US20160154070A1

US20160154070A1 - Wafer bonding method for use in making a mems gyroscope

Info

Publication number: US20160154070A1
Application number: US14/512,469
Authority: US
Inventors: Biao Zhang; Tao Ju
Original assignee: INSIGHTECH LLC
Current assignee: INSIGHTECH LLC
Priority date: 2012-07-27
Filing date: 2014-10-13
Publication date: 2016-06-02
Also published as: US20150033856A1; US20150033854A1; US10012670B2; US20140026660A1; US20140026659A1; US20160154020A1; US20140190257A1; US20150226555A1; US20160153780A1; US20150033855A1; US20160154019A1; US20140026658A1; US20150034620A1; US20140026661A1

Abstract

A MEMS gyroscope is disclosed herein, wherein the MEMS gyroscope comprised a magnetic sensing mechanism on a magnetic sensor wafer and a magnetic source on a MEMS wafer that further comprises a proof-mass. The magnetic sensor wafer and MEMS wafer are bonded through a bonding mechanism that comprises a hearting mechanism.

Description

CROSS-REFERENCE
This US utility patent application claims priority from co-pending US utility patent application “A HYBRID MEMS DEVICE,” Ser. No. 13/559,625 filed Jul. 27, 2012, which claims priority from US provisional patent application “A HYBRID MEMS DEVICE,” filed May 31, 2012, Ser. No. 61/653,408 to Biao Zhang and Tao Ju. This US utility patent application also claims priority from co-pending US utility patent application “A MEMS DEVICE,” Ser. No. 13/854,972 tiled Apr. 2, 2013 to the same inventor of this US utility patent application, the subject matter of each of which is incorporated herein by reference in its entirety.

TECHNICAL FIELD OF THE DISCLOSURE

The technical field of the examples to be disclosed in the following sections is related generally to the art of operation of microstructures, and, more particularly, to operation of MEMS devices comprising MEMS magnetic sensing structures.

BACKGROUND OF THE DISCLOSURE

Microstructures, such as microelectromechanical (hereafter MEMS) devices (e.g. accelerometers, DC relay and RF switches, optical cross connects and optical switches, microlenses, reflectors and beam splitters, filters, oscillators and antenna system components, variable capacitors and inductors, switched banks of filters, resonant comb-drives and resonant beams, and micromirror arrays for direct view and projection displays) have many applications in basic signal transduction. For example, a MEMS gyroscope measures angular rate.
A gyroscope (hereafter “gyro” or “gyroscope”) is based on the Coriolis effect as diagrammatically illustrated in FIG. 1. Proof-mass 100 is moving with velocity V_d. Under external angular velocity Ω, the Coriolis effect causes movement of the poof-mass (100) with velocity V_s. With fixed V_d, the external angular velocity can be measured from V_d. A typical example based on the theory shown in FIG. 1 is capacitive MEMS gyroscope, as diagrammatically illustrated in FIG. 2.
The MEMS gyro is a typical capacitive MEMS gyro, which has been widely studied. Regardless of various structural variations, the capacitive MEMS gyro in FIG. 2 includes the very basic theory based on which all other variations are built. In this typical structure, capacitive MEMS gyro 102 is comprised of proof-mass 100, driving mode 104, and sensing mode 102. The driving mode (104) causes the proof-mass (100) to move in a predefined direction, and such movement is often in a form of resonance vibration. Under external angular rotation, the proof-mass (100) also moves along the V_sdirection with velocity V_s. Such movement of V_sis detected by the capacitor structure of the sensing mode (102). Both of the driving and sensing modes use capacitive structures, whereas the capacitive structure of the driving mode changes the overlaps of the capacitors, and the capacitive structure of the sensing mode changes the gaps of the capacitors.
Current capacitive MEMS gyros, however, are hard to achieve submicro-g/rtHz because the capacitance between sensing electrodes decreases with the miniaturization of the movable structure of the sensing element and the impact of the stray and parasitic capacitance increase at the same time, even with large and high aspect ratio proof-masses.
Therefore, what is desired is a MEMS device capable of sensing angular velocities and methods of operating the same.

SUMMARY OF THE DISCLOSURE

In view of the foregoing, a MEMS gyroscope is disclosed herein, wherein the gyroscope comprises: a first substrate having a movable portion that is movable in response to an external angular velocity, said movable portion comprising a magnetic source for generating magnetic field; a second substrate having a magnetic sensor for detecting the magnetic field from said magnetic source; and a bonding structure for bonding the first and second wafer with a predetermined distance, said bonding structure comprising: a heating mechanism for generating heating; and a bonding material.

