Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
MEMS DEVICES AND METHODS OF MANUFACTURE
Document Type and Number:
WIPO Patent Application WO/2003/072486
Kind Code:
A2
Abstract:
Microelectromechanical (MEMS) devices that use MEMS electromagnetic actuators to selectively generate displacement forces and methods of fabrication are disclosed herein. According to one exemplary embodiment disclosed herein, a MEMS device may include a substrate having a surface, an actuable element at least partially formed from the substrate, and an electromagnetic actuator disposed on the substrate for selectively applying a first force to the actuable element to displace the actuable element along a path. The actuable element may have a base and an arm coupled to the base. The base may include a portion comprised of a magnetic material. The electromagnetic actuator may comprise an electrically conductive coil, and the path of the actuable element may pass through a coil gap in the coil. The electromagnetic actuator may also comprise a magnetic core about which the electrically conductive coil may be wound. MEMS devices and methods of fabrication that can minimize derailing of actuable elements during fabrication and the effects of fabrication tolerances on the operation of the devices are also disclosed herein.

Inventors:
TAMURA HIROKAZU
NEAL MATTHEW J
MUGINO AKIRA
BORSKI JUSTIN C
SIDMAN ALAN L
ZHE JIANG
Application Number:
PCT/US2003/004440
Publication Date:
September 04, 2003
Filing Date:
February 13, 2003
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
ADVANCED MICROSENSORS (US)
International Classes:
B81B5/00; F04B19/00; F04B43/04; G02B6/35; H01F7/08; H02K33/16; H01F7/124; (IPC1-7): B81B5/00
Domestic Patent References:
WO2000005734A12000-02-03
Foreign References:
US5808384A1998-09-15
US5644177A1997-07-01
EP1164601A22001-12-19
EP1081722A22001-03-07
US6122149A2000-09-19
Attorney, Agent or Firm:
Sears, Jeffrey M. (Patent Group 155 Seaport Boulevar, Boston MA, US)
Download PDF:
Claims:
Claims
1. A MEMS device comprising: a substrate having a surface, an actuable element at least partially formed from the substrate, and an electromagnetic actuator disposed on the substrate for selectively applying a first force to the actuable element to displace the actuable element along a path.
2. The MEMS device of claim 1, wherein the actuable.
3. The MEMS device of claim 2, wherein the magnetic material is comprised of at least one of a permanent magnetic material and soft magnetic material.
4. The MEMS device of claim 2, wherein the magnetic material is comprised of at least one of ferrites, remalloy, vicalloy, AINiCo, Co, CoPt, a rare earth metal, NiFe (permalloy), CoFe (permendur), CoZr, FeN, AlSiFe (sendust), NiFeMo (supermalloy), NiFeCuCr (mumetal), NiFeCo, CoFeB, CoFeV (supermendur), CoFeCr (hiperco), CoZrTa, FeAIN, FeTaN, and combinations thereof.
5. The MEMS device of claim 1, wherein the electromagnetic actuator comprises an electrically conductive coil.
6. The MEMS device of claim 5, wherein the path of the actuable element passes through a gap in the coil.
7. The MEMS device of claim 5, wherein the coil is comprised of a conductive material having a resistivity less than approximately 1 x 107 Qm at 2 (P C.
8. The MEMS device of claim 5, wherein the coil is comprised of at least one of copper, aluminum, gold, silver, and alloys thereof.
9. The MEMS device of claim 5, wherein the coil has an inductance of at least approximately 50 nH.
10. The MEMS device of claim 5, wherein the electromagnetic actuator further comprises a magnetic core about which the electrically conductive coil is wound.
11. The MEMS device of claim 10, wherein the core has a magnetic permeability of at least approximately 4 x 103.
12. The MEMS device of claim 10, wherein the core is comprised of a soft magnetic material.
13. The MEMS device of claim 12, wherein the soft magnetic material includes at least one of NiFe (permalloy), CoFe (permendur), CoZr, FeN, AlSiFe (sendust), NiFeMo (supermalloy), NiFeCuCr (mumetal), NiFeCo, CoFeB, CoFeV (supermendur), CoFeCr (hiperco), CoZrTa, FeAIN, FeTaN, and combinations thereof.
14. The MEMS device of claim 10, wherein the core has a first core end and a second core end, the first core end and the second core end being separated by a core gap, the path of the actuable element extending at least partially into the core gap.
15. The MEMS device of claim 14, wherein the core gap has an extent in a direction substantially perpendicular to the path of at least approximately 5 im.
