Compositions: ceramic – Ceramic compositions – Glass compositions – compositions containing glass other than...
Reexamination Certificate
2000-09-01
2003-03-25
Fiorilla, Christopher A. (Department: 1731)
Compositions: ceramic
Ceramic compositions
Glass compositions, compositions containing glass other than...
C501S069000
Reexamination Certificate
active
06537938
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a glass which is useful for e.g. a member of a semiconductor sensor and which can be effectively bonded by anodic bonding to a silicon crystal substrate (hereinafter referred to simply as “the silicon”).
2. Discussion of Background
Heretofore, semiconductor sensors for measuring the pressure of a gas or liquid, or the acceleration of a body in motion, have been widely practically used in the field of automobiles or measuring instruments. These sensors are designed mainly to detect a change in the electrical capacitance or strain exerted to the silicon, and miniaturization, cost reduction and improvement in sensitivity have been advanced by micromachining technology.
On the other hand, as a member for a semiconductor sensor, a glass having a thermal expansion coefficient close to the silicon is employed as a support or substrate for supporting the silicon. This glass also has a feature that it can be bonded to the silicon by an anodic bonding method which employs no adhesive, whereby the residual strain at the bond interface can be suppressed to a minimum level, and thus it has contributed to an improvement of the sensor properties.
Anodic bonding is a method wherein a glass is heated to a temperature at which readily mobile cations contained in the glass are readily mobile, and using the silicon as an anode and the glass as a cathode, a DC voltage is applied to heat-bond them. It is considered that cations in the glass move to the cathode, and as a result, non-bridging oxygen ions in the glass at the interface with the silicon will establish covalent bond with the silicon, whereby firm bonding will be established between them.
Heretofore, as a glass suitable for such an application, an aluminosilicate glass having a low thermal expansion coefficient has been invented and disclosed in JP-A-4-83733. The curve of the thermal expansivity of such a glass resembles the curve of the thermal expansivity of the silicon, and it has a feature that it contains sodium as readily mobile cations essential for anodic bonding with the silicon.
However, sodium is a component which sharply increases the thermal expansion coefficient of the glass, whereby its content in the glass is limited. Consequently, the amount of sodium ions mobile during the anodic bonding is also limited. In order to carry out the anodic bonding efficiently, it is essential to let sodium ions move as many as possible. For that purpose, a high temperature and a high voltage are required. Specifically, the anodic bonding is presently carried out at about 400° C. at about 800V.
On the other hand, along with the progress in the micromachining technology in recent years, the sensors have been modified for high integration and more complicated structures, and lamination of silicon or glass, or an element having a sandwich structure, has also been developed, whereby anodic bonding is carried out several times for a single member. Further, there has been an increasing trend for adopting a step of anodic bonding after forming a circuit or a pattern on a substrate.
Under these circumstances, it has been desired to lower the temperature during the anodic bonding to prevent thermal damage of a sensor element during the bonding, in addition to the demand for efficiency in the anodic bonding. JP-A-5-9039 discloses a glass-ceramic having crystals of &bgr;-quartz solid solution precipitated and having sodium introduced in a large amount, which has a minus thermal expansion coefficient. This publication discloses Examples wherein anodic bonding was carried out at 150° C. However, there has been a problem that as the glass used, is a glass-ceramic, not only the process steps increase for e.g. precipitation of crystals, but also the curve of the thermal expansivity of the glass does not necessarily resemble the curve of the thermal expansivity of the silicon, as compared with the aluminosilicate glass.
SUMMARY OF THE INVENTION
The present invention is intended to solve the above-mentioned conventional problems and to provide a glass whereby anodic bonding is possible at a temperature lower than 300° C. without losing the matching to the curve of the thermal expansivity of the silicon.
The present inventors have conducted an extensive study to solve the above problems, and as a result, have found that lithium has a substantial effect to reduce the volume resistivity of glass, which is an index indicating the efficiency for anodic bonding. Besides, it has been found that lithium has a little effect to increase the thermal expansion coefficient of glass, and accordingly, a large amount of lithium can be introduced to effectively reduce the volume resistivity. Further, it has been found that even without containing sodium, firm anodic bonding can be carried out in the same manner as with the conventional sodium-containing glass.
The present invention has been accomplished on the basis of the above discoveries, and firstly, the present invention provides a glass for anodic bonding, which is a glass to be anodically bonded to a silicon crystal plate and which contains substantially no Na
2
O and contains from 4 to 8 mol % of Li
2
O.
Secondly, the present invention provides such a glass which is an aluminosilicate glass and which has an average thermal expansion coefficient within a range of from room temperature to 300° C. within a range of from 25×10
−7
/° C. to 40×10
−7
/° C. and is capable of being anodically bonded to a silicon crystal plate at a temperature lower than 300° C.
Thirdly, the present invention provides the above-mentioned first or second glass which has a composition consisting essentially of from 56 to 70 mol % of SiO
2
, from 7 to 17 mol % of Al
2
O
3
, from 4 to 8 mol % of Li
2
O, from 1 to 11 mol % of MgO, from 4 to 12 mol % of ZnO, from 14 to 23 mol % of Li
2
O+MgO+ZnO, from 0 to 9 mol % of B
2
O
3
and from 0 to 3 mol % of CaO+BaO.
Fourthly, the present invention provides the above-mentioned third glass which preferably has a composition consisting essentially of from 60 to 68 mol % of SiO
2
, from 7 to 12 mol % of Al
2
O
3
, from 5 to 7 mol % of Li
2
O, from 3 to 9 mol % of MgO, from 5 to 10 mol % of ZnO, from 15 to 21 mol % of Li
2
O+MgO+ZnO, from 5 to 9 mol % of B
2
O
3
and from 0 to 2 mol % of CaO+BaO.
REFERENCES:
patent: 3642504 (1972-02-01), Petzold et al.
patent: 3811901 (1974-05-01), Bacon
patent: 4229220 (1980-10-01), Hirota
patent: 5736061 (1998-04-01), Fukada et al.
patent: 5998037 (1999-12-01), Sridharan et al.
patent: 2577493 (1996-11-01), None
patent: WO 97/07701 (1990-07-01), None
L.J. Spangler, et al., pp. 140-143, “A New Silicon-On-Glass Process for Integrated Sensors,” 1988.
Asahi Techno Glass Corporation
Fiorilla Christopher A.
LandOfFree
Glass for anodic bonding does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Glass for anodic bonding, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Glass for anodic bonding will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3032264