U.S. patent application number 13/739587 was filed with the patent office on 2013-05-16 for mems relay and method of forming the mems relay.
This patent application is currently assigned to NATIONAL SEMICONDUCTOR CORPORATION. The applicant listed for this patent is National Semiconductor Corporation. Invention is credited to Aditi Dutt CHAUDHURI, Peter JOHNSON, Dok Won LEE.
Application Number | 20130122628 13/739587 |
Document ID | / |
Family ID | 46600267 |
Filed Date | 2013-05-16 |
United States Patent
Application |
20130122628 |
Kind Code |
A1 |
LEE; Dok Won ; et
al. |
May 16, 2013 |
MEMS Relay and Method of Forming the MEMS Relay
Abstract
A micro-electromechanical systems (MEMS) relay includes a switch
with a first contact region and a second contact region that are
vertically separated from each other by a gap. The MEMS relay
requires a small vertical movement to close the gap and therefore
is mechanically robust. In addition, the MEMS relay has a small
footprint and, therefore, can be formed on top of small integrated
circuits.
Inventors: |
LEE; Dok Won; (Mountain
View, CA) ; JOHNSON; Peter; (Sunnyvale, CA) ;
CHAUDHURI; Aditi Dutt; (Mountain View, CA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
National Semiconductor Corporation; |
Santa Clara |
CA |
US |
|
|
Assignee: |
NATIONAL SEMICONDUCTOR
CORPORATION
Santa Clara
CA
|
Family ID: |
46600267 |
Appl. No.: |
13/739587 |
Filed: |
January 11, 2013 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
13020052 |
Feb 3, 2011 |
8378766 |
|
|
13739587 |
|
|
|
|
Current U.S.
Class: |
438/52 |
Current CPC
Class: |
H01H 49/00 20130101;
H01H 50/005 20130101; H01H 50/36 20130101; H01H 50/44 20130101;
H01L 29/84 20130101; H01H 1/0036 20130101; Y10T 29/49073
20150115 |
Class at
Publication: |
438/52 |
International
Class: |
H01L 29/84 20060101
H01L029/84 |
Claims
1. A method of forming a MEMS relay comprising: providing a
semiconductor structure having a top surface; forming a metal
interconnect structure touching the semiconductor structure top
surface; depositing a top passivation layer, having top and bottom
surfaces, performing a mask and etch process to form openings in
the top passivation layer to expose a plurality of conductive pads,
wherein the plurality of conductive pads provide points for
external electrical connections, and points for electrical
connections to overlying devices, wherein a first pad of the
plurality of conductive pads is coupled to a first switch contact
and a magnetic cantilever, a second pad of the plurality of
conductive pads is coupled to a second switch contact, a third pad
of the plurality of conductive pads is coupled lower magnetic core
section and a first pair of pads are coupled_to a magnetic coil;
depositing a photoimageable epoxy or polymer forming a first
non-conductive layer, having top and bottom surfaces, that touches
the top surface of the passivation layer having a second set of
openings for switch metal plugs, magnetic coil metal plugs and a
lower magnetic core section, formed by a mask and develop process;
forming the lower magnetic core section, having top and bottom
surfaces, that extends through the first non-conductive layer,
touching the top surface of the top passivation layer and
communicating with the third pad of the plurality of conductive
pads, by depositing a seed layer and then plating the seed layer
with a soft magnetic material, wherein the top surface of the lower
magnetic core section and the top surface of the first
non-conductive layer lie substantially in the same plane; forming a
first plurality of metal plugs, having top and bottom surfaces,
formed in the second set of openings of the first non-conductive
layer for coupling with a switch and a magnetic coil, wherein the
top surfaces of the first plurality of metal plugs and the top
surface of the first non-conductive layer lie substantially in the
same plane; depositing a second non-conductive layer, having top
and bottom surfaces, that touches the top surface of the first
non-conductive layer and the top surface of the lower magnetic core
section with a third set of openings for switch metal plugs and
magnetic coil metal plugs and via openings that expose regions on
the top surface of the lower magnetic core section formed by a mask
and etch process; forming a coil of a continuous series of loops,
having top and bottom surfaces, touching the top surface of the
second non-conductive layer and the top surfaces of the magnetic
switch metal plugs of the first plurality of metal plugs; forming a
second plurality of metal plugs, having top and bottom surfaces,
formed on and in contact within the second set of openings of the
second non-conductive layer for coupling with the switch contact;
