High-K dielectric materials and processes for manufacturing...

Details High-K dielectric materials and processes for manufacturing... High-K dielectric materials and processes for manufacturing...

2002-07-02

2004-09-07

Nhu, David (Department: 2818)

Semiconductor device manufacturing: process

Making passive device

Planar capacitor

C438S239000

Reexamination Certificate

active

06787429

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates generally to semiconductor devices. More particularly, the invention pertains to materials with high dielectric constants and methods for incorporating them in semiconductor devices.
2. State of the Art
In the manufacture and use of integrated circuit (IC) devices, new applications continually drive the development of devices with enhanced miniaturization and increased circuit density. Current and future developments in reducing the size of dynamic random access memory (DRAM) devices and the like result in a need for storage capacitor materials having higher dielectric constants than currently available.
Capacitor cells are generally formed as “stacked” capacitors, i.e., positioned above the working surface of the chip or wafer, or “trench” capacitors, which are formed in a trench in the wafer or chip substrate. Because of the need to make the best use of available space in a device, current capacitor designs include non-planar structures which may be formed in various configurations. References which describe examples of non-planar capacitor constructions include U.S. Pat. No. 5,981,333 to Parekh et al., U.S. Pat. No. 5,981,350 to Geusic et al., U.S. Pat. No. 5,985,714 to Sandhu et al., and U.S. Pat. No. 5,985,732 to Fazan et al., each of which is incorporated herein by reference.
The number of high dielectric materials from which capacitor cells may be satisfactorily formed is limited. Insulating inorganic metal oxide materials such as ferro-electric or perovskite material have high dielectric constants and generally low leakage current. However, these materials require a step of “oxidation-densification” to produce the desired dielectric capacitor layer. Unfortunately, such oxidation-densification undesirably oxidizes the underlying electrode of conductively doped polysilicon. As practiced currently, an intervening oxygen-barrier layer is placed between the electrode and dielectric material. The barrier layer must be electrically conductive, inasmuch as the underlying polysilicon must be in electrical connection with the dielectric layer. The materials which may be used as oxygen barrier layers are limited in number. Elemental platinum on polysilicon has been suggested as a barrier layer for a lower capacitor plate but undergoes physical degradation with thermal cycling due to silicon diffusion through the platinum. Sputtered TiN and CVD-applied TiN have been known to fail due to diffusion along grain boundaries.
As known in the art, an alloy of titanium and tungsten may be used as a barrier layer between a silicon layer and an aluminum ohmic contact, where the junction is very shallow, i.e., less than about 0.2 &mgr;m.
In U.S. Pat. No. 5,985,714 having patentees of Sandhu et al. and of even assignment with this application, a condenser construction is described which uses a wide variety of dielectric materials including titanates of barium; barium and strontium; strontium; lead; barium and lead; lead and zirconium; lead and lanthanum; lead and lanthanum and zirconium; and bismuth. Lithium tantalate is also mentioned.
Several materials which have been used or undergone evaluation include Ta
2
O
5
and (Ba, Sr)TiO
3
, the latter commonly known as BST. Ta
2
O
5
has a dielectric constant k which is about 15 to 25; the dielectric constant is too low to meet the requirements for use in advanced DRAM and other semiconductor construction, i.e., a much higher dielectric constant generally exceeding about 100.
BST materials have dielectric constants, i.e., about 300-600, which are higher than dielectric materials in current use. However, the processes for producing BST are not yet fully developed. The processing of BST is intrinsically difficult because of the low volatility of the precursors used in the chemical vapor deposition (CVD) step, and by difficulty in controlling the complex stoichiometry to maintain the desired material characteristics.
Alternative dielectric materials have appeared to offer potential advantages in dielectric constant value and ease of manufacture. For example, TiO
2
films are well known as high dielectric materials. TiO
2
films have a dielectric constant greater than 100, which is considerably higher than that of Ta
2
O
5
. In addition, TiO
2
films may be formed using current manufacturing methods. However, it has been found that capacitors made of pure TiO
2
have a high leakage current unacceptable in high-density devices required by current and developing electronic technology.
It has been shown by Matsushita (
Jpn. J. Appl. Phys.
30 (1991) 3594) that doping TiO
2
with SiO
2
may dramatically improve the leakage current of the TiO
2
materials used in capacitors. However, this doped material is generally comparable to Ta
2
O
5
in dielectric constant, i.e., in a low range of about 15-25.
Other materials considered for high dielectric use include tungsten trioxide (WO
3
) but it has an unacceptably high leakage current.
Commercial production of semiconductor devices requires a sequence of basic physical/chemical processes, many of which are typically performed on a large number of dice in a semiconductor wafer prior to singulating and packaging the devices. The minimal time required to carry out the process from beginning to end is extensive, with high attendant cost. For example, it usually takes about 6-8 weeks or more to produce a potentially finished memory chip from an uncut multi-wafer crystal. It is desirable to shorten the processing time as much as possible, to reduce manpower cost and increase the throughput rate of processing equipment.
The instant invention addresses the need for new dielectric materials having high dielectric constants (K) of about 100 or more, and the capability of being processed more quickly, easily and precisely, and at lower cost than other high dielectric material candidates.
BRIEF SUMMARY OF THE INVENTION
In accordance with the invention, dielectric materials are presented which have dielectric constants greater than either Ta
2
O
5
or SiO
2
-doped TiO
2
, have low leakage current, and may be prepared using cost-effective deposition and annealing processes.
In this invention, materials for use in making integrated circuit devices are formed of the oxides of mixed transition metals such as titanium plus tungsten, or titanium plus tantalum. To form a capacitative cell, the transition metals are deposited on a conductive plate (electrode) such as doped polysilicon, and annealed and oxidized under controlled conditions. The resulting materials have a dielectric constant k greater than either Ta
2
O
5
or SiO
2
-doped TiO
2
, and an oxygen barrier is integrally formed to prevent oxidation of the underlying doped polysilicon.
These new materials may be readily prepared through a chemical vapor deposition (CVD) process, which provides excellent conformality, an important advantage in the manufacture of capacitor cells with non-planar structures. Unlike the problems in preparing BST, all precursors used for deposition of the new materials by CVD are volatile and easily used. Use of the high dielectric materials to form capacitor cells enables further density increases in DRAM and other devices.
The high dielectric materials presented herein may be readily used to produce capacitative structures ranging from simple two-plate cells to multi-plate stacked non-planar cells.


REFERENCES:
patent: 4152490 (1979-05-01), Witzke
patent: 4585312 (1986-04-01), Ishiwata et al.
patent: 5189503 (1993-02-01), Suguro et al.
patent: 5195018 (1993-03-01), Kwon et al.
patent: 5225286 (1993-07-01), Fujikawa et al.
patent: 5290609 (1994-03-01), Horiike et al.
patent: 5362632 (1994-11-01), Mathews
patent: 5387480 (1995-02-01), Haluska et al.
patent: 5589733 (1996-12-01), Noda et al.
patent: 5719425 (1998-02-01), Akram et al.
patent: 5719607 (1998-02-01), Hasegawa et al.
patent: 5793057 (1998-08-01), Summerfelt
patent: 5841186 (1998-11-01), Sun et al.
patent: 5859760 (1999-01-01), Park et al.
patent: 5866460 (1999-02-01), Akram et al.
patent: 5910880 (1999-06-01

Affiliated with

Also associated with

Rating

High-K dielectric materials and processes for manufacturing... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with High-K dielectric materials and processes for manufacturing..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and High-K dielectric materials and processes for manufacturing... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3227879