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1 diagrammatically illustrates the Coriolis effect in a MEMS structure;

FIG. 2 is a top view of a typical existing capacitive MEMS gyroscope having a driving mode and a sensing mode, wherein both of the driving and sensing mode utilize capacitance structures;

FIG. 3 illustrates an exemplary MEMS gyroscope having a magnetic sensing mechanism;

FIG. 4 illustrates a top view of a portion of an exemplary implementation of the MEMS gyroscope illustrated in FIG. 3, wherein the MEMS gyroscope illustrated in FIG. 4 having a capacitive driving mode and a magnetic sensing mechanism;

FIG. 5 illustrates a perspective view of a portion of another exemplary implementation of the MEMS gyroscope illustrated in FIG. 3, wherein the MEMS gyroscope illustrated in FIG. 5 having a magnetic driving mechanism for the driving mode and a magnetic sensing mechanism for the sensing mode

FIG. 6 illustrates an exemplary magnetic driving mechanism of the MEMS gyroscope in FIG. 5;

FIG. 7 illustrates an exemplary magnetic source of the MEMS gyroscope illustrated in FIG. 3;

FIG. 8 illustrates an exemplary magnetic sensing mechanism that can be used in the MEMS gyroscope illustrated in FIG. 3;

FIG. 9 shows an exemplary thin-film stack that can be configured into a CIP or CPP structure for use in the magnetic sensing mechanism illustrated in FIG. 8;

FIG. 10 illustrates an exemplary MEMS gyroscope that comprises multiple magnetic sensing structures;

FIG. 11 illustrates an exemplary wafer bonding scheme for use in making the MEMS gyroscope shown in FIG. 3; and

FIG. 12 illustrates a perspective view of the water bonding scheme in FIG. 11.

