Semiconductor device manufacturing: process – Making device or circuit emissive of nonelectrical signal – Mesa formation
Reexamination Certificate
2000-01-18
2002-03-12
Christianson, Keith (Department: 2813)
Semiconductor device manufacturing: process
Making device or circuit emissive of nonelectrical signal
Mesa formation
Reexamination Certificate
active
06355497
ABSTRACT:
FIELD OF THE PRESENT INVENTION
The present invention relates to the art of epitaxially grown semiconductors. It finds specific application in the growth of Group III-nitride laser diodes and light emitting diodes (LEDs) and will be described with particular reference thereto. It will be appreciated, however, that the invention is also amenable to other semiconductor devices and integrated circuits.
BACKGROUND OF THE PRESENT INVENTION
The data storage capacity of an optical data storage device, such as a compact disk read only memory (CD ROM) or a digital video disk (DVD), is limited by the wavelength of light used for reading/writing data to/from the storage device. If shorter wavelength light is used, more data may be stored on the storage device because it is possible to “pack” the data in a tighter fashion. Until recently, the light sources for reading/writing data to/from optical data storage devices produced light having relatively long wavelengths (i.e., light in the red and infra red regions of the light spectrum). New laser diodes and light emitting diodes (LEDs), are being developed for use in optical data storage devices. These new laser diodes and LEDs produce light having relatively short wavelengths (i.e., light in the blue, violet, and ultra violet regions of the spectrum). These new light sources have great potential in many areas such as, high resolution full-color printing, advanced display systems, optical communications, electronic device, and high-density optical storage.
One promising group within these new light sources are those based on crystals of Group III-nitrides (e.g., aluminum gallium indium nitride (AlGaInN)). However, progress in developing such Group III-nitride devices has been hampered by difficulties in separating films from the base substrates they are grown on, and by difficulties in producing defect free crystals on which to grow the devices.
A perfect crystal is a form of matter comprised of a regularly repeating arrangement of atoms. The regular repeating nature of the internal arrangement of atoms in a crystal is often apparent to the unaided eye. The plane faces or facets of a crystal, such as a quartz crystal or a sugar crystal, are the result of the regular repeating arrangement of its atoms. Imperfections, or interruptions in that regular atomic pattern, are often visible as well (e.g., when two crystals grow out of one another).
The properties of semiconductor devices stem from the properties of their underlying component crystals. Imperfections or irregularities in the crystals that make up a semiconductor device, at least in some cases, lead to reduced performance characteristics, such as a reduced tolerance to heat, or a shortened operating life time. Laser diodes and LEDs are examples of devices that are adversely affected by imperfections in their component crystals.
The preferred method used to make the new Group III-nitride devices is referred to as “epitaxial growth.” Epitaxy is the growth, on a crystalline substrate, of a crystalline substance that mimics the orientation of the atoms in the substrate. The most common substrate for the growth of Group III-nitride light sources known up until recently has been sapphire.
Directly growing Group III-nitrides on sapphire, however, has been found to result in a material having a very large defect density (e.g., approximately 10
10
/cm
2
). Bulk gallium nitride (GaN) is a better substrate than sapphire for growing Group III-nitride semiconductors. However, methods for growing bulk GaN are problematic. Some require working at high pressures and have not been successful. Other methods, using epitaxial lateral overgrowth (ELOG) techniques to grow GaN films, typically result in the creation of suture defects roughly in the center of what would otherwise be a desirable low defect density GaN film. Furthermore, it is difficult to separate the devices from the base substrates they are grown on.
The detrimental effect of suture defects in the standard ELOG technique is illustrated in
FIG. 1. A
GaN nucleation layer
12
covers a base sapphire substrate
10
. An SiO
2
mask has windows
16
for allowing nucleation and vertical GaN crystal growth. The process of creating the windows
16
in the SiO
2
mask also creates mesas
20
of SiO
2
. The mesas
20
prevent GaN nucleation. During GaN film growth, high defect density GaN
22
grows vertically in the windows
16
. The GaN
22
that grows in the windows
16
has a high defect density because it takes on the defect pattern of the underlying nucleation layer
12
. The GaN nucleation layer
12
has a high defect density because of the chemical and lattice mismatch with the base sapphire substrate
10
. The base sapphire substrate
10
is not a perfect epitaxial substrate for GaN, though it is among the best available.
As the high defect density GaN
22
growth reaches the top of the mesas
20
, it begins to laterally overgrow the mesas
20
. The mesas
20
block the dislocations of the underlying GaN nucleation layer
12
. Therefore, the GaN that overgrows the mesas
20
is relatively free of vertical defects, and, therefore, constitutes a low defect density GaN film
24
.
Lateral crystal growth is accompanied by continued vertical crystal growth. In order to have a reasonable final film thickness, it is necessary to use a series of windows in the SiO
2
mask As the lateral growth fronts of crystals started from adjacent windows coalesce, dislocations, or irregularities in the pattern of the atoms that make up the crystal, are created, and detrimental suture defects
26
are formed.
These detrimental suture defects
26
effectively cut the usable low defect density area in half. Very accurate lithographic techniques are then required in order to use the low defect area that is produced. Furthermore, one way to separate Group III-nitride devices from sapphire substrates is by laser ablation. Separation by laser ablation requires the use of a laser homogenizer and a stepper to move the beam around the substrate. Very accurate lithography and laser ablation techniques are slow and expensive. A better technique is needed for providing bulk substrates that are nearly lattice matched to III-nitride materials for epitaxial growth of semiconductor devices. Furthermore, a simpler and less expensive method for separating newly grown Group III-nitride film from its base substrate is also needed.
The present invention takes advantage of the fact that the Group III-nitrides and other films of interest, are impervious to most mask/release layer material etchants and provides a new and improved method for releasing films from substrates. Furthermore it provides a new and improved method for creating a suitable substrate for epitaxially growing Group III-nitride semiconductor devices. Therefore, it also provides new and improved Group III-nitride semiconductor devices.
SUMMARY OF THE INVENTION
One aspect of the present invention is a method for separating a film from a base substrate. The method comprising the steps of: depositing a release layer material above the base substrate for forming a release layer; growing a film over the release layer; and, etching the release layer with an etchant to separate the film from the base substrate.
Another aspect of the present invention is a method for the fabrication of a semiconductor device. The method comprises the following steps: Growing a nucleation layer on a base substrate; Depositing a release layer over the nucleation layer; Manipulating the release layer, providing points of access to the nucleation layer for uses as a seed crystal for a film, and blocking defects in the nucleation layer from propagating into at least one region of the film; Growing the film, producing at least one low defect density region in the film large enough for use as a substrate for growing a semiconductor device; Growing at least one semiconductor device on the low defect density region of the film; Removing the substrate and nucleation layer from the rest of the wafer; Applying appropriate contact metallization; and cleavin
Chua Christopher L.
Johnson Noble M.
Krusor Brent S.
Romano Linda T.
Walker Jack
Christianson Keith
Fay Sharpe Fagan Minnich & McKee LLP
Xerox Corporation
LandOfFree
Removable large area, low defect density films for led and... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Removable large area, low defect density films for led and..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Removable large area, low defect density films for led and... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2860887