Semiconductor device manufacturing: process – Coating with electrically or thermally conductive material – To form ohmic contact to semiconductive material
Reexamination Certificate
2001-07-17
2002-10-01
Nelms, David (Department: 2818)
Semiconductor device manufacturing: process
Coating with electrically or thermally conductive material
To form ohmic contact to semiconductive material
C438S680000
Reexamination Certificate
active
06458698
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates generally to methods of forming electronic devices, to methods of forming electro-mechanical devices, to methods of forming force-sensing devices, to methods of forming accelerometers, to electronic devices, to electro-mechanical devices, to force-sensing devices, and to accelerometers.
BACKGROUND OF THE INVENTION
Electronic devices and electro-mechanical devices are devices which can include both electrical and mechanical parts. Sometimes, because of the nature of these parts, the way they interact with one another, and the environment within which they are used, the parts can be adversely affected. Exemplary electronic and electro-mechanical devices in the form of force-sensing transducers are described in U.S. Pat. Nos. 5,367,217, 5,339,698, and 5,331,242, the disclosures of which are incorporated by reference. Exemplary accelerometers are described in U.S. Pat. Nos. 5,594,170, 5,501,103, 5,379,639, 5,377,545, 5,456,111, 5,456,110, and 5,005,413, which are incorporated by reference herein. Other types of electronic and electro mechanical devices can be used as rate sensors. Exemplary rate sensors are described in U.S. Pat. Nos. 5,717,140, 5,376,217, 5,696,323, 5,691,472, and 5,668,329, which are incorporated by reference herein. Yet other electronic or electro-mechanical devices can be used as acceleration and rate sensors. Exemplary sensors are described in U.S. Pat. Nos. 5,627,314, 5,557,046, 5,341,682, 5,331,853, 5,331,854, and 5,319,976, the disclosures of which are incorporated by reference herein.
One type of electronic or electro mechanical device is an accelerometer. Accelerometers can have mechanical parts which experience acceleration forces, and electrical parts to help set up conditions conducive to sensing such forces, and to assist in sensing such forces. Yet, these electrical and mechanical parts can sometimes experience problems which adversely impact their performance.
For example, micromachined silicon accelerometers can exhibit undesirable resonator frequency drift at temperatures below 200° C. and even at room temperature. This drift is believed to be due to “creep” or plastic deformation in a current-carrying, evaporated chromium/gold or chromium/molybdenum/gold metallization layer following 400° C. wafer bonding in air. This drift can degrade the performance of a device and can limit its accuracy and its marketability.
The frequency drift can be directly attributed to structural changes which take place in the top, thicker gold layer (typically around 0.2-0.7 micron in thickness). The plastic deformations and creep in the gold layer are due to a low Young's modulus, low hardness and low tensile strength of gold. A much thinner chromium layer (around 100 Angstrom) is used to improve the adhesion of the gold or molybdenum to an underlying thermal silicon dioxide (SiO2) and does not appreciably contribute to the overall resistance or current-carrying capacity. In addition to the above-described creep, gold and chromium, or gold and molybdenum in thin-film form show significant undesirable interdiffusion among themselves in the above temperature range. This can result in an undesirable increased resistivity and visual splotchiness in an otherwise uniform mirror-like appearance of the layers.
Accordingly, there is a need for a more stable metallization, and one which can mitigate some of the problems experienced in electronic or electro-mechanical devices having mechanical and electrical parts. Particularly, there is a need within the context of accelerometers for stable metallization which would mitigate resonator frequency drift, and which would have reduced or no interdiffusion when in contact with chromium or other adhesion-promoting layers. Further, there is a need for materials which exhibit stable resistivity and visual appearance. In addition, an improved metallization scheme should desirably exhibit relatively low electrical sheet resistance, e.g. about 0.25 ohm/square at a total thickness of less than one micron, and good bondability to gold wire. Stress in the metallization should also be desirably as low as possible.
In parent application Ser. No. 09/187,288, the inventors of the present invention described novel metallizations using iridium or rhodium, which are significantly more thermally stable than the above gold or molybdenum/gold metallizations, as the current-carrying conductive material. Each of the iridium and rhodium metallizations was separately and individually deposited on a thin chromium adhesion layer previously deposited on an oxidized silicon wafer substrate. The iridium or rhodium metallizations were deposited using physical vapor deposition in high vacuum. Each of the iridium and rhodium metallizations was shown to possess several desirable characteristics useful for accelerometers, rate sensors and other electronic and electro-mechanical devices, including relatively low and thermally stable sheet resistance for a total metallization thickness smaller than 1 micrometer and no interdiffusion with the underlying chromium adhesion layer after annealing at 400° C. in air. Deposition parameters such as the substrate temperature, metal film thickness and metal deposition rate were varied, and preferred deposition conditions were identified which resulted in the lowest sheet resistance and lowest residual stress.
Although preferred deposition conditions result in the lowest sheet resistance and lowest residual stress, even under optimized deposition conditions, such iridium and rhodium metallization films retained a level of residual tensile stress in the range of about 300 MPa to 1000 MPa at room temperature. This residual tensile stress limits the practical upper thicknesses for such metallization films, and therefore may also limit the lower sheet resistances attainable. High tensile stress in the metallization film in excess of the film's ultimate strength may render the film useless for an intended device through delamination from the substrate or microcracking,. The stress related force on the metallization film increases with film thickness. Residual stresses in thin films may in some cases result in increased rates of failures due to metal fatigue, delamination and microcracking, such effects increasing as a function of increasing film thickness. An improved deposition process which enables the formation of such stable metallization thin films with lower stress, and ideally with much lower stress than 300 MPa, is therefore highly desirable.
The inventors are familiar with efforts to bombard films with ions during deposition. To date, such efforts have been restricted primarily to optical materials rather than current carrying metallization films. The work in ion bombardment of optical materials is not applicable to metallization films. Optical films are typically electrically insulating and optically transparent and usually formed of an oxide, a nitride or a compound which is transparent in the visible or infrared spectrum, rather than a current carrying metal, which is in general optically opaque. Some transparent electrically conductive film materials, such as Indium-Tin-Oxide and Indium-Zinc-Oxide, are known, but these materials are oxides and have electrical conductivities far lower than those of metal films, such as gold, iridium, and rhodium. Furthermore, such ion bombardment in optical materials usually damages electrical conduction properties.
Ion bombardment during deposition in the integrated circuit industry has ignored the current carrying metallization films other than those used in integrated circuits. In particular, the inventors are experts in the field of metallization film deposition and are not aware of efforts to reduce or eliminate stress while retaining or improving electrical conductivity in a metallization film formed of metals, such as iridium, rhodium, tungsten, and osmium, which are not generally used in the integrated circuit industry. In contrast to the prior art, what is needed is a deposition method that significantly reduces or c
Eagan Margaret
Golecki Ilan
Dang Phu T.
Honeywell International , Inc.
Nelms David
LandOfFree
Controlled-stress stable metallization for electronic 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 Controlled-stress stable metallization for electronic and..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Controlled-stress stable metallization for electronic and... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2998805