Metal working – Method of mechanical manufacture – Electrical device making
Reexamination Certificate
2002-08-29
2004-12-14
Tugbang, A. Dexter (Department: 3729)
Metal working
Method of mechanical manufacture
Electrical device making
C029S025420, C029S025410, C438S033000, C438S042000, C438S045000, C381S170000
Reexamination Certificate
active
06829814
ABSTRACT:
CROSS REFERENCE TO RELATED APPLICATIONS
Not applicable.
STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH
Not applicable.
BACKGROUND OF THE INVENTION
(1) Field of the Invention
The present invention generally relates to semiconductor micromachined devices. More particularly, this invention relates to a process of making an all-silicon capacitive microphone, in which a membrane used to sense sound vibration is formed of substantially stress-free single-crystal silicon bonded to a support substrate.
(2) Description of the Related Art
There is a continuing desire for audio devices that are smaller in size, lower in cost, and can be manufactured using high-volume manufacturing practices, yet are characterized by high reliability and sensitivity. An example is acoustic transducers such as microphones that make use of a silicon sensing membrane, examples of which are disclosed in U.S. Pat. No. 5,146,435 to Bernstein, and U.S. Patent Application Publication No. 2001/0015106 to Aigner et al. In both Bernstein and Aigner et al., the silicon sensing membrane is movable and capacitively coupled to a stationary silicon membrane, such that sound waves impinging on the silicon sensing membrane are sensed by changes in the capacitive output of the device.
Processes for fabricating all-silicon microphones of the type disclosed by Bernstein and Aigner et al. are typically long, cumbersome, expensive, and not compatible with high-volume processes. In addition, the silicon sensing membranes can be prone to process-induced deformation and package-induced stresses that can prevent or interfere with proper operation of the device. For example, the silicon membrane disclosed in Aigner et al. is formed of a deposited silicon film and capacitively coupled to a stationary membrane formed of epitaxially-grown silicon. As well known in the art, stresses in deposited films such as the silicon sensing membrane of Aigner et al. are difficult to control, and high temperature steps required to form and process the stationary membrane of Aigner et al. can lead to plastic deformation of surrounding structures, including the silicon sensing membrane. A further disadvantage of capacitive audio devices such as those taught by Bernstein and Aigner et al. is the difficulty with which the distance between the capacitively coupled membranes can be precisely predetermined. For example, the capacitive gap of Bernstein's device is established by the shape of the stationary silicon membrane, while in the device of Aigner et al. the capacitive gap is established by a deposited spacer layer.
In view of the above, there is a continuing need for a process of making a relatively low-cost all-silicon sound transducer that is compatible with high-volume manufacturing practices, yet yields a device characterized by high reliability and performance characteristics.
BRIEF SUMMARY OF THE INVENTION
The present invention is a process of forming a capacitive audio transducer, preferably having an all-silicon monolithic construction that includes capacitive plates defined by doped single-crystal silicon layers. The capacitive plates are defined by etching the single-crystal silicon layers, and the capacitive gap therebetween is accurately established by wafer bonding, yielding a transducer that can be produced by high-volume manufacturing practices, yet is characterized by high reliability and performance characteristics.
The process generally makes use of a first wafer having thereon a first single-crystal silicon layer, which is doped with boron and germanium so as to be p-type. A second p-type single-crystal silicon layer is formed on the first single-crystal silicon layer, and a recess is defined in the second single-crystal silicon layer so as to expose a portion of the first single-crystal silicon layer therebeneath. The portion of the first single-crystal silicon layer exposed by the recess will subsequently define a first capacitor plate of the capacitive audio transducer. A second wafer is provided to have a third single-crystal silicon layer, also doped with boron and germanium so as to be p-type. The first and second wafers are then bonded together so that the recess in the second single-crystal silicon layer defines a cavity between the first and third single-crystal silicon layers of the first and second wafers, respectively. At least portions of the first and second wafers are then removed to expose a portion of the first single-crystal silicon layer defining the first capacitor plate and to expose a portion of the third single-crystal silicon layer, which is spaced apart from the first single-crystal silicon layer by the cavity. The exposed portion of the third single-crystal silicon layer thereby defines a second capacitor plate that is capacitively coupled to the first capacitor plate. One of the first and second capacitor plates is configured to be movable in response to impingement by sound vibrations. Finally, a vent is provided to the cavity through one of the first or third single-crystal silicon layers. A capacitive output signal is produced in response to changes in the distance between the first and second capacitor plates.
From the above, it can be appreciated that the present invention provides an audio transducer characterized by an uncomplicated fabrication process, which can be readily modified to promote both performance and processing characteristics of the transducer. The single-crystal silicon layers that define the first and second capacitor plates are preferably doped to be low stress, and are less prone to process-induced deformation and package-induced stresses than the corresponding structures of prior art all-silicon monolithic audio transducers. Because the single-crystal silicon layers are not deposited films requiring high processing temperatures, the surrounding structures are also less vulnerable to plastic deformation. Another advantage of the invention is that the distance between the capacitive plates is established by the thicknesses of etched layers as a result of the wafer bonding process, enabling the capacitive gap to be precisely predetermined.
Other objects and advantages of this invention will be better appreciated from the following detailed description.
REFERENCES:
patent: 4420790 (1983-12-01), Golke et al.
patent: 4670092 (1987-06-01), Motamedi
patent: 5146435 (1992-09-01), Bernstein
patent: 5343064 (1994-08-01), Spangler et al.
patent: 5369544 (1994-11-01), Mastrangelo
patent: 5555448 (1996-09-01), Thiede et al.
patent: 5610971 (1997-03-01), Vandivier
patent: 5706565 (1998-01-01), Sparks et al.
patent: 5725729 (1998-03-01), Greiff
patent: 6156585 (2000-12-01), Gogoi et al.
patent: 2001/0015106 (2001-08-01), Aigner et al.
Baney William J.
Betzner Timothy M.
Chilcott Dan W.
Christenson John C.
Freeman John E.
Chmielewski Stefan V.
Delphi Technologies Inc.
Nguyen Tai
Tugbang A. Dexter
LandOfFree
Process of making an all-silicon microphone does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Process of making an all-silicon microphone, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Process of making an all-silicon microphone will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3328328