DETAILED DESCRIPTION OF SELECTED EXAMPLES

Disclosed herein is a MEMS gyroscope for sensing an angular velocity, wherein the MEMS gyroscope utilizes a magnetic sensing mechanism. It will be appreciated by those skilled in the art that the following discussion is for demonstration purposes, and should not be interpreted as a limitation. Many other variations within the scope of the following disclosure are also applicable. For example, the MEMS gyroscope and the method disclosed in the following are applicable for use in accelerometers.
Referring to FIG. 3, an exemplary MEMS gyroscope is illustrated herein. In this example, MEMS gyroscope 106 comprises magnetic sensing mechanism 114 for sensing the target angular velocity through the measurement of proof-mass 112. Specifically, MEMS gyroscope 106 comprises mass-substrate 108 and sensor substrate 110. Mass-substrate 108 comprises proof-mass 112 that is capable of responding to an angular velocity. The two substrates (108 and 110) are spaced apart, for example, by a pillar (not shown herein for simplicity) such that at least the proof-mass (112) is movable in response to an angular velocity under the Coriolis effect. The movement of the proof-mass (112) and thus the target angular velocity can be measured by magnetic sensing mechanism 114.
The magnetic sensing mechanism (114) in this example comprises a magnetic source 116 and magnetic sensor 118. The magnetic source (116) generates a magnetic field, and the magnetic sensor (118) detects the magnetic field and/or the magnetic field variations that is generated by the magnetic source (116). In the example illustrated herein in FIG. 3, the magnetic source is placed on/in the proof-mass (112) and moves with the proof-mass (112). The magnetic sensor (118) is placed on/in the sensor substrate (120) and non-movable relative to the moving proof-mass (112) and the magnetic source (116). With this configuration, the movement of the proof-mass 12) can be measured from the measurement of the magnetic field from the magnetic source (116).
Other than placing the magnetic source on/in the movable proof-mass (1112), the magnetic source (116) can be placed on/in the sensor substrate (120); and the magnetic sensor (118) can be placed/in the proof-mass (112).
It is also noted that the MEMS gyroscope illustrated in FIG. 3 can also be used as an accelerometer.
The MEMS gyroscope as discussed above with reference to FIG. 3 can be implemented in many ways, one of which is illustrated in FIG. 4. Referring to FIG. 4, the proof-mass (120) is driven by capacitive, such as capacitive comb. The sensing mode, however, is performed using the magnetic sensing mechanism illustrated in FIG. 3. For this reason, capacitive combs can be absent from the proof-mass (120).
Alternatively, the proof-mass can be driven by magnetic force, an example of which is illustrated in FIG. 5. Referring to FIG. 5, the mass substrate (108) comprises a movable proof-mass (126) that is supported by flexible structures such as flexures 128, 129, and 130. The layout of the flexures enables the proof-mass to move in a plane substantially parallel to the major planes of mass substrate 108. In particular, the flexures enables the proof-mass to move along the length and the width directions, wherein the length direction can be the driving mode direction and the width direction can be the sensing mode direction of the MEMS gyro device. The proof-mass (126) is connected to frame 132 through flexures (128, 129, and 130). The frame (132) is anchored by non-movable structures such as pillar 134. The mass-substrate (108) and sensing substrate 110 are spaced apart by the pillar (134). The proof-mass 12) in this example is driving by a magnetic driving mechanism (136). Specifically, the proof-mass (126) can move (e.g. vibrate) in the driving mode under magnetic force applied by magnetic driving mechanism 136, which is better illustrated in FIG. 6.
Referring to FIG. 6, the magnetic driving mechanism 136 comprise a magnet core 138 surrounded by coil 140. By applying an alternating current through coil 140, an alternating magnetic field can be generated from the coil 140. The alternating magnetic field applies magnetic force to the magnet core 140 so as to move the magnet core. The magnet core thus moves the proof-mass.
The magnetic source (114) of the MEMS gyroscope (106) illustrated in FIG. 3 can be implemented in many ways, one of which is illustrated in FIG. 7. Referring to FIG. 7, conductive wire 142 is displaced on/in proof-mass 112. In one example, conductive wire 142 can be placed on the lower surface of the proof-mass (112), wherein the lower surface is facing the magnetic sensors (118 in FIG. 3) on the sensor substrate (110, in FIG. 3). Alternatively, the conductive wire (142) can be placed on the top surface of the proof-mass (112), i.e. on the opposite side of the proof-mass (112) in view of the magnetic sensor (118). In another example, the conductive wire (142) can be placed inside the proof-mass, e.g. laminated or embedded inside the proof-mass (112), which will not be detailed herein as those examples are obvious to those skilled in the art of the related technical field.
The conductive wire (142) can be implemented in many suitable ways, one of which is illustrated in FIG. 7. In this example, the conductive wire (142) comprises a center conductive segment 146 and tapered contacts 144 and 148 that extend the central conductive segment to terminals, through the terminals of which current can be driven through the central segment. The conductive wire (142) may have other configurations. For example, the contact tapered contacts (144 and 148) and the central segment (146) maybe U-shaped such that the tapered contacts may be substantially parallel but are substantially perpendicular to the central segment, which is not shown for its obviousness.