16. The MEMS device of claim 14, wherein the coil has a first coil end and a second coil end, the first coil end and the second coil end being separated by a coil gap, the coil gap having an extent equal to or greater than an extent of the core gap.
17. The MEMS device of claim 10, wherein the coil continuously covers at least approximately 80% of the surface area of the core.
18. The MEMS device of claim 10, wherein the coil includes at least ten windings about the core.
19. The MEMS device of claim 1, further comprising a clamp for selectively clamping the actuable element at a desired position on the path.
20. The MEMS device of claim 19, wherein the clamp is coupled to at least one of the MEMS actuator and the actuable element.
21. The MEMS device of claim 1, further comprising a stop for selectively inhibiting displacement of the actuable element beyond a desired position on the path.
22. The MEMS device of claim 21, wherein the stop is coupled to the substrate.
23. The MEMS device of claim 21, wherein the stop comprises a wall disposed on the substrate, the wall protruding at least partially into the path.
24. The MEMS device of claim 1, further comprising a control mechanism operable to selectively apply a second force to the actuable element in a direction opposite to the first force.
25. The MEMS device of claim 24, wherein the control mechanism is formed from the substrate.
26. The MEMS device of claim 24, wherein the control mechanism is coupled to the actuable element at one end of the substrate at another end.
27. The MEMS device of claim 24, wherein the control mechanism comprises at least one cantilever.
28. The MEMS device of claim 27, wherein the at least one cantilever extends substantially parallel to the surface of the substrate.
29. The MEMS device of claim 27, wherein the actuable element has a first side, and a plurality of cantilevers are coupled to the first side.
30. The MEMS device of claim 27, wherein the actuable element has a first side and a second side disposed opposite to the first side and at least one cantilever is coupled to the first side and at least one cantilever is coupled to the second side.
31. The MEMS device of claim 30, wherein at least one cantilever coupled to the first side of the actuable element is substantially coaxial with at least one cantilever coupled to the second side of the actuable element.
32. The MEMS device of claim 24, wherein the control mechanism comprises at least one spring coupled to the actuable element at one end and the substrate at another end.
33. The MEMS device of claim 24, wherein the control mechanism comprises a second MEMS actuator.
34. The MEMS device of claim 33, wherein the second MEMS actuator is a second electromagnetic MEMS actuator.
35. The MEMS device of claim 34, wherein the second electromagnetic MEMS actuator includes a magnetic core and an electrically conductive coil wound about the core.
36. The MEMS device of claim 1, wherein the MEMS device has an extent in a direction substantially perpendicular to the surface of the substrate of less than approximately 50 pm.
37. The MEMS device of claim 1, further comprising a suspension mechanism for selectively controlling the location of the actuable element in a direction substantially perpendicular to the surface of the substrate.
38. The MEMS device of claim 37, wherein the suspension mechanism comprises at least one of one or more clamps, one or more springs, one or more cantilevers, one or more permanent magnets, and one or more electromagnets.
39. The MEMS device of claim 1, wherein the actuable element includes an optical element for attenuating an optical beam lying in the path.
40. The MEMS device of claim 39, wherein the optical beam comprises at least one of a light beam and a particle beam.
41. The MEMS device of claim 39, wherein the optical beam lies in at least one of a plane substantially parallel to the surface of the substrate and a plane substantially perpendicular to the surface of the substrate.
42. The MEMS device of claim 1, wherein the optical element is a shutter for selectively blocking the optical beam.
43. The MEMS device of claim 42, wherein the shutter includes at least one of an opaque, a semitransparent, a semireflective, and a reflective surface.
44. The MEMS device of claim 27, wherein at least a portion of the cantilever is formed from the substrate.
45. A method of fabricating a MEMS device on a substrate, the method comprising: constructing an electromagnetic MEMS actuator on the surface of the substrate by building an electrically conductive coil on the surface of the substrate, the coil being arranged to form a gap between two ends of the coil, and forming an actuable element from a layer of the substrate at position on the substrate to facilitate displacement of at least a portion of the actuable element relative to the gap upon activation of the electromagnetic MEMS actuator.
46. The method of claim 45 further comprising constructing a cantilever on the substrate, the cantilever being coupled at one end to the substrate and at another end to the actuable element.
47. The method of claim 46, wherein constructing the cantilever comprises forming the cantilever from a layer of the substrate.
48. The method of claim 45, wherein constructing the electromagnetic MEMS actuator further includes building a core of magnetic material within the electrically conductive coil such that the coil winds about the core.
