Active solid-state devices (e.g. – transistors – solid-state diode – Responsive to non-electrical signal – Physical deformation
Reexamination Certificate
2000-04-07
2003-02-11
Tran, Minh Loan (Department: 2826)
Active solid-state devices (e.g., transistors, solid-state diode
Responsive to non-electrical signal
Physical deformation
C257S076000, C257S077000, C257S625000
Reexamination Certificate
active
06518637
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates generally to aluminum nitride (AlN) and more particularly, to epitaxial cubic (zinc-blende) AlN films that may have a thickness on the order of 1000 Å or greater and a method of making same by plasma source molecular beam epitaxy (PSMBE).
2. Description of the Related Art
The Group III-V nitride semiconductors (GaN, AlN, and InN) are of great interest for their potential as optoelectronic materials. These materials have an equilibrium crystal structure which is wurtzite, or hexagonal. The bandgaps of the wurtzite nitride semiconductors are all direct and their alloys have a continuous range of direct bandgaps values ranging from 1.9 eV for InN to 4.0 eV for GaN to 6.2 eV for AlN. As optical materials, these semiconductors are active from the orange into the ultraviolet.
Formation of nitride semiconductors for device applications requires, among other things, achieving the correct stoichiometry, inducing the correct energy to form a highly crystalline matrix, maintaining high purity, and matching the lattice parameters of the semiconductor and the substrate. Much effort was expended in the 1960's and 1970's to grow and characterize Group III-V nitride semiconductors. However, the effort was ineffectual to achieve high-quality material. Recently, there has been renewed effort to create higher quality Group III-V nitride semiconductors. However, GaN, AlN, and InN produced by conventional methods have high n-type background carrier concentrations resulting from native defects commonly thought to be nitrogen vacancies. Nitrogen vacancies affect the electrical and optical properties of the film. Oxygen contamination is also a major problem. Thin layers of AlN have been prepared by magnetron sputtering, chemical vapor deposition, ion beam sputtering, and ion beam assisted deposition. However, these methods operate at elevated temperatures and generally do not result in epitaxial growth (i.e., growth oriented in one direction). Moreover, while these techniques have been successful in producing polycrystalline AlN films, they have not been successful in producing electronic-grade single crystal films.
AlN, in particular, is a promising material for high-power, high-temperature optoelectronic devices since it has very high chemical and thermal stability, good thermal conductivity, and fast Rayleigh velocity. AlN crystallizes, under normal conditions, into the thermodynamically stable hexagonal wurtzite structure. However, the metastable cubic zinc-blende structure is expected to be easier to dope and to have decreased phonon scattering, and therefore, to have higher ballistic electron velocities, thermal conductivity, and acoustic velocities due to its higher symmetry. These properties give rise to many exciting potential device applications.
There have been several reports of AlN having the metastable cubic zinc-blende structure. These reports, however, lack detail on the physical, electrical, and optical properties of cubic zinc-blende AlN because the films were too thin for such studies, and certainly too thin to be useful for optoelectronic devices which require thicknesses on the order of at least 2000 Å, and preferably 4000 Å to 8000 Å. The lattice constant of zinc-blende AlN was calculated theoretically to be 4.38 Å using data from the elastic constants of wurtzite AlN. This value was later confirmed experimentally on a 12 nm thick film of zinc-blende AlN grown pseudomorphically on cubic TiN sandwiched between a tetragonal Al
3
Ti overlayer. To date, however, there have been no reports of successful fabrication of thick, device-quality films of zinc-blende AlN. The known AlN films have been mixed hexagonal and possibly cubic (which could be the rock salt structure).
It is, therefore, an object of the invention to prepare zinc-blende AlN of sufficient quality and thickness to characterize it for its mechanical, optical, and electrical properties and to be useful for device fabrication.
It is also an object of the invention to prepare device quality, single crystal, epitaxial films of cubic zinc-blende AlN.
It is a further object of the invention to produce a semiconductor devices that include an epitaxial film(s) of single crystal zinc-blende AlN.
It is an additional object of this invention to provide a method of making an epitaxial film of zinc-blende AlN.
SUMMARY OF THE INVENTION
The foregoing and other objects, features and advantages are achieved by this invention which is, in a first device embodiment, a film of device quality, single crystal cubic zinc-blende AlN. In other embodiments, the zinc-blende AlN film is deposited on a substrate, and preferably on a substrate having cubic symmetry on its surface, such as a silicon (100) wafer (Si(100)). In a particularly preferred embodiment, there is a buffer, or an interfacial, layer of cubic 3C—SiC between the epitaxial film of zinc-blende AlN and the substrate which may be a Si(100) wafer.
In a specific illustrative embodiment, a semiconductor device comprises a Si(100)-oriented substrate; an interfacial layer of 3C—Si(C) on the Si(100) substrate; and a film of single crystal zinc-blende AlN having a thickness of at least 800 Å, and preferably in the range of 1000 Å to 2000 Å, which is epitaxial with respect to the Si(100) substrate. The epitaxial relationship between film and substrate is (100)AlN∥(100)Si and [101]AlN∥[101]Si. The interfacial layer may have a thickness ranging from several atomic layers (e.g., about 25-30 nm) and up.
In accordance with the principles of the invention, the metastable zinc-blende form of AlN is deposited on the substrate by a plasma beam of aluminum ions and activated nitrogen ion species produced in a molecular beam epitaxy system by applying a pulsed d.c. power to a hollow cathode source. In this manner, films having a thickness of at least 800 Å were produced. Thickness, of course, is a function of deposition time, and films ranging from 10 Å to several microns, are possible by the method of the present invention. The lattice parameter of the as-deposited films was calculated to be approximately 4.373 Å which corresponds closely to the theoretical calculation (4.38 Å) for cubic zinc-blende AlN.
The zinc-blende AlN epilayer films of the present invention have a wide bandgap (experimentally determined to be about 5.34 eV); thermal stability (up to about 800° C.), and extraordinary piezoelectric properties. In addition to the foregoing, the films have been of sufficient quality to enable experimental confirmation that zinc-blende AlN is an indirect semiconductor. When characterized in situ by Reflection High Energy Electron Diffraction (RHEED), the films show four-fold symmetry rather than the six-fold symmetry which is typical for hexagonal AlN in (0001) orientation. Furthermore, the RHEED patterns appear to be very similar to those for the Si(001) substrates, except for different streak spacings. X-ray diffraction (XRD) revealed broad peaks at diffraction angle (2&thgr;) values of approximately 41° and 89.8°. These peaks match the (002) and (004) peaks of zinc-blende AlN with a lattice parameter of 4.38 Å. Transmission electron microscopy (TEM) confirmed that the AlN produced by the method of the present invention is cubic, single crystal and epitaxial with respect to the Si(100) substrate.
In accordance with the invention, the growth surface of the substrate should preferably have a cubic structure to act as a template for cubic (zinc-blende) growth. Specific illustrative examples include, but are not limited to Si(100) and magnesium oxide (MgO (100)). Of course, the substrate may comprise one or more layers. Preferably, the growth surface of the substrate should have an good lattice match with AlN. A good lattice match is defined as being within about 1%. However, as is known in the art, substrates with seemingly poor lattice matches (e.g, Si(100) has a 19% mismatch), may be used since epitaxially dep
Auner Gregory W.
Thompson Margarita P.
Rohm & Monsanto, PLC
Tran Minh Loan
Wayne State University
LandOfFree
Cubic (zinc-blende) aluminum nitride does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Cubic (zinc-blende) aluminum nitride, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Cubic (zinc-blende) aluminum nitride will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3138005