depositing a photoimageable epoxy or polymer forming a third
non-conductive layer, having top and bottom surfaces, that touches
the top surface of the second non-conductive layer, the top surface
of the lower magnetic core section and covering the coil having a
fourth set of openings to expose the tops of the set of switch
metal plugs of the second plurality of metal plugs and a pair of
core openings to expose regions on the top surface of the lower
magnetic core section formed by a mask and develop process; forming
first and second magnetic core vias, having top and bottom
surfaces, formed in the pair of core openings, by depositing a seed
layer and then plating the seed layer with the soft magnetic
material, the bottom surfaces of the first and second magnetic core
vias touching the top surface of the lower magnetic core section,
wherein the tops surfaces of the first and second of magnetic core
vias and the top surface of the third non-conductive layer lie
substantially in the same plane; depositing a fourth non-conductive
member, having a top and bottom surfaces, touches the top surface
of the first magnetic core via; forming a first metal trace, having
a top and bottom surfaces, touching the top surface of third
non-conductive layer the top of the first metal switch plug of the
second plurality of metal plugs by a mask and etch process; forming
a second metal trace, having a top and bottom surfaces, touching
the top surface of third non-conductive layer and the top of the
second metal switch plug of the second plurality of metal plugs by
a mask and etch process; forming a third metal trace, having a top
and bottom surfaces, touching the top surface of the fourth
non-conductive member and the first metal trace; wherein the second
and third metal traces have first and second metal contact regions
respectively, disposed at the ends and top surface of the second
metal trace the bottom surface of the third metal trace, forming
the switch contacts; forming a magnetic cantilever core section
that touches the top surface of the third metal trace; wherein the
second contact region is spaced apart from the first contact region
when the second contact region is in the first position, the second
contact region touching the first contact region when the second
contact region is in the second position; and wherein the lower
magnetic core section and the third conductive trace are
electrically coupled to a holding voltage source to
electrostaticaly maintain a closed position after the coil is
deenergized.
2. The method of claim 1 wherein the seed layer is comprised of a
first layer of titanium; a second layer of titanium and a layer of
copper therebetween, the first layer of titanium forming the bottom
of the seed layer and the second layer of titanium forming the top
of the seed layer.
3. The method of claim 1 wherein the soft magnetic material is
comprised of an alloy of nickel and iron.
4. The method of claim 1 wherein forming a coil of a continuous
series of loops and the second plurality of metal plugs includes:
depositing the seed layer touching the top surface of the second
non-conductive layer and the top surfaces of the magnetic switch
coil metal plugs of the first plurality of metal plugs; forming a
sacrificial coil and second plurality of metal plugs mold; etching
an removing the second layer of titanium down to the copper of the
seed layer exposed by the sacrificial coil and second plurality of
metal plugs mold; plating the coil metal and the second plurality
of metal plugs on the seed layer; and removing the sacrificial mold
and the remaining seed layer.
5. The method of claim 1 wherein the first, second and third metal
traces are comprised of gold.
6. The method of claim 1 wherein the coil metal and the second
plurality of metal plugs are comprised of copper.
7. The method of claim 1 wherein forming third metal trace and the
magnetic cantilever core section_structure includes: forming a
sacrificial structure to touch a top surface of the third
non-conductive layer and the second metal trace, the third metal
trace touching and lying over the non-conductive member and the
first metal trace and the sacrificial structure; and forming an
upper magnetic core section that touches the second metal
trace.
8. The method of claim 7 and further comprising removing the
sacrificial structure after the upper magnetic core section has
been formed
Description
RELATED APPLICATION
[0001] This application is a divisional of co-pending application
Ser. No. 13/020,052 filed on Feb. 3, 2011, which is the subject of
a Notice of Allowance mailed on Dec. 19, 2012. application Ser. No.
13/020,052 is hereby incorporated by reference herein in its
entirety.