49. The method of claim 45, wherein forming the actuable element comprises: applying and patterning a magnetic layer of the actuable element on the substrate, applying and patterning a mask layer on the substrate to define a shape of the actuable element, etching a top layer of the substrate in accordance with the mask to form the actuable element, and releasing the actuable element from the substrate to permit displacement of the actuable element relative to the substrate.
50. The method of claim 49, wherein applying and patterning a magnetic layer of the actuable element on the substrate comprises applying a layer of magnetic material to a top layer of the substrate, applying a layer of a first material to the layer of magnetic material to form a first mask layer over the layer of magnetic material, applying a layer of a second material to the first mask layer to form a second mask layer, patterning the second mask layer to forma first window through the second mask layer to the first mask layer, patterning the first mask layer through the first window to form a second window through the first layer to the layer of magnetic material, removing the second mask, patterning the layer of magnetic material through the second window to remove magnetic material exposed through the second window, and removing the first mask.
51. The method of claim 50, wherein the first material is resistant to highpowered non reactive ion beam etching.
52. The method of claim 51, wherein the first material is at least one of alumina and silica.
53. The method of claim 50, wherein the second material is a photoresist material.
54. The method of claim 50, wherein patterning the layer of magnetic material includes nonreactive ion beam etching the layer of magnetic material.
55. The method of claim 50, wherein patterning the second mask layer includes reactive ion beam etching the second mask layer.
56. The method of claim 49, wherein releasing the actuable element from the substrate comprises etching away at least a portion of a second layer of the substrate beneath the top layer.
57. The method of claim 49, further comprising applying and patterning a spacer layer on the substrate before applying a layer of magnetic material on the substrate.
58. The method of claim 49, wherein the magnetic layer comprises at least one of ferrites, remalloy, vicalloy, AINiCo, Co, CoPt, a rare earth metal, NiFe (permalloy), CoFe (permendur), CoZr, FeN, AlSiFe (sendust), NiFeMo (supermalloy), NiFeCuCr (mumetal), NiFeCo, CoFeB, CoFeV (supermendur), CoFeCr (hiperco), CoZrTa, FeAlN, FeTaN, and combinations thereof.
59. The method of claim 45, wherein constructing the electromagnetic MEMS actuator comprises: applying a lower layer of electrically conductive material to substrate, patterning the lower layer to form a lower set of winding elements, applying a second layer of electrically conductive material over the lower set of winding elements, patterning the second layer to form connectors over the winding elements of the lower set of winding elements, applying a third layer of electrically conductive material over the connectors, and patterning the third layer to form an upper set of winding elements that are electrically connected to the winding elements of the lower set of winding elements by the connectors.
60. The method of claim 59, further comprising applying a metal seed layer on the substrate, wherein the first layer of electrically conductive material is applied to the metal seed layer.
61. The method of claim 60, further comprising applying a photoresist layer over metal seed layer.
62. The method of claim 61, further comprising patterning the photoresist layer to open windows in the photoresist layer to the metal seed layer.
63. The method of claim 62, wherein patterning the photoresist layer comprises patterning a mask defining the winding elements of the lower set of winding elements, exposing the photoresist layer through the mask, and developing the photoresist layer to open windows in the photoresist layer to the metal seed layer in the photoresist layer.
64. The method of claim 62, wherein applying the first layer of electrically conductive material comprises depositing an electrically conductive material on the metal seed layer in the windows defined by the photoresist layer to form the winding elements of the lower set of winding elements.
65. The method of claim 64, wherein the first layer of electrically conductive material is electroplated onto the metal seed layer in the windows defined by the photoresist layer.
66. The method of claim 65, wherein the electrically conductive material of the first layer is comprised of at least one of copper, gold, and silver, and alloys thereof.
67. The method of claim 64, further comprising removing the metal seed layer.
68. The method of claim 67, wherein removing the metal seed layer comprises etching away the metal seed layer in accordance with a mask to facilitate electrical isolation of the winding elements of the lower set of winding elements.
69. The method of claim 67, further comprising applying a layer of polymeric material between the winding elements of the lower set of winding elements.
70. The method of claim 69, further comprising heating the polymeric material to cross link the polymeric material.
71. The method of claim 69, wherein the polymeric material comprises at least on of baked photoresist, polyamide, and BCB.
72. The method of claim 60, wherein applying a metal seed layer comprises applying alternating layers of chrome and electrically conductive material to the substrate.
