Optical waveguides – With optical coupler – Particular coupling structure
Reexamination Certificate
1999-11-23
2001-06-26
Ngo, Hung N. (Department: 2874)
Optical waveguides
With optical coupler
Particular coupling structure
C385S014000, C385S131000, C385S132000, C372S050121
Reexamination Certificate
active
06253009
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention relates to the field of semiconductor optical components used for optical transmission or optical digital data processing. It pertains especially to all optical components comprising active and passive waveguides buried in sheathing layers, for example components such as semiconductor lasers, semiconductor amplifiers or again modulators for example.
At present, it is being sought to reduce the cost of optical modules to the maximum. The greatest part of the cost here most usually arises from the joining of such components to optical fiber, as for example when it is necessary to couple optical modes of very different sizes. Thus, when a laser and a flat-end single-mode optical fiber are joined together, the optical mode of the laser, whose diameter is for example 2 &mgr;m, has to be coupled with the optical mode of the optical fiber whose diameter is far greater, for example in the range of 9 &mgr;m.
To enable the coupling of these optical modes of very different sizes, spot-size converters are made in order to increase the size of the mode at the output of the optical component and make its profile compatible with that of the mode guided in the optical fiber. However, this mode matching must be done while preserving the performance characteristics of the component. Thus, for a semiconductor laser for example, it must be possible to maintain its performance characteristics in terms of threshold current and efficiency. It is necessary indeed that, in the range of operating temperature (generally between −40° C. and +85° C.), the laser component should have a threshold current that is as low as possible and efficiency that is as high as possible. To thus preserve the performance characteristics of the component and not debase the current threshold and efficiency parameters, it appears to be preferable to carry out the expansion of the optical mode in a passive waveguide.
2. Description of the Prior Art
Several approaches have been envisaged to the deconfining of the optical mode of a component. A first method called butt coupling that enables the coupling of a passive waveguide with an active waveguide is very common today. The drawings of
FIGS. 1A and 1B
respectively show a top sectional view and a longitudinal sectional view of an integrated optical component during its manufacture by the butt coupling method.
This method consists, in a first stage, in achieving the growth, on a substrate
1
, of a first layer
5
constituting the active waveguide formed for example by a quaternary material and in burying this layer in a sheathing layer
6
constituted for example by InP. These two layers
5
and
6
are then etched locally according to a standard etching method on a zone reserved for the integration of a passive type of waveguide. An epitaxial regrowth operation is performed to make this passive waveguide. For this purpose, a layer of quaternary material
2
; capable of acting as the passive waveguide, is deposited on the substrate
1
in the zone that is locally etched beforehand. Then it is buried in a sheathing layer
3
made of InP for example. The structure of the active waveguide
5
is different from that of the passive waveguide
2
. The coupling interface
7
between the two types of waveguides is called a butt joint. Furthermore, to enable the deconfining of the optical mode, the thickness of the passive guide
2
tapers evenly all along the passive section.
This method of manufacture is fully mastered at the present time. However, it requires an additional step of etching and epitaxy regrowth, thus giving rise to an increase in the cost price of the component. Furthermore, for aligning the active and passive guides, the alignment tolerance values remain low. Although the technique of butt coupling is well mastered, it is an extremely important step. This method is relatively complex to implement. It entails costs that are still high.
A second method, known as the method of selective epitaxial growth, has been considered. It is shown schematically in
FIGS. 2A and 2B
which respectively show a top sectional view and a longitudinal sectional view of an integrated optical component during its manufacture. In this method, the composition of a waveguide
7
is made to vary continuously, to make it gradually go from an active waveguide state
7
A to a passive waveguide state
7
B. The selective growth of the material constituting the waveguide
7
is achieved on a substrate
1
by the use of two dielectric masks, made of silica (SiO
2
) or silicon nitride (Si
3
N
4
) for example, placed side by side. The species under growth do not get deposited on these masks, and a phenomenon of diffusion of species under growth is created. The shape of the masks is determined so that the phenomenon of diffusion of the species is pronounced to a greater or to a lesser extent, depending on the regions of the waveguide that are considered. Just as in the butt coupling method, the thickness of the waveguide in the passive section
7
B tapers down in order to permit the deconfinement of the optical mode and therefore the increasing of its size. The optical guide
7
A,
7
B is furthermore buried in a sheathing layer
9
.
This second method has the advantage of comprising only one epitaxial step. However, it cannot be used to optimize the two waveguides, namely the active waveguide
7
A and the passive waveguide
7
B, separately. This means that it necessitates compromises. Furthermore, this method does not enable a clear definition of the boundary between the two types of guides, active and passive, because the change in state is gradual. The fact of not being able to define this boundary enables penalties because it is difficult to know where to position the electrode necessary for the operation of the component. This electrode must indeed be positioned above the active guide to ensure efficient operation of the component. By contrast, if it covers a part of the passive guide, electrical leaks are created that penalize and degrade the threshold current and efficiency parameters.
A third approach that has been envisaged in order to obtain a spot-size converter integrated into an optical component is shown schematically in the top sectional view of FIG.
3
. In this approach, an active waveguide
15
and a passive waveguide
12
are superimposed so as to create a damped vertical coupling zone SC, wherein the width of the active waveguide
15
tapers down gradually to deconfine the mode, and the width of the passive waveguide
15
increases very rapidly and then becomes constant throughout the length of this section. The two guides, namely the active guide
15
and the passive guide
12
, are furthermore buried in a sheathing layer
17
. In this case, the width of the coupling section SC must be sufficient to enable complete deconfinement of the optical mode in the entire active guide. This length is generally greater than 150 &mgr;m. Furthermore, as and when it gets deconfined, the optical mode transits through the passive guide
12
. The passive guide has a constant width of about 4 &mgr;m and a very small thickness of about 50 nm to enable the deconfinement of the mode. For, an excessive thickness of this passive guide would prevent the deconfinement of the mode in the active guide. This is why the passive guide generally has a thickness of less than 100 nm.
The major drawback of this approach lies in the fact that the deconfinement of the mode is done entirely in the active guide. This leads to a deterioration of the performance characteristics of the component, especially its threshold current and its efficiency. Furthermore, the optical coupling losses with a single-mode optical fiber are still high. They are about 4.5 dB.
The present invention makes it possible to resolve the problems related to the prior art components. For this purpose, it proposes an optical component comprising an integrated spot-size converter wherein two waveguides, one active and one passive, are superimposed and each of them has a
Colson Veronique
Lestra Alexis
Alcatel
Ngo Hung N.
Sughrue Mion Zinn Macpeak & Seas, PLLC
LandOfFree
Semiconductor optical component comprising a spot-size... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Semiconductor optical component comprising a spot-size..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Semiconductor optical component comprising a spot-size... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2538848