1. FIELD OF THE INVENTION
[0002] The present invention relates to MEMS devices and, more
particularly, to a MEMS relay and a method of forming the MEMS
relay.
2. DESCRIPTION OF THE RELATED ART
[0003] A switch is a well-known device that connects, disconnects,
or changes connections between devices. An electrical switch is a
switch that provides a low-impedance electrical pathway when the
switch is "closed," and a high-impedance electrical pathway when
the switch is "opened." A mechanical-electrical switch is a type of
switch where the low-impedance electrical pathway is formed by
physically bringing two electrical contacts together, and the
high-impedance electrical pathway is formed by physically
separating the two electrical contacts from each other.
[0004] An actuator is a well-known mechanical device that moves or
controls a mechanical member to move or control another device.
Actuators are commonly used with mechanical-electrical switches to
move or control a mechanical member that closes and opens the
switch, thereby providing the low-impedance and high-impedance
electrical pathways, respectively, in response to the actuator.
[0005] A relay is a combination of a switch and an actuator where
the mechanical member in the actuator moves in response to
electromagnetic changes in the conditions of an electrical circuit.
For example, electromagnetic changes due to the presence or absence
of a current in a coil can cause the mechanical member in the
actuator to close and open the switch.
[0006] One approach to implementing actuators and relays is to use
micro-electromechanical systems (MEMS) technology. MEMS devices are
formed using the same fabrication processes that are used to form
conventional semiconductor structures, such as the interconnect
structures that provide electrical connectivity to the transistors
on a die.
BRIEF DESCRIPTION OF THE DRAWINGS
[0007] FIGS. 1A-1D through FIGS. 22A-22D are views illustrating an
example of a method of forming a MEMS relay in accordance with the
present invention. FIGS. 1A-22A are plan views. FIGS. 1B-22B are
cross-sectional views taken along lines 1B-1B through 22B-22B in
FIGS. 1A-22A. FIGS. 1C-22C are cross-sectional views taken along
lines 1C-1C through 22C-22C in FIGS. 1A-22A. FIGS. 1D-22D are
cross-sectional views taken along lines 1D-1D through 22D-22D in
FIGS. 1A-22A.
[0008] FIG. 23 is a cross-sectional view taken along line 22B-22B
of FIG. 22A illustrating MEMS relay 230 in the closed position in
accordance with the present invention.
[0009] FIG. 24 is a cross-sectional view taken along line 22B-22B
of FIG. 22A illustrating an example of a MEMS relay 2400 in
accordance with an alternate embodiment of the present
invention.
DETAILED DESCRIPTION OF THE INVENTION
[0010] FIGS. 1A-1D through FIGS. 22A-22D show views that illustrate
an example of a method of forming a MEMS relay in accordance with
the present invention. FIGS. 1A-22A show plan views, while FIGS.
1B-22B show cross-sectional views taken along lines 1B-1B through
22B-22B in FIGS. 1A-22A, FIGS. 1C-22C show cross-sectional views
taken along lines 1C-1C through 22C-22C in FIGS. 1A-22A, and FIGS.
1D-22D show cross-sectional views taken along lines 1D-1D through
22D-22D in FIGS. 1A-22A.
[0011] As shown in FIGS. 1A-1D, the method utilizes a
conventionally-formed semiconductor wafer 100 that has a
semiconductor structure 110, and a metal interconnect structure 112
that touches the top surface of semiconductor structure 110. In the
present example, semiconductor structure 110 includes a large
number of electrical devices, such as transistors, resistors,
capacitors, and diodes.
[0012] Further, metal interconnect structure 112, which
electrically connects together the electrical devices in
semiconductor structure 110 to form a circuit, includes a number of
levels of metal traces, a large number of contacts that connect the
bottom metal trace to electrically conductive regions on
semiconductor structure 110, and a large number of inter-metal vias
that connect the metal traces in adjacent layers together.
[0013] In addition, metal interconnect structure 112 includes a top
passivation layer 114 with openings that expose a number of
conductive pads 116. The pads 116, in turn, are selected regions of
the metal traces in the top metal layer that provide points for
external electrical connections, and points for electrical
connections to overlying devices.