73. The method of claim 60, wherein applying a metal seed layer includes at least one of sputtering and vacuum depositing the metal seed layer.
74. The method of claim 59, further comprising applying an insulation layer over the winding elements of the first set of winding elements.
75. The method of claim 59, further comprising forming an oxide layer on a silicon substrate and forming a silicon device layer on the oxide layer, wherein the first layer of electrically conductive material is applied to the silicon device layer.
76. The method of claim 56, wherein etching away at least a portion of the second layer of the substrate comprises release etching at least a portion of the second layer of the substrate.
77. The method of claim 49, wherein releasing the actuable element from the substrate comprises etching away at least a portion of the top layer of the substrate.
78. The method of claim 77, wherein etching away at least a portion of the top layer of the substrate comprises deed silicon etching at least a portion of the top layer of the substrate.
79. The method of claim 45, further comprising cleaning the MEMS device with supercritical carbon dioxide.
80. A MEMS device comprising: a substrate, an actuable element, an actuator disposed on the surface of the substrate for selectively applying a first force to the actuable element to displace the actuable element along a displacement path adjacent at least one sidewall of a channel formed in the substrate, and a stop located on the at least one sidewall of the channel to restrict displacement of the actuable element along the path.
81. The MEMS device of claim 80, further comprising a second stop, wherein the stops are located on opposing faces of indented portion of the at least one sidewall.
82. The MEMS device of claim 81, wherein the actuable element comprises a stop arm extending from the actuable element in a direction towards the independent portion to a point between the stops on the opposing faces of the indented portion.
83. The MEMS device of claim 82, comprising at least one cantilever coupled to the substrate at a first end and coupled to the actuable element at a second end adjacent to the stop arm.
84. The MEMS device of claim 82, wherein at least one of the stops and the stop arm have contact surfaces configured to reduce stiction between the stops and the stop arm.
85. The MEMS device of claim 84, wherein the contact surfaces include at least one dimple.
86. The MEMS device of claim 82, wherein a distance between the actuable element and the at least one sidewall of the channel is generally constant along the displacement path.
87. The MEMS device of claim 80, further comprising a second stop, wherein the stops are located on opposing sidewalls of the channel and spaced apart in a direction along the displacement path.
88. The MEMS device of claim 87, wherein the actuable element comprises an offset portion between the stops, the offset portions having offset faces on opposite ends thereof, each offset face being in an opposed relationship with one of the stops.
89. The MEMS device of claim 88, comprising at least one cantilever coupled to the substrate at a first end and coupled to the actuable element at a second end adjacent the offset portion.
90. The MEMS device of claim 88, wherein at least one of the stops and the offset faces have contact surfaces configured to reduce stiction between the stops and the offset faces.
91. The MEMS device of claim 90, wherein the contact surfaces include at least one dimple.
92. The MEMS device of claim 88, wherein a distance between the actuable element and the sidewalls of the channel is generally constant along the displacement path.
93. The MEMS device of claim 80, further comprising a second stop, wherein the stops are located on opposing sidewalls of the channel, the stops extending from the sidewalls into the channel to reduce a width of the channel to less than a width of a first portion of the actuable element.
94. The MEMS device of claim 93, wherein the stops extend equally into the channel.
95. The MEMS device of claim 93, wherein the channel passes through a gap formed by spaced apart ends of a magnetic core of the actuator and the stops are located adjacent the gap.
96. The MEMS device of claim 95, wherein the stops extend equally into the channel.
97. The MEMS device of claim 93, wherein the first portion of the actuable element comprises magnetic material, a magnetic field being applied to the magnetic material within the gap by the actuator.
98. The MEMS device of claim 93, wherein the first portion comprises tabs extending from the actuable element towards the sidewalls of the channel.
99. The MEMS device of claim 98, wherein the tabs extend equally from opposing sides of the actuable element.
100. The MEMS device of claim 93, wherein at least one of the stops and the first portion of the actuable element have contact surfaces configured to reduce stiction between the stops and the first portion of the actuable element.
101. The MEMS device of claim 100, wherein the contact surfaces include at least one dimple.
102. A method of restricting displacement along a displacement path of an actuable element of a MEMS device so as to minimize derailing of the actuable element from a channel formed by at least one sidewall along the displacement path, comprising indenting the at least one sidewall to form opposed stop faces thereon, extending an arm portion of the actuable element to a position between the opposed stop faces, and maintaining a width between the at least one sidewall and the actuable element generally constant along the displacement path.