The magnetic sensor (118) illustrated in FIG. 3 can be implemented to comprise a reference sensor (150) and a signal sensor (152) as illustrated in FIG. 8. Referring to FIG. 8, magnetic senor 118 on/in sensor substrate 120 comprises reference sensor 150 and signal sensor 152. The reference sensor (150) can be designated for dynamically measuring the magnetic signal background in which the target magnetic signal (e.g. the magnetic field from the conductive wire 146 as illustrated in FIG. 7) co-exists. The signal sensor (152) can be designated for dynamically measuring the target magnetic signal (e.g. the magnetic field from the conductive wire 146 as illustrated in FIG. 7). In other examples, the signal sensor (152) can be designated for dynamically measuring the magnetic signal background in which the target magnetic signal (e.g. the magnetic field from the conductive wire 146 as illustrated in FIG. 7) co-exists, while the signal sensor (150) can be designated for dynamically measuring the target magnetic signal (e.g. the magnetic field from the conductive wire 146 as illustrated in FIG. 7).
The reference sensor (150) and the signal sensor (152) preferably comprise magneto-resistors, such as AMRs, giant-magneto-resistors (such as spin-valves, hereafter SV), or tunneling-magneto-resistors (TMR). For demonstration purpose, FIG. 9 illustrates a magneto-resistor structure, which can be configured into CIP (current-in-plane, such as a spin-valve) or a CPP (current-perpendicular-to-plane, such as IMP, structure). As illustrated in FIG. 9, the magneto-resistor stack comprises top pin-layer 154, free-layer 156, spacer 158, reference layer 160, bottom pin layer 162, and substrate 120. Top pin layer 154 is provided for magnetically pinning free layer 156. The top pin layer can be comprised of IrMn, PtMn or other suitable magnetic materials. The free layer (156) can be comprised of a ferromagnetic material, such as NiFe, CoFe, CoFeB, or other suitable materials or the combinations thereof. The spacer (158) can be comprised of a non-magnetic conductive material, such as Cu, or an oxide material, such as Al₂O₃or MgO or other suitable materials The reference layer (160) can be comprised of a ferromagnetic magnetic material, such as NiFe, CoFe, CoFeB, or other materials or the combinations thereof. The bottom pin layer (162) is provided for magnetic pinning the reference layer (160), which can be comprised of a IrMn, PtMn or other suitable materials or the combinations thereof. The substrate (120) can be comprised of any suitable materials, such as glass, silicon, or other materials or the combinations thereof.
In examples wherein the spacer (158) is comprised of a non-magnetic conductive layer, such as Cu, the magneto-resistor (118) stack can be configured into a CIP structure (i.e. spin-valve, SV), wherein the current is driven in the plane of the stack layers. When the spacer (158) is comprised of an oxide such as Al₂O₃, MgO or the like, the magneto-resistor stack (118) can be configured into a CPP structure (i.e. TMR), wherein the current is driven perpendicularly to the stack layers.
In the example as illustrated in FIG. 9, the free layer (156) is magnetically pinned by the top pin layer (154), and the reference layer (160) is also magnetically pinned by bottom pin layer 162. The top pin layer (154) and the bottom pin layer (162) preferably having different blocking temperatures. In this specification, a blocking temperature is referred to as the temperature, above which the magnetic pin layer is magnetically decoupled with the associated pinned magnetic layer. For example, the top pin layer (154) is magnetically decoupled with the free layer (156) above the blocking temperature T_Bof the top pin layer (154) such that the free layer (156) is “freed” from the magnetic pinning of top pin layer (154). Equal to or below the blocking temperature T_Bof the top pin layer (154), the free layer (156) is magnetically pinned by the top pin layer (154) such that the magnetic orientation of the free layer (156) is substantially not affected by the external magnetic field. Similarly, the bottom pin layer (162) is magnetically decoupled with the reference layer (160) above the blocking temperature T_Bof the bottom pin layer (162) such that the reference layer (160) is “freed” from the magnetic pinning of bottom pin layer (162). Equal to or below the blocking temperature T_Bof the bottom pin layer (162), the reference layer (160) is magnetically pinned by the bottom pin layer (162) such that the magnetic orientation of the reference layer (162) is substantially not affected by the external magnetic field.
The top and bottom pin layers (154 and 162, respectively) preferably have different blocking temperatures. When the free layer (156) is “freed” from being pinned by the top pin layer (154), the reference layer (160) preferably remains being pinned by the bottom pin layer (162). Alternatively, when the free layer (156) is still pinned by the top pin layer (154), the reference layer (160) can be “freed” from being pinned by the bottom pin layer (162). In the later example, the reference layer (160) can be used as a “sensing layer” for responding to the external magnetic field such as the target magnetic field, while the free layer (156) is used as a reference layer to provide a reference magnetic orientation.
The different blocking temperatures can be accomplished by using different magnetic materials for the top pin layer (154) and bottom pin layer (162). In one example, the top pin layer (154) can be comprised of IrMn, while the bottom pin layer (162) can be comprised of PtMn, vice versa. In another example, both of the top and bottom pin layers (154 and 162) may be comprised of the same material, such as IrMn or PtMn, but with different thicknesses such that they have different blocking temperatures.
It is noted by those skilled in the art that the magneto-resistor stack (118) is configured into sensors for sensing magnetic signals. As such, the magnetic orientations of the free layer (156) and the reference layer (160) are substantially perpendicular at the initial state. Other layers, such as protective layer Ta, seed layers for growing the stack layers on substrate 120 can be provided, It is further noted that the magnetic stack layers (118) illustrated in FIG. 9 are what is often referred to as “bottom pin” configuration in the field of art. In other examples, the stack can be configured into what is often referred as “top pinned” configuration in the field of art, which will not be detailed herein.