[0014] In the present example, the pads 116 include a pair of
switch pads 116A and 116B which provide input and output electrical
connections for a to-be-formed switch, and a pair of coil pads 116C
and 116D which provide input and output electrical connections for
a to-be-formed coil. (Only the pads 116A-116D, and not the entire
metal interconnect structure, are shown for clarity.)
[0015] As further shown in FIGS. 1A-1D, metal interconnect
structure 112 includes a pair of switch metal plugs 118A and 118B
that sit on top of and extend through passivation layer 114 to make
electrical connections with the switch pads 116A and 116B,
respectively. In addition, metal interconnect structure 112 also
includes a pair of coil metal plugs 118C and 118D that sit on top
of and extend through passivation layer 114 to make electrical
connections with the coil pads 116C and 116D, respectively. The
metal plugs 118A-118D can be formed in a conventional copper
electroplating process.
[0016] As additionally shown in FIGS. 1A-1D, the method begins by
forming a non-conductive layer 120 that touches the top surface of
passivation layer 114 and the side wall surfaces of the metal plugs
118A-118D, and a lower magnetic core section 122 that extends
through non-conductive layer 120 to touch the top surface of
passivation layer 114. Non-conductive layer 120 and lower magnetic
core section 122 are formed so that the top surfaces of the metal
plugs 118A-118D, non-conductive layer 120, and lower magnetic core
section 122 lie substantially in the same plane P.
[0017] For example, as shown in FIG. 2A-2D, non-conductive layer
120 can be formed by depositing a layer of photoimageable epoxy or
polymer, such as SU-8, which is substantially self planarizing, on
the top surface of passivation layer 114. Once the photoimageable
epoxy or polymer has been deposited, a lower core opening 124 is
formed by projecting a light through a mask to form a patterned
image on the layer of photoimageable epoxy or polymer. The light
hardens the regions of the layer of photoimageable epoxy or polymer
that are exposed to light. Following this, the softened regions
(the regions protected from light) of the layer of photoimageable
epoxy or polymer are removed to form non-conductive layer 120 with
lower core opening 124 that exposes the top surface of passivation
layer 114.
[0018] As shown in FIGS. 3A-3D, once non-conductive layer 120 with
lower core opening 124 has been formed, lower magnetic core section
122 is formed by depositing a seed layer 126 on passivation layer
114 and non-conductive layer 120. For example, seed layer 126 can
be formed by depositing 300 .ANG. of titanium, 3000 .ANG. of
copper, and 300 .ANG. of titanium. (Seed layer 126 can also include
a barrier layer to prevent copper electromigration if needed.)
[0019] After seed layer 126 has been formed, the top titanium layer
is stripped and a soft magnetic material, such as an alloy of
nickel and iron like permalloy or orthonol, is deposited by
electroplating to a thickness of, for example, 10 .mu.m to form a
plated layer 130. After this, as shown in FIGS. 4A-4D, wafer 100 is
planarized in a conventional manner, such as by chemical-mechanical
polishing, to expose the top surfaces of the plugs 118A-118D and
form lower magnetic core section 122.
[0020] As shown in FIGS. 5A-5D, after non-conductive layer 120 and
lower magnetic core section 122 have been formed, a coil structure
is next formed. The coil structure is formed by forming a
non-conductive layer 134 on non-conductive layer 120 and lower
magnetic core section 122. Non-conductive layer 134 has a number of
plug openings 136A-136D that expose the plugs 118A-118D,
respectively, and a pair of via openings 138A and 138B that expose
regions on the top surface of lower magnetic core section 122.
[0021] For example, non-conductive layer 134 can be formed with a
layer of photoimageable epoxy or polymer, such as SU-8, which is
substantially self planarizing. Once the photoimageable epoxy or
polymer has been deposited, the openings 136A-136D and 138A-138B
are formed by projecting a light through a mask to form a patterned
image on the photoimageable epoxy or polymer. The light hardens the
regions of the photoimageable epoxy or polymer that are exposed to
the light. Following this, the softened regions (the regions
protected from light) of the photoimageable epoxy or polymer are
removed to form non-conductive layer 134 with the openings
136A-136D and 138A-138B.
