Silicon-based functional matrix substrate and optical...

Details Silicon-based functional matrix substrate and optical... Silicon-based functional matrix substrate and optical...

1999-08-31

2001-05-08

Lee, Eddie C. (Department: 2815)

Active solid-state devices (e.g., transistors, solid-state diode

Incoherent light emitter structure

In combination with or also constituting light responsive...

C257S080000, C257S085000, C257S098000, C257S103000, C257S431000

Reexamination Certificate

active

06229159

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to a silicon-based functional matrix substrate and an optical integrated oxide device, especially suitable for use in optical integrated oxide electronics spread out on silicon.
2. Description of the Related Art
It is a well-known fact that oxide thin film materials have been remarkably developed in recent several years, starting from high-temperature superconductive oxides reported in 1986 ((1) Z. Phys. B., 64, 189-193(1986), (2) MRS Bulletin, XVII, No. 8, 16-54(1992), (3) MRS Bulletin, XIX, No. 9, 21-55(1994)).
On the other hand, memory devices using ferroelectric materials, which were energetically studied in a certain period of 1950s ((4) Electrical Engineering, 71, 916-922(1952), (5) Bell Labs. Record, 33, 335-342(1955)) but failed to penetrate into industries because of difficulties in, for example, controlling interfaces, have recently come to be highlighted, and researches and developments thereon have progressed rapidly. The current aspect of the ferroelectric nonvolatile memory devices were reported in detail (for example, (6) Appl. Phys. Lett., 48, 1439-1440(1986), (7) U.S. Pat. No. 4,713,157, (8) IEDM Tech. Dig., 850-851(1987), (9) IEEE J. Solid State Circuits, 23, 1171-1175(1988), (10) Tech. Dig. ISSCC 88, 130-131 (1988), (11) Applied Physics, Vol. 62, No. 12, 1212-1215(1993), (12) Electronic Ceramics, Vol. 24, July, 6-10(1993), (13) Electronic Materials, Vol. 33, No. 8 (1994) (Special Vol. entitled “Application of Ferroelectric Thin Films to Nonvolatile Memory”), (14) Ceramics, Vol. 27, 720-727(1992)).
It is needless to say so on oxide superconductive devices (see Literatures (2) and (3)), it is well known also that researches and developments have been proceeded recently on applications of oxide nonlinear optical devices and elements, as well. While fields of superconductive devices and ferroelectric nonvolatile memory devices are under remarkable development, in the field of optical devices, conciliation with lithographic techniques favorably used for silicon devices has not been prosecuted, as shown by the fact that bulk materials are still used, for example.
The Inventor, however, has recognized formerly that the importance of epitaxial thin films of oxides on silicon is not limited only to superconductive devices or ferroelectric nonvolatile memory device, and made some reports or proposals on oxide stacked structures made by stacking oxide thin films on silicon substrates and ferroelectric nonvolatile memory devices using them ((15) Japanese Patent Laid-Open No. Hei 8-330540, (16) Japanese Patent Laid-Open No. Hei 8-335672, (17) Japanese Patent Laid-Open No. Hei 8-340087, (18) Japanese Patent Application No. Hei 8-336158, (19) J. Ceram. Soc. Japan. Int. Edition, 103, 1088-1099(1995), (20) Mater. Sci. Eng. B., 41, 166-173(1966).
On the other hand, in optical integrated oxide electronics, it is required to integrate a semiconductor laser or other semiconductor light emitting device together with an oxide optical device or other oxide device on a common substrate. However, as far as the Inventor is aware, there has been almost no substantial report on concrete device structures. Especially on optical integrated oxide devices in which oxide elements are formed by epitaxial growth of oxides, no report has been heard of.
OBJECT AND SUMMARY OF THE INVENTION
It is therefore an object of the invention to provide a silicon-based functional matrix substrate for integrating thereon an oxide device, such as oxide optical device, ferroelectric nonvolatile memory and oxide superconductive device, and a semiconductor light emitting device, such as semiconductor laser, in an optimum structure with a high density. Another object of the invention is to provide an optical integrated oxide device using the silicon-based functional matrix substrate.
