becomes the smallest when the angle α ° = 0 °.

The film thickness d of the antireflection film 106 is set to an angle α ° = 0.
When the film thickness is such that the reflectance R has a minimum value at 0 °, the reflectance R can be set to 10 ⁻² or less, which is practically preferable, by setting the angle α ° to approximately 12 ° or less, Further, by setting the angle α ° to about 7.8 ° or less, the reflectance R can be set to 10 ⁻³ or less, which is more preferable in practical use.

As can be understood from the above description with reference to FIGS. 14 and 15, the angle α ° is set to 0 ° <α ° ≦ 12 °, whereby the antireflection film 10 is formed.
It can be understood that the reflectance R in the case of providing 6 can be reduced to a practically satisfactory level.

On the other hand, as shown in FIG. 13, let us consider a case where the lower waveguide layer emission end face 44a is coupled to the core 108 of the optical fiber. In this case, crosstalk occurs when light emitted from the upper waveguide layer emission end face 48a enters the core 108. However, by adjusting the angle α ° and / or the distance t, this crosstalk can be reduced or eliminated. You can The distance t is the core 108 when viewed two-dimensionally.
The distance t is the distance between the incident end face 108a and the upper waveguide layer emission end face 48a. Here, the distance t is the distance on the extension line of the optical axis T2.

The angle θ (hereinafter referred to as the radiation angle θ) of the light emitted from the upper waveguide layer emission end face 48a into the air through the antireflection film 106 is θ when the angles α ° = 5 ° and 10 °, respectively. = 11.2 ° and 23.8 °. t ・ t
If an θ ≧ Φ / 2, it is considered that crosstalk can be sufficiently reduced or eliminated in practical use, and therefore an extinction ratio that is practically satisfactory can be obtained. Φ is the diameter of the core 108.

The core diameter Φ of commonly used optical fibers
Is about 10 μm, so Φ = 10 μm and angle α °
It is considered that crosstalk can be practically sufficiently reduced or eliminated by setting t> 25 μm in the case of = 5 ° and t> 11 μm in the case of Φ = 10 μm and α ° = 10 °.

Further, the core diameter Φ = 10 μm and the distance t is 5
When it is set to 0 μm, it is considered that the crosstalk can be practically sufficiently reduced or eliminated when the radiation angle θ satisfies tan θ ≧ 1/10. θ = tan ⁻¹ (1/10) is almost 5.
7 °, and if the angle α ° is approximately 1.8 °, the radiation angle θ
Can be approximately 5.7 °.

The present invention is not limited to the above-mentioned embodiment, and therefore, the shape, size, forming material, disposition position and the like of each component can be changed suitably.

For example, in the above-mentioned embodiments, the antireflection film 106 is provided on the upper waveguide layer emission end face 48a, but the antireflection film 106 may not be provided. If the antireflection film 106 is not provided, there is a possibility that the return light reflected by the emitting end face 48a and returning to the upper waveguide layer 48 may increase. However, since the emitting end face 48a is parallel to the normal line direction H of the substrate main surface 42a, the reflection direction of the return light is the direction along the substrate main surface 42a. Therefore, it is also possible to radiate the return light from the upper waveguide layer 48 to the buried layer 56 or 58 and attenuate the light, thereby preventing the return light from entering the lower waveguide layer 44.

In the above-mentioned embodiment, the optical modulator 40a is used.
Although the semiconductor laser 40b and the semiconductor laser 40b are provided on the same substrate 42, the optical modulator 40a is provided without providing the semiconductor laser 40b on the substrate 42.
Only one may be provided.

[0057]

As is apparent from the above description, according to the optical modulator of the present invention, the notch end face inclined by the angle α ° with respect to the emitting end face of the lower waveguide layer emits light from the upper waveguide layer. Since it is provided at the end, the light emitting direction of the upper waveguide layer and the light emitting direction of the lower waveguide layer can be shifted. Therefore, when the light emitted from the lower waveguide layer is coupled to the optical fiber, the light emitted from the upper waveguide can be prevented from entering the optical fiber. As a result, the extinction ratio can be improved.

Moreover, since the notched end face is parallel to the normal line direction of the main surface of the substrate, it is assumed that the light guided through the upper waveguide layer is reflected by the notched end face to be returned light and is incident on the upper waveguide layer. Also, the reflection direction of the return light can be a direction along the substrate surface, that is, a direction not toward the lower waveguide layer. As a result, it is possible to prevent the returning light from entering the lower waveguide layer.

[Brief description of drawings]

FIG. 1 is an overall perspective view schematically showing a configuration of an embodiment of the present invention.

FIG. 2 is a sectional view schematically showing the configuration of an embodiment of the present invention.

FIG. 3 is a sectional view schematically showing the configuration of an embodiment of the present invention.

FIG. 4 is a sectional view schematically showing a configuration of an embodiment of the present invention.

FIG. 5 is a drawing for explaining the manufacturing process in the example.

FIG. 6 is a diagram for explaining the manufacturing process of the example.

FIG. 7 is a drawing for explaining the manufacturing process in the example.

FIG. 8 is a drawing for explaining the manufacturing process in the example.

FIG. 9 is a drawing for explaining the manufacturing process in the example.

FIG. 10 is a drawing for explaining the manufacturing process in the example.

FIG. 11 is a drawing for explaining the manufacturing process in the example.

FIG. 12 is a drawing for explaining the manufacturing process in the example.

FIG. 13 is a plan view schematically showing a main part configuration of an embodiment of the present invention.

FIG. 14 is a diagram for explaining the relationship between the angle α ° of the cutout end surface and the reflectance R at the exit end surface of the upper waveguide layer.

FIG. 15 is another diagram for explaining the relationship between the angle α ° of the cutout end face and the reflectance R at the emission end face of the upper waveguide layer.

FIG. 16 is a side view showing an example of the configuration of a conventional optical modulator.

FIG. 17 is a diagram showing an example of a usage pattern of a light modulation device.

FIG. 18 is a side view showing another example of the configuration of the conventional light modulator.

[Explanation of symbols]

 40: Optical Modulator 40a: Optical Modulator 40b: Semiconductor Laser 42: Substrate 42a: Substrate Main Surface 44: Lower Waveguide Layer 46: Separation Layer 48: Upper Waveguide Layer 50, 52, 70, 72: Upper Cladding Layer 54: Notched end face 68: Active layer

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI technical display location H01L 31/14 A

Claims (3)

[Claims] 1. A laminated directional coupler type optical modulator comprising a lower waveguide layer, a separation layer, an upper waveguide layer and an upper cladding layer, which are sequentially provided on a main surface of a semiconductor substrate. A cutout end face for deflecting outgoing light is provided at the exit end of the wave layer, and the cutout end face is parallel to the normal direction of the main surface of the substrate and is at a predetermined angle α ° with respect to the exit end face of the lower waveguide layer. An optical modulator characterized by having an inclined plane. 2. The optical modulation device according to claim 1, wherein an antireflection film is provided on a notched end surface of the upper waveguide layer. 3. The optical modulator according to claim 1, further comprising a semiconductor laser in which an active layer and an upper clad layer are sequentially provided on a main surface of a semiconductor substrate, the incident end of the lower waveguide layer and the active layer. An optical modulation device characterized in that it is formed by coupling with the emission end of.