[0022] As shown in FIGS. 6A-6D, following the formation of
non-conductive layer 134 with the openings 136A-136D and 138A-138B,
a coil 140 which has a continuous series of loops is formed on
non-conductive layer 134 so that a portion of each loop lies
directly vertically over lower magnetic core section 122. Coil 140
is electrically connected to the coil plugs 118C and 118D. In
addition, a pair of switch plugs 142A-142B is formed through
non-conductive layer 134 to make electrical connections with the
switch plugs 118A-118B, respectively.
[0023] For example, as shown in FIGS. 7A-7D, coil 140 and the
switch plugs 142A-142B can be formed by forming a seed layer 144 on
the plugs 118A-118D, lower magnetic core section 122, and
non-conductive layer 134. Seed layer 144 can be formed by
depositing 300 .ANG. of titanium, 3000 .ANG. of copper, and 300
.ANG. of titanium. (Seed layer 144 can also include a barrier layer
to prevent copper electromigration if needed.)
[0024] Once seed layer 144 has been formed, a plating mold 146 is
formed on the top surface of seed layer 144. Plating mold 146, in
turn, has an opening that exposes a portion of seed layer 144 that
lies over the plugs 118C and 118D and defines the shape of the
to-be-formed coil, and openings that expose portions of seed layer
144 that lie over the plugs 118A-118B.
[0025] As shown in FIGS. 8A-8D, following the formation of plating
mold 146, the top titanium layer is stripped and copper is
deposited by electroplating to form coil 140 and the switch plugs
142A-142B. After the electroplating, plating mold 146 and the
underlying regions of seed layer 144 are removed.
[0026] After coil 140 and the switch plugs 142A-142B have been
formed, as shown in FIGS. 9A-9D, a non-conductive layer 150 is
deposited on non-conductive layer 134, coil 140, and the plugs
142A-142B to complete the formation of the coil structure.
Non-conductive layer 150 has a pair of switch openings 152A and
152B that expose the top surfaces of the switch plugs 142A and
142B, and a pair of core openings 152C and 152D that expose regions
on the top surface of lower magnetic core section 122.
[0027] For example, non-conductive layer 150 can be formed with a
layer of photoimageable epoxy or polymer, such as SU-8, which is
substantially self planarizing. Once the photoimageable epoxy or
polymer has been deposited, the openings 152A-152D are formed by
projecting a light through a mask to form a patterned image on the
photoimageable epoxy or polymer. The light hardens the regions of
the photoimageable epoxy or polymer that are exposed to the light.
Following this, the softened regions (the regions protected from
light) of the photoimageable epoxy or polymer are removed to
forming non-conductive layer 150 with the openings 152A-152D.
[0028] As shown in FIGS. 10A-10D, once non-conductive layer 150
with the openings 152A-152D have been formed, a magnetic core via
154A and a magnetic core via 154B are formed in the openings 152C
and 152D, respectively, to touch the opposite ends of lower
magnetic core section 122. For example, as shown in FIGS. 11A-11D,
the magnetic core vias 154A and 154B can be formed by depositing a
seed layer 160 on the top surfaces of lower magnetic core section
122, the switch plugs 142A-142B, and non-conductive layer 150.
[0029] Seed layer 160 can be formed by depositing 300 .ANG. of
titanium, 3000 .ANG. of copper, and 300 .ANG. of titanium. (Seed
layer 160 can also include a barrier layer to prevent copper
electromigration if needed.) After seed layer 160 has been formed,
a plating mold 162 is formed on the top surface of seed layer 160.
Plating mold 162, in turn, has openings that expose portions of
seed layer 160 that lie over the ends of lower magnetic core
section 122.
[0030] As shown in FIGS. 12A-12D, following the formation of
plating mold 162, the top titanium layer is stripped and a soft
magnetic material, such as an alloy of nickel and iron like
permalloy or orthonol, is deposited by electroplating to a
thickness of, for example, 4 .mu.m to form plated regions 164A and
164B. After the electroplating, plating mold 162 and the underlying
regions of seed layer 164 are removed. Following this, wafer 100 is
planarized to remove portions of plated regions 164A and 164B and
form the magnetic core vias 154A and 154B. As a result, the
continuous series of loops of coil 140 is wound around magnetic
core via 154B.
