Active solid-state devices (e.g. – transistors – solid-state diode – Incoherent light emitter structure – With reflector – opaque mask – or optical element integral...
Reexamination Certificate
2001-01-10
2003-10-07
Crane, Sara (Department: 2811)
Active solid-state devices (e.g., transistors, solid-state diode
Incoherent light emitter structure
With reflector, opaque mask, or optical element integral...
C257S096000, C438S031000
Reexamination Certificate
active
06630693
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to electro-optic semiconductor devices which incorporate multiple active and passive component parts and methods for making the same. In particular the present invention relates to semiconductor lasers and detectors with integrated semiconductor waveguides and extended cavity semiconductor lasers and methods for making the same.
2. Discussion of Prior Art
Electro-optic semiconductor devices are used in the quickly developing field of high speed analogue and digital optical signal processing circuitry which have applications in high speed, wide-bandwidth optomicrowave and optical transmission techniques particularly in the field of telecommunications.
Electro-optic semiconductor components, such as semiconductor lasers and waveguides, are currently grown separately in different growth processes to generate different component structures on separate crystal substrates. Subsequently, the separate component structures are assembled into a single subsystem. There are significant losses at interfaces between the separate component structures which lead to high input power requirements. Also, problems of accurately optically aligning the separate component structures leads to high complexity and cost.
Alternatively, electro-optic semiconductor components, such as semiconductor lasers and waveguides are grown with quantum well layers, which extend over the whole area of crystal growth. Selected areas of these quantum well crystal growth layers are subsequently destroyed by the process of quantum well intermixing in which the selected parts of the quantum well layers are caused to diffuse into surrounding crystal growth layers. Unfortunately, the quantum well diffusion method only allows device to be grown within a passive waveguide. Also, the diffuse material can change the properties of the surrounding crystal growth layers, for example, passive waveguide core layers, in an undesirable manner. The diffusion process is significantly alloy dependent, ie. what works for a GaAlAs structure does not work for a InGaP structure. This places a constraint on the type of materials that can be used in the quantum well layers. Also, this method of crystal growth requires a post processing quantum well intermixing step to destroy the selected areas of the layers of quantum wells, which adds expense.
A further method of growing layers of quantum wells within a passive waveguide core is to grow the layers of quantum wells over the whole area of crystal growth and then stop the crystal growth and selectively etch away unwanted areas of the layers of quantum wells. After the etching process a passive waveguide layer is grown over the remaining areas of quantum well layers. The problem with this process is that the growth process has to be interrupted to perform at least one etching process. This adds complexity and cost to the growth process and can reduce yield.
A method and device are described in U.S. Pat. No. 5,418,183 in which the device is made using selective area epitaxy in which a mask is deposited on the partly grown device and the mask is defined by an etching process. This is followed by further epitaxial growth in the areas exposed by the mask and then by removal of the mask by etching. The structure and thus the resulting characteristics of the epitaxial growth in the areas exposed by the mask are critically dependent on the dimensions of the deposited mask. Also, the removal of the partly grown structure from the epitaxial growth chamber for the two etching stages and the processes involved in the etching stages can introduce contamination into the resulting device. Accordingly, the method described is relatively complex and can have associated low yields.
SUMMARY OF THE INVENTION
The present invention aims to overcome at least some of the above mentioned problems by providing electo-optic semiconductor devices which are cheap, have low losses and so reduced power requirements and are robust.
Therefore, according to the present invention there is provided an electro-optic semiconductor device comprising a semiconductor waveguide with a core region within which is located at least one active area wherein the core of the waveguide outside of the active area is not contaminated with diffuse active area material and the active area and waveguide are monolithic and are grown in an additive growth process.
Also, according to the present invention, there is provided an electro-optic semiconductor device comprising a semiconductor waveguide with a core region within which is located at least one active area wherein the profile of the core region follows the profile of the active area, and the active area and waveguide are monolithic and are grown in an additive crystal growth process.
The electro-optic semi-conductor device according to the present invention is monolithic and so is a structure composed of a single continuous crystalline mass of material.
The term additive growth process is used to define a growth process in which a monolithic crystal structure is grown without any intermediate stages in which material is removed from the crystal previously grown. For example, it excludes growth processes which have an intermediate etching stage. Clearly, an additive growth process will generally be simpler and so will provide a cheaper way of making a monolithic crystal substrate comprising an active area located within a passive waveguide region.
The present invention combines this advantage with the advantage that no post processing stage is require to destroy selected parts of the active areas. Also, as the resulting device is monolithic there are very low coupling losses between the active area and the waveguide.
In post processing steps the active areas are configured as electro-optic components, such as lasers and detectors. Integration of at least one active area in the core of a waveguide enables optical connection between a plurality of electro-optical components, configured from the active areas, within a single crystal structure. Subsequent device configuration uses conventional and established processing technology to configure the active areas into electro-optic components. This monolithic integration will allow repeatable and reliable fabrication on a single substrate with low cost if high volumes are achieved. The resultant devices are simple, compact and robust because the entire device is a single crystalline structure. Optical alignment of, for example, a laser/waveguide interface requires no mechanical alignment, and so good optical alignment is achieved cheaply and without adding complexity.
Preferably the active area is located between two adjacent growth layers of the core of the waveguide. The growth method for this structure is very simple and enables multiple active areas to be grown in a single epitaxial growth which does not require multistage processing or regrowth.
An active area preferably comprises at least one quantum well layer or thin bulk layer having a smaller bandgap than the core of the waveguide. The parts of the resulting single crystal structure outside an active area become the passive optical interconnections defined by the waveguide whereas the regions with quantum wells become, for example, active laser or detector devices on subsequent processing. This is because the waveguide, having a wider bandgap is transparent to the light generated or detected by the active area.
It is preferable that the bandgap in an active area is less than the bandgap in a transition region between that active area and the core of the waveguide. This reduces losses due to band gap absorption outside of the active area.
The quantum well layers may be configured to act as a laser by post growth processing in which a laser stripe and an electrode are added. Alternatively, the quantum well layers may be configured to act as a detector by virtue of their absorption of light when they are reverse biased.
Preferably, the waveguide and active areas are grown using chemical beam epitaxy (CBE). CBE
Kane Michael J
Martin Trevor
Crane Sara
Nixon & Vanderhye P.C.
QinetiQ Limited
LandOfFree
Electro-optic semiconductor devices and method for making... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Electro-optic semiconductor devices and method for making..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Electro-optic semiconductor devices and method for making... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3113759