In some applications, multiple magnetic sensing mechanisms can be provided, an example of which is illustrated in FIG. 10. Referring to FIG. 10, magnetic sensing mechanisms 116 and 164 are provided for detecting the movements of proof-mass 112. The multiple magnetic sensing mechanisms can be used for detecting the movements of proof-mass 112 in driving mode and sensing mode respectively. Alternatively, the multiple magnetic sensing mechanisms 116 and 164 can be provided for detecting the same modes (e.g. the driving mode and/or the sensing mode).
The MEMS gyroscope as discussed above can be fabricated in many ways. During fabrication, special concerns on the fact that magnetic sensor and MEMS proof-mass may having different properties need to be addressed. In some examples, the magnetic sensor comprises spintronic structures, such as spin-valve (SV), magnetic-tunnel-junction (MTJ) or other similar structures. These spintronic structures in general has a blocking temperature that is 220 C or less, which means that these spintronic structures can be processed at a temperature mot higher than the blocking temperature. However, the MEMS structures generally are processed at a much higher temperature, such as 250 C or higher. IN particular, bonding the MEMS wafer having the proof-mass and the magnetic sensor wafer having the magnetic sensor generally requires a bonding temperature of 250 C or higher to secure a reliable bonding strength. This problem can be solved by using a localized heating, as illustrated in FIG. 11.
Referring to FIG. 11, MEMS wafer 108 has proof-mass 112, on which magnetic source 116 is provided. Magnetic sensor wafer 110 has magnetic sensor 118. MEMS wafer 108 and magnetic sensor wafer 110 can be bonded by using bonding structure 170 that comprises a pillar (e.g. pillar 172 and/or 174), localized heater 176 and bonding material 178. FIG. 12 shows a perspective view of the structure in FIG. 11 to better illustrate the structure.
Referring to FIG. 12, localized heater 176 is disposed on the surface of pillar 174 wherein pillar 174 in this example forms a hermetic bonding even though not required in some examples. The localized heater is conductive such that it generates localized heating to raise the temperature in the vicinity of the heater when current is driven through. The localized heater has two terminals 182 fur feeding current. Bonding material 178, which can be any suitable bonding materials, such as a glass frit, metal alloy, or metal can be disposed on the heater. In this way, the bonding material (178) and the pillar (174) laminate the localized heater therebetween. In some examples especially when the bonding material is conductive, an insulating layer can be disposed between the localized heater (176) and bonding material 178, which is not shown in the figure.
In a bonding process, current is driven into the localized heater through terminals 182. As current flows through the localized heater, the temperature in the vicinity of the heater (176) is elevated. When this temperature equals to or higher than the melting temperature of the bonding material 178, the pillar (174) can be bonded to the magnetic sensor wafer (120) (which is not shown in FIG. 12). In this bonding process, pressure can be applied.
By using the localized heater (176), only the area in the vicinity of the heater (176) raises its temperature. The magnetic sensor may not experience temperature raise. As such, the bonding area (the area wherein the bonding material is disposed) can have a temperature higher than 220 C, such as 350 C or 400 C, while the area wherein the magnetic sensor is disposed may still keep a safe temperature that is lower than 200 C.
the example shown in FIG. 12, pillar 174 forms a hermetic bonding. The same bonding scheme and process are also applicable examples wherein the MEMS wafer and the magnetic sensor wafer are bonded non-hermetically. In those examples, multiple pillars can be provided; and each bonding area at individual pillars may be provided with a localized heater for generating localized heating.
The pillar (172, 174) can be separately provided. In another example, the pillar can be directly formed from the MEMS wafer (108) shown in FIG. 11. Alternatively, the pillar can be formed from the magnetic sensor wafer, in examples of which, the localized heater and the bonding materials can be disposed in the vicinity of the MEMS wafer (108).
It will be appreciated by those of skilled in the art that a new and useful MEMS gyroscope has been described herein. In view of the many possible embodiments, however, it should be recognized that the embodiments described herein with respect to the drawing figures are meant to be illustrative only and should not be taken as limiting the scope of what is claimed. Those of skill in the art will recognize that the illustrated embodiments can be modified in arrangement and detail. Therefore, the devices and methods as described herein contemplate all such embodiments as may come within the scope of the following claims and equivalents thereof. In the claims, only elements denoted by the words “means for” are intended to be interpreted as means plus function claims under 35 U.S.C. §112, the sixth paragraph.

Claims

We claim:

1. A MEMS gyroscope, comprising:

a first substrate having a movable portion that is movable in response to an external angular velocity, said movable portion comprising a magnetic source for generating magnetic field;

a second substrate having a magnetic sensor for detecting the magnetic field from said magnetic source; and

a bonding structure for bonding the first and second wafer with a predetermined distance, said bonding structure comprising:

a heating mechanism for generating heating; and

a bonding material.

2. The MEMS gyroscope of claim 1, wherein the bonding structure further comprises an insulating layer between the heating mechanism and the bonding material.

3. The MEMS gyroscope of claim 2, wherein the bonding material comprises a metal or a metal alloy.

4. The MEMS gyroscope of claim 2, wherein the magnetic sensor comprises a giant-magnetic-resistor.

5. The MEMS gyroscope of claim 2, wherein the magnetic sensors comprises a spin-valve structure.

6. The MEMS gyroscope of claim 2, wherein the magnetic sensors comprises a tunnel-magnetic-resistor.

7. The MEMS gyroscope of claim 2, wherein the magnetic sensors comprises a magnetic pickup coil that is an element of a fluxgate.