[0031] After the magnetic core vias 154A-154B have been formed, as
shown in FIGS. 13A-13D, an upper structure is next formed. The
upper structure can be formed by forming a non-conductive member
170 on non-conductive layer 150 and magnetic core via 154A to cover
core magnetic via 154A. For example, as shown in FIGS. 14A-14D,
non-conductive member 170 can be formed by depositing a
non-conductive layer 172 on the plugs 142A-142B, non-conductive
layer 150, and the magnetic core vias 154A and 154B. After this, a
patterned photoresist layer 174 is formed on the top surface of
non-conductive layer 172.
[0032] Patterned photoresist layer 174 is formed in a conventional
manner, which includes depositing a layer of photoresist, and
projecting a light through a patterned black/clear glass plate
known as a mask to form a patterned image on the layer of
photoresist. The light softens the photoresist regions exposed to
the light. Following this, the softened photoresist regions are
removed. After patterned photoresist layer 174 has been formed, the
exposed regions of non-conductive layer 172 are etched in a
conventional manner to form non-conductive member 170. Patterned
photoresist layer 174 is then removed with conventional solvents
and processes.
[0033] As shown in FIGS. 15A-15D, following the formation of
non-conductive member 170, a metal plug 180 and a metal trace 182
are formed to make electrical connections to switch plugs 142A and
142B, respectively. For example, as shown in FIGS. 16A-16D, metal
plug 180 and metal trace 182 can be formed by depositing a gold
layer 184 approximately 3000 .ANG. thick that touches the switch
plugs 142A-142B, non-conductive layer 150, core via 154B, and
non-conductive member 170.
[0034] After this, a patterned photoresist layer 186 is formed on
the top surface of gold layer 184. Patterned photoresist layer 186
is formed in a conventional manner. After patterned photoresist
layer 186 has been formed, the exposed regions of gold layer 184
are etched in a conventional manner to form metal plug 180 and
metal trace 182. Patterned photoresist layer 186 is then removed
with conventional solvents and processes.
[0035] As shown in FIGS. 17A-17D, following the formation of metal
plug 180 and metal trace 182, a sacrificial structure 190 is formed
to touch the top surface of non-conductive layer 150, magnetic core
via 154B, non-conductive member 170, and metal trace 182. For
example, as shown in FIGS. 18A-18D, sacrificial structure 190 can
be fabricated by forming a sacrificial layer 192 on non-conductive
layer 150, magnetic core via 154B, non-conductive member 170, metal
plug 180, and metal trace 182.
[0036] After this, a patterned photoresist layer 194 is formed on
the top surface of sacrificial layer 192. Patterned photoresist
layer 194 is formed in a conventional manner. After patterned
photoresist layer 194 has been formed, the exposed regions of
sacrificial layer 192 are etched in a conventional manner to form
sacrificial structure 190. Patterned photoresist layer 194 is then
removed with conventional solvents and processes.
[0037] As shown in FIGS. 19A-19D, following the formation of
sacrificial structure 190, a metal trace 210 is formed to touch the
top surface of non-conductive member 170, metal plug 180, and
sacrificial structure 190 to lie over lower magnetic core section
122 and a contact region of metal trace 182. After the formation of
metal trace 210, a magnetic cantilever core section 212 is formed
to touch the top surface of metal trace 210 and lie over lower
magnetic core section 122.
[0038] For example, as shown in FIGS. 20A-20D, metal trace 210 and
magnetic cantilever core section 212 can be formed by depositing a
gold layer 220 approximately 3000 .ANG. thick on the top surfaces
of non-conductive layer 150, non-conductive member 170, metal plug
180, metal trace 182, and sacrificial structure 190 to lie over
lower magnetic core section 122.
[0039] After gold layer 220 has been formed, a plating mold 222 is
formed on the top surface of gold layer 220. Plating mold 222, in
turn, has an opening that exposes a portion of gold layer 220 that
defines the shape of the to-be-formed magnetic cantilever core
section 212.
