Ammonia gas passivation on nitride encapsulated devices

Details Ammonia gas passivation on nitride encapsulated devices Ammonia gas passivation on nitride encapsulated devices

2000-08-30

2003-04-08

Chaudhuri, Olik (Department: 2823)

Semiconductor device manufacturing: process

Radiation or energy treatment modifying properties of...

Ionized irradiation

C438S792000

Reexamination Certificate

active

06544908

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates in general to methods for passivating semiconductor device structures during fabrication thereof and, more particularly, to methods for passivating semiconductor-insulator interfaces in semiconductor device structures.
2. State of the Art
Insulative structures of semiconductor devices have long been formed from silicon-containing materials, such as silicon dioxide and silicon nitride. Silicon dioxide structures are typically fabricated by forming a silicon layer over a semiconductor device structure and oxidizing the silicon layer or by deposition processes that employ materials such as tetraethylorthosilicate (TEOS). Silicon dioxide layers so formed are then patterned by known processes to define insulative structures of the semiconductor device.
Silicon nitride insulative structures may be formed by first forming a layer of silicon on a semiconductor device structure, then nitridating the layer of silicon so as to form a silicon nitride layer. Conventionally, nitridation of silicon has been effected by exposing the silicon to nitrogen-free radicals from sources such as nitric oxide (NO), nitrous oxide (N
2
O) and ammonia (NH
3
). Free radicals may be generated from these nitrogen-containing species by use of plasmas. Once a silicon nitride layer has been formed, the silicon nitride layer may be patterned by known processes to form insulative structures of the semiconductor device.
Conventional semiconductor devices typically include conductive lines with thicknesses of at least about 0.25 microns. The insulative structures of the semiconductor devices, which are fabricated by conventional processes, have comparable thicknesses and impart to the semiconductor device the desired dielectric properties.
The trend in the semiconductor industry is toward fabricating semiconductor devices including structures of ever-decreasing size. While the widths of conductive lines of state-of-the-art semiconductor devices are currently in the range of about 0.25 microns down to about 0.18 microns, the widths of conductive lines and, thus, of the insulative structures adjacent thereto, are continuing to decrease. The goal in the industry is to decrease the thicknesses of conductive lines and their adjacent insulative structures down to dimensions that are measurable in terms of a few molecules or even single molecules.
In semiconductor device structures, the silicon-oxide interface contains interface states and surface defects caused by unsatisfied chemical bonds, or “dangling silicon bonds.” Unsatisfied bonds in silicon atoms contribute to a charge at the oxide surface and the existence of the interface states cause the threshold voltage to fluctuate.
As the thicknesses of insulative structures of semiconductor devices decrease, dangling silicon bonds at interfaces between insulative structures and adjacent silicon structures, such as source/drain regions (i.e., n-wells and p-wells) and polysilicon conductive structures, become problematic. In conventional semiconductor devices, the dangling silicon bonds are present at concentrations of about 10
11
to about 10
12
per square centimeter. While these concentrations of dangling silicon bonds do not cause significant problems in conventional semiconductor devices, dangling silicon bonds may cause defects in much thinner insulative structures, which may cause electrical shorting in semiconductor devices including such thin insulative structures and, thus, failure of the semiconductor devices.
A common solution to this problem has been to passivate the interface by exposing the interface to high concentrations of molecular hydrogen (H
2
), or hydrogen gas, which acts as a source for hydrogen atoms (H). The passivation is effected by rapid thermal processing (“RTP”) and furnace tools before encapsulation with a thick nitride film. It is believed that the hydrogen atoms bond to the dangling silicon bonds to passivate the interface. Unfortunately, the use of H
2
raises serious safety concerns. The explosive nature of hydrogen gas narrows the window of acceptable process conditions involving its use. Argon or another inert gas can be mixed with H
2
to mitigate some of the safety risks, but this dilution reduces the overall rate of reaction, which results in unsatisfactory processing times. Additionally, hydrogen passivation must be done at the end of the processing of a semiconductor device structure. Otherwise, the hydrogen will escape from the interface region.
Moreover, the passivating hydrogen species are more readily driven from the passivated structures when a semiconductor device structure under fabrication is exposed to high process temperatures, such as processes that require temperatures of about 600° C. or greater.
Another method that has been proposed for passivating semiconductor device structures during fabrication thereof so as to prevent dangling silicon bonds from causing defects to be formed in insulative structures includes the use of deuterium species derived from molecular deuterium (D
2
). When deuterium is used to passivate the various structures, including insulative structures, of a semiconductor device structure under fabrication, deuterium species, including deuterium-free radicals, permeate the various structures of the semiconductor device structure. It has been further proposed that by encapsulating deuterium-passivated structures with a suitable material, such as silicon nitride, the concentration of deuterium that permeates, and thus, passivates the semiconductor device structures will not be significantly diminished when the semiconductor device structure under fabrication is exposed to high process temperatures. Further, deuterium has been used to form silicon nitride layers by reaction with silane or dichlorosilane (DCS), ammonia, and molecular deuterium. Nonetheless, the use of molecular deuterium poses many of the same threats as those present during the use of molecular hydrogen.
Thus, it can be appreciated that it would be advantageous to develop a technique for passivating the silicon-silicon dioxide interface by trapping a passivating species at the interface using a method that mitigates the problems present in the prior art.
BRIEF SUMMARY OF THE INVENTION
The present invention includes a method for passivating with hydrogen species derived from ammonia (NH
3
) semiconductor device structures under fabrication and, particularly, insulative structures of the semiconductor device structures. The method of the invention includes obtaining the hydrogen species from ammonia so as to allow the hydrogen species to permeate the semiconductor device structure under fabrication, as well as the insulative structures thereof, and forming an encapsulant layer over the semiconductor device structure so as to retain the hydrogen species within at least the insulative structures of the semiconductor device structure. The encapsulant may be formed following exposure of the semiconductor device structure to the hydrogen species or concurrently with exposure of the semiconductor device structure to the hydrogen species.
Hydrogen species, including hydrogen-free radicals, may be derived from ammonia by known processes. High temperature processes, for example, known rapid thermal processing (RTP) and other batch system techniques, may be employed. As hydrogen species from the ammonia permeate the semiconductor device structure, the concentration of dangling silicon bonds present at interfaces between silicon and silicon oxide structures is decreased from about 10
11
to about 10
12
dangling silicon bonds per square centimeter to about 10
9
to about 10
10
dangling silicon bonds per square centimeter, or to about 1% of the original concentration of dangling silicon bonds present at these interfaces. This represents a decrease in concentration of dangling silicon bonds of about two orders of magnitude.
As the passivating hydrogen species that permeate the semiconductor device structure, including the interfaces between the insulative and semico

Affiliated with

Also associated with

Rating

Ammonia gas passivation on nitride encapsulated devices does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Ammonia gas passivation on nitride encapsulated devices, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Ammonia gas passivation on nitride encapsulated devices will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3099798