To attain the object, the Inventor made specific researches on optimum structures, materials, processes, and so forth, for integrating an oxide device, such as oxide optical device, ferroelectric nonvolatile memory and oxide superconductive device, and a semiconductor light emitting device, such as semiconductor laser, on a common substrate. These researches are summarized below.
Selected as the substrate is a single-crystal silicon substrate which is a proven basic material of semiconductor memory, inexpensive, easily available and excellent in crystallographic property.
To make an oxide device on a single-crystal silicon substrate, it would be advantageous to epitaxially grow an oxide thin film on the single-crystal silicon substrate. Usually, however, it is difficult to epitaxially grow an oxide thin film directly on a single-crystal silicon substrate. It is therefore considered that a buffer layer of a material in lattice match with a single-crystal silicon substrate be first grown epitaxially or oriented preferentially on the single-crystal silicon substrate, and an oxide thin film in lattice match with the buffer layer be thereafter grown epitaxially on the buffer layer. In this manner, a material most suitable for the oxide device to be made can be selected from a wide variety of oxides. The buffer layer is preferably made of an oxide to ensure epitaxial growth of an oxide thin film thereon. The buffer layer made of an oxide is preferably epitaxially grown directly on the single-crystal silicon substrate.
Oxide materials that can be epitaxially grown directly on a single-crystal silicon substrate are currently only five, namely, magnesium oxide (MgO), cerium oxide (ceria) (CeO
2
), &agr; alumina (&agr;-Al
2
O
3
), yttrium stabilized zirconium (YSZ) and magnesium aluminum spinel (MgAl
2
O
4
), as shown in Table 1. This does not mean that the other oxides have been proved not to epitaxially grow directly on single-crystal silicon substrates, and there might be one or more, other than the above five oxides, which can be epitaxially grown directly. Table 1 shows lattice constants (a, c) of oxide crystals and thermal expansion coefficients (&agr;) (some of the data relies on “Structure and Properties of Inorganic Solids” by F. S. Galasso (Int. Series of Monographs in Solid State Physics, Vol.7). The lattice constant and thermal expansion coefficient of silicon (Si) are a=0.5430884 nm and &agr;=3.0×10
−6
/K, respectively.
TABLE 1
Oxide
Type of Crystal
Lattice Constant
Crystal
Structure
(nm)
&agr;[10
−6
/K]
MgO
halite: NaCl
a = 0.4213
13.6
CeO
2
fluorite: CaF
2
a = 0.5411
 8.9
&agr;-Al
2
O
3
corundum: &agr;-Al
2
O
3
a = 0.476 c = 1.299
 8.3
YSZ
halite: CaF
2
a = 0.514 (a
P
= 3.63)
 7.6 (ZrO
2
)
MgAl
2
O
4
spinel: MgAl
2
O
4
a = 0.8083
 7.18
Among these oxide materials, MgO, CeO
2
, &agr;-Al
2
O
3
, YSZ and MgAl
2
O
4
, CeO
2
and MgAl
2
O
4
are most prospective because they are less subject to problems by diffusion of their component elements, and promise a higher possibility of epitaxial growth of perovskite oxides thereon. However, both have merits and demerits.
FIG. 1
shows dependency of lattice constants on temperature on CeO
2
and MgAl
2
O
4
together with dependency of the lattice constant of Si on temperature.
As shown in
FIG. 1
, CeO
2
is much more excellent than MgAl
2
O
4
from the viewpoint of lattice match with the single-crystal silicon substrate (when its surface orientation is (100)). However, as to crystallographic stacking alignment when lattice match is established, MgAl
2
O
4
was very easy for perovskite oxide (ABO
3
) to be stacked thereon, but there was a serious technical bar in front of CeO
2
, as shown in
FIGS. 2 and 3
.
That is, there are many reports stating that CeO
2
(100) does not epitaxially grow even on Si(100), but CeO
2
(110) epitaxially grows thereon. Actually, almost all on CeO
2
among these reports conclude that nothing but CeO
2
(110)/Si(100) structure is obtained ((21) Appl. Phys. Lett., 59, 3604-3606(1991)). Therefore, in the crystallographic stacking alignment, it is considered difficult to epitaxia

Affiliated with

Also associated with

Rating

Silicon-based functional matrix substrate and optical... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Silicon-based functional matrix substrate and optical..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Silicon-based functional matrix substrate and optical... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2481437