[0040] As shown in FIGS. 21A-21D, following the formation of
plating mold 222, a soft magnetic material, such as an alloy of
nickel and iron like permalloy or orthonol, is deposited by
electroplating to a thickness of, for example, 2 .mu.m to form
magnetic cantilever core section 212. After the electroplating,
plating mold 222 and the underlying regions of gold layer 220 are
removed to form metal trace 210. Thus, gold layer 220 is used to
form metal trace 210 as well as a seed layer for electroplating the
soft magnetic material.
[0041] After this, as shown in FIGS. 22A-22D, sacrificial structure
190 is etched away to form a MEMS relay 230. As a result,
sacrificial structure 190 can be implemented with any material
which can selectively etched away without removing excessive
amounts of the exposed elements of MEMS relay 230. Thus, as further
shown in FIGS. 22A-22D, MEMS relay 230 includes a switch 232, which
has a contact region 234 at the end of metal trace 182, and a
contact region 236 at the end of metal trace 210 that opposes
contact region 234.
[0042] In operation, contact region 236 is movable between a first
position and a second position. Switch 232 is open when no current
flows through coil 140. In this condition, contact region 234 is in
the first position, which is vertically spaced apart from contact
region 236 by a gap 240. FIG. 22A-22D show MEMS relay 230 in the
open position. FIG. 23 shows a cross-sectional view taken along
line 22B-22B of FIG. 22A that illustrates MEMS relay 230 in the
closed position in accordance with the present invention.
[0043] As shown in FIG. 23, switch 232 closes when a current flows
through coil 140. The current generates a magnetic field that pulls
magnetic cantilever core section 212 towards magnetic core via 154B
and lower magnetic core section 122 which, in turn, causes contact
region 236 to move to the second position and touch contact region
234.
[0044] One of the advantages of MEMS relay 230 is that MEMS relay
230 only requires a small vertical movement to close gap 240
between the contacts 234 and 236 and therefore is mechanically
robust. In addition, MEMS relay 230 has a small footprint and,
therefore, can be formed on top of small integrated circuits.
[0045] In order for switch 232 to close when current flows through
coil 140, the electromagnetic force generated by coil 140 must be
greater than the spring force of magnetic cantilever core section
212 (the force required to deflect contact region 236 of magnetic
cantilever core section 212 the amount required to close gap 240)
combined with a contact force (the force required to ensure that
contact region 236 fully touches contact region 234).
[0046] The spring force of magnetic cantilever core section 212, in
turn, is a function of the thickness of magnetic cantilever core
section 212. In the present example, the thickness of magnetic
cantilever core section 212 is much thinner (two microns) than the
thickness of lower magnetic core section 122 (ten microns). As a
result, the cross-sectional area of magnetic cantilever core
section 212 (thickness of two microns times a width) is much less
than the cross-sectional area of lower magnetic core section 122
(thickness of ten microns times the same width).
[0047] The maximum amount of magnetic flux that can flow through a
core member is a function of the cross-sectional area of the core
member and the permeability of the core member. Thus, if lower
magnetic core section 122, the magnetic core vias 154A and 154B,
and magnetic cantilever core section 212 are formed from the same
material, substantially more magnetic flux flows through lower
magnetic core section 122 than flows through the magnetic core vias
154A and 154B and magnetic cantilever core section 212. (The
magnetic core vias 154A and 154B and magnetic cantilever core
section 212 can be formed to have the same cross-sectional
areas.)
[0048] To increase the amount of magnetic flux that flows through
magnetic cantilever core section 212, and thus better contain the
magnetic flux around magnetic cantilever core section 212, lower
magnetic core section 122 can be formed from a material that has a
different permeability than the material used to form the magnetic
core vias 154A and 154B and magnetic cantilever core section
212.
[0049] For example, magnetic cantilever core section 212 and the
magnetic core vias 154A and 154B can be formed from permalloy,
which has a high permeability. Permalloy is approximately 80%
nickel and 20% iron. Adjusting the relative percentages of the
materials lowers the permeability. For example, orthonol is a
nickel-iron alloy of 50% nickel and 50% iron that has a lower
permeability than permalloy.
[0050] Thus, the differences between the cross-sectional areas of
lower magnetic core section 122 and magnetic cantilever core
section 212, which effect the maximum amount of flux that can pass
through sections 122 and 212, can be compensated for by forming
lower magnetic core section 122 with a material that has a lower
permeability than the material used to form the magnetic core vias
154A and 154B and magnetic cantilever core section 212.
