Stock material or miscellaneous articles – Composite – Of metal
Reexamination Certificate
2001-03-28
2004-06-08
Thomas, Tom (Department: 2815)
Stock material or miscellaneous articles
Composite
Of metal
C428S615000, C428S620000, C428S641000, C428S457000, C428S689000, C428S700000, C257S009000, C257S183000
Reexamination Certificate
active
06746777
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention is related to substrates for epilayer epitaxial growth in which the epilayers are lattice mismatched to the substrate and, in particular, to alternative substrates for fabrication of electronic and optoelectronic devices, such as semiconductor diode lasers, for example vertical-cavity surface-emitting lasers (VCSELs).
2. Description of the Related Art
The following descriptions and examples are not admitted to be prior art by virtue of their inclusion within this section.
Lasers, such as semiconductor diode lasers, have a wide range of industrial and scientific uses. The use of semiconductor diode lasers as sources of optical energy is attractive for a number of reasons. For example, diode lasers have a relatively small volume and consume a small amount of power as compared to conventional laser devices. Further, as monolithic devices, they do not require a combination of a resonant cavity with external mirrors and other structures to generate a coherent output laser beam. One disadvantage of the semiconductor diode laser, however, is the relatively low power of the output beam, as compared to other types of laser devices.
Group III-V (“III≧V”) semiconductor materials have been used to construct semiconductor lasers. Processing of III-V semiconductor devices includes vital steps for depositing III-V materials on a semiconductor substrate. For the deposition of a thick III-V layer, the lattice constant of the substrate material has to be very close to that of the deposited III-V layers (epi layers) with the same crystalline structure. Otherwise, crystalline defects, especially threading dislocations, will form during material deposition. When the defect density in the deposited material is high, it will significantly degrade device performance. These threading dislocation defects can create leakage paths for current, provide undesired carrier recombination centers and reduce device lifetime.
It is thus very difficult to grow high quality thin film materials on conventional prior art substrates with a large lattice mismatch. This lattice-matching requirement for compound semiconductor material deposition severely limits the possible choice of compound semiconductor material compositions and device material structure designs due to the limited choice of available substrates with the appropriate crystalline structures and lattice constants. Such substrates include Si, GaAs, InP, GaSb, InAs, and sapphire, inter alia.
For material systems for which there are no lattice-matched prior art substrates, however, some alternative approaches have been used. E.g., either a thick buffer layer is grown on the substrate, as proposed in U.S. Pat. No. 5,285,086, or a special technique, such as the lateral growth method proposed by Parillaud et al.,
Appl. Phys. Lett
. vol. 68 (1996), p. 2654, is employed before the growth of the device structure layers. It is known that defects, in particular threading dislocations, induced by lattice mismatch can be reduced from 10
11
/cm
2
to 10
5
/cm
2
by using the lateral growth method, for example. However, lattice-mismatched material growth techniques that result in defects, especially threading dislocation defects, often cause undesirable performance or characteristics of optoelectronic or electronic devices grown with such techniques.
It is desirable to epitaxially fabricate a variety of types of structures or devices, using a given epi material system, grown on a given substrate. Such epitaxially fabricated devices include electronic devices, such as transistors and integrated circuits, and optoelectronic devices, such as semiconductor lasers, light-emitting diodes, and photodetectors.
One such optoelectronic device in which there has recently been an increased interest is the vertical-cavity surface-emitting laser (VCSEL). The conventional VCSEL has several advantages, such as emitting light perpendicular to the surface of the die, and the possibility of fabrication of two dimensional arrays. VCSELs typically have a circular laser beam and a smaller divergence angle, and are therefore more attractive than edge-emitting lasers in some applications. Long infra-red spectrum wavelength (e.g., the range from approximately 1.2 &mgr;m to approximately 1.8 &mgr;m, including closely-spaced wavelengths around 1.3 &mgr;m or closely-spaced ITU grid wavelengths around 1.55 &mgr;m) VCSELs are also of great interest in the optical telecommunications industry because of the minimum fiber dispersion at 1.32 &mgr;m and the minimum fiber loss at 1.55 &mgr;m. The dispersion shifted fiber will have both minimum dispersion and minimum loss at 1.55 &mgr;m. The long wavelength VCSEL is typically based on an In
x
Ga
1−x
As
y
P
1−y
active layer lattice matched to InP cladding layers.
The structure of a typical VCSEL usually consists of an active region sandwiched between two distributed Bragg reflector (DBR) mirrors, as shown schematically in FIG.
1
. For the fabrication of long wavelength (e.g., 1.3 or 1.55 &mgr;m) VCSELs, it is very difficult to form the desired materials in one single growth step on a substrate. For instance, it is difficult to grow either the desired 1.3 &mgr;m active region on a GaAs substrate or to grow proper DBR mirrors on an InP substrate, despite the maturity of the technology for growing the DBR structure on GaAs substrates. Likewise, it is difficult to grow a 1.3 &mgr;m wavelength DBR structure on an InP substrate, despite the maturity of the technology for growing the active region. Recently, some alternative material systems, such as InGaNAs, GaAsSb and InGaAs quantum dots, have been developed to grow directly on a GaAs substrate using an Al
x
Ga
1−x
As/Al
y
Ga
1−y
As DBR for a 1.3 &mgr;m wavelength active region. However, these material systems are very difficult to grow and not easy to reproduce.
Another alternative approach to fabricate a long wavelength VCSEL is by using the so-called wafer bonding technique. However, this approach requires at least two to three wafer growth and one to two wafer-to-wafer bonding processes, which leads to very high fabrication cost and very low device yield. Therefore, a single wafer growth approach would be preferable to the wafer bonding approach, other considerations being equal.
One alternative approach to fabricate a long wavelength VCSEL with a single wafer growth step is to use the (In,Ga,Al)As material system lattice matched to In
x
(Al
y
Ga
1−y
)
1−x
As, where, e.g., 0.15<x<0.45, and growth of an InAlGaAs/InAlAs DBR structure and a moderately strained InGaAs quantum well (QW) structure active region. (Depending on the value of x, y is selected such that the material utilized has a bandgap absorption edge less than the lasing wavelength, e.g. less than 1.3 &mgr;m for a 1.3 &mgr;m VCSEL.) However, there is no commercially available substrate that is lattice matched to this material system. It is very difficult to control the composition precisely of a ternary In
x
Ga
1−x
As substrate uniformly over a whole wafer. Therefore, a high quality alternative substrate needs to be developed for this application.
One approach is to create a substrate that has the same crystalline structure and the same surface lattice constant as those of non-strained In
x
(Al
y
Ga
1−y
)
1−x
As, where 0.15<x<0.45. Another approach is to make a substrate that has a thin layer that is physically attached to the substrate, but can freely expand in a direction parallel to the substrate surface during material growth. This thin surface layer must have the same crystalline structure and a similar lattice constant as those of non-strained In
x
(Al
y
Ga
1−y
)
1−x
As, where 0.15<x<0.45.
For lattice-mismatched epitaxial layers, it is widely accepted that there exists a critical thickness beyond which misfit dislocations are introduced causing the breakdown of coherence between the substrate and epitaxial layers. The relaxation mechanism for lattice-mismatched epilayers known as the
Applied Optoelectronics, Inc.
Gruss Mason A.
Kinsella N. Stephan
Nguyen Joseph
Thomas Tom
LandOfFree
Alternative substrates for epitaxial growth does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Alternative substrates for epitaxial growth, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Alternative substrates for epitaxial growth will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3358253