[0051] Alternately, the amount of magnetic flux that flows through
magnetic cantilever core section 212 can be increased by increasing
the widths of the magnetic core vias 154A and 154B and magnetic
cantilever core section 212. Increasing the widths increases the
cross-sectional areas of the magnetic core vias 154A and 154B and
magnetic cantilever core section 212.
[0052] FIG. 24 shows a cross-sectional view taken along line
22B-22B of FIG. 22A that illustrates an example of a MEMS relay
2400 in accordance with an alternate embodiment of the present
invention. MEMS relay 2400 is similar to MEMS relay 230 and, as a
result, utilizes the same reference numerals to designate the
structures which are common to both relays.
[0053] As shown in FIG. 24, MEMS relay 2400 differs from MEMS relay
230 in that MEMS relay 2400 utilizes a lower magnetic core section
2410 in lieu of lower magnetic core section 122. Lower magnetic
core section 2410 is the same as lower magnetic core section 122
except that lower magnetic core section 2410 also extends through
passivation layer 114 to make an electrical connection with a pad
116E, which allows a voltage to be placed on lower magnetic core
section 2410 and the magnetic core vias 154A and 154B.
[0054] In operation, after switch 232 has closed in response to
current flowing through coil 140, a holding voltage with a
magnitude that is sufficient to electrostaticly hold switch 232 in
the closed position is placed on lower magnetic core section 2410
and the magnetic core vias 154A and 154B by way of pad 116E. (The
voltage required to electrostaticly hold switch 232 closed is
substantially less than the voltage required to electrostaticly
close switch 232.)
[0055] After the holding voltage has been applied, the current fed
into coil 140 is stopped, utilizing the holding voltage to keep
switch 232 closed. One of the advantages of the present embodiment
is that no current is required, and thus no power is consumed, to
maintain switch 232 in the closed position. (The holding voltage
can also be applied to lower magnetic core section 2410 before
current is fed into coil 140 to close switch 232.)
[0056] For example, if switch 232 is a ground switch such that
ground is placed on the metal traces 182 and 210 when switch 232 is
closed, then a positive holding voltage can placed on lower
magnetic core section 2410 and the magnetic core vias 154A and 154B
by way of pad 116E after switch 232 has been closed. (If the
positive holding voltage is less than a power supply voltage, the
power supply voltage can be placed on lower magnetic core section
2410 and the magnetic core vias 154A and 154B by way of pad 116E
after switch 232 has been closed.) The current fed into coil 140 is
then stopped, utilizing the holding voltage to keep switch 232
closed.
[0057] Similarly, if switch 232 is a power switch such that a power
supply voltage is placed on the metal traces 182 and 210 when
switch 232 is closed, then a voltage equal to the power supply
voltage less the holding voltage can be placed on lower magnetic
core section 2410 and the magnetic core vias 154A and 154B by way
of pad 116E after switch 232 has been closed. (If the holding
voltage is less than the power supply voltage, ground can be placed
on lower magnetic core section 2410 and the magnetic core vias 154A
and 154B by way of pad 116E after switch 232 has been closed.) The
current fed into coil 140 is then stopped, utilizing the holding
voltage to keep switch 232 closed.
[0058] If switch 232 is a signal switch such that the voltage
placed on the metal traces 182 and 210 varies between ground and
the power supply voltage when switch 232 is closed, then a voltage
equal to the power supply voltage plus the holding voltage can be
placed on lower magnetic core section 2410 and the magnetic core
vias 154A and 154B by way of pad 116E after switch 232 has been
closed. (Alternately, a voltage equal to ground less the holding
voltage can be placed on lower magnetic core section 2410 and the
magnetic core vias 154A and 154B by way of pad 116E after switch
232 has been closed.) The current fed into coil 140 is then
stopped, utilizing the holding voltage to keep switch 232
closed.
[0059] It should be understood that the above descriptions are
examples of the present invention, and that various alternatives of
the invention described herein may be employed in practicing the
invention. Thus, it is intended that the following claims define
the scope of the invention and that structures and methods within
the scope of these claims and their equivalents be covered
thereby.
* * * * *