103. The method of claim 102, comprising minimizing stiction between the stop faces and the arm portion of the actuable element.
104. The method of claim 103, comprising dimpling at least one of the stop faces and the arm portion of the actuable element.
105. The method of claim 103, comprising applying an antistiction coating to at least one of the stop faces and the arm portion of the actuable element.
106. A MEMS device comprising: an actuable element having a magnetic portion comprising a magnetic material, and an electromagnetic MEMS actuator comprising an electrically conductive coil wound about a magnetic core, the coil and the magnetic core being arranged to generate a magnetic field within a gap formed by spaced apart ends of the magnetic core upon application of a current to the coil, the actuable element being displaceable along a displacement axis relative to the gap upon application of the magnetic field on the magnetic portion of the actuable element, wherein at least one of the gap and the magnetic portion varies along the displacement axis.
107. The MEMS device of claim 106, wherein the displacement axis passes through the gap.
108. The MEMS device of claim 106, wherein the ends of the core are spacedapart about a gap distance to form the gap and the gap distance varies along the displacement axis.
109. The MEMS device of claim 108, wherein the gap distance decreases along the displacement axis.
110. The MEMS device of claim 108, wherein the gap distance increases along the displacement axis.
111. The MEMS device of claim 106, wherein the magnetic portion has a width and the width varies along the displacement axis.
112. The MEMS device of claim 111, wherein the width increases along the displacement axis in a direction corresponding to displacement of the actuable element away from the gap.
113. The MEMS device of claim 111, wherein the width decreases along the displacement axis.
114. The MEMS device of claim 106, wherein the magnetic portion has a first angled surface oriented at an angle to the displacement axis.
115. The MEMS device of claim 114, wherein the magnetic portion has a second angled surface oriented at an angle to the displacement axis.
116. The MEMS device of claim 115, wherein the second angled surface and the first angled surface are symmetrically oriented about the displacement axis.
117. The MEMS device of claim 116, wherein the angle is greater than 45.
118. The MEMS device of claim 116, wherein the angle is greater than 70.
119. The MEMS device of claim 114, wherein the angle is greater than 45.
120. The MEMS device of claim 114, wherein the angle is greater than 70.
121. The MEMS device of claim 114, wherein the magnetic portion includes a surface oriented parallel to the displacement axis.
122. A MEMS device comprising: a substrate, an actuable element, an actuator disposed on the surface of the substrate for selectively applying a first force to the actuable element to displace the actuable element along a displacement path, at least one control mechanism coupled to the actuable element at a first end and coupled to the substrate at a second end to control displacement of actuable element along the displacement path, and a pin coupled to the substrate and positioned in a deflection path of the control mechanism, the pin altering a bending resistance of the control mechanism upon contact with the cantilever.
123. The MEMS device of claim 122, wherein the pin has a contact surface configured to minimize stiction between the pin and the control mechanism.
124. The MEMS device of claim 123, wherein the contact surface is coated with anti stiction coating.
125. The MEMS device of claim 123, wherein the contact surface is textured.
126. The MEMS device of claim 123, wherein the contact surface is arcuate in shape.
127. The MEMS device of claim 122, wherein the pin includes at least one of an arcuate shape, a cylindrical shape, and a polygonal crosssection.
128. The MEMS device of claim 127, wherein the pin includes a polygonal crosssection, the control mechanism is a cantilever, and the cantilever contacts the pin at two or more points on the polygonal crosssection.
129. The MEMS device of claim 122, wherein the control mechanism is a cantilever and the pin is positioned a distance D from the first end of the cantilever, the distance D being selected to increase the bending resistance of the cantilever.
130. The MEMS device of claim 122, wherein the control mechanism is a cantilever and the pin is positioned a distance Dx from the displacement path and a distance Dy from a longitudinal axis of the cantilever.
131. The MEMS device of claim 122, further comprising at least one additional pin coupled to the substrate and positioned in a deflection path of the control mechanism.
132. The MEMS device of claim 122, wherein the control mechanism is a cantilever and the cantilever has a crosssection defined by a cantilever width W and a cantilever height H and the crosssection of the cantilever varies along a longitudinal axis of the cantilever by varying at least one of the cantilever width W and the cantilever height H along the longitudinal axis of the cantilever.
Description:
MEMS DEVICES AND METHODS OF MANUFACTURE Background [001] Advances in semiconductor fabrication technology have enabled the creation of Micro Electro Mechanical (MEMS) devices capable of motion and applying a force at the micron level. MEMS devices having components and features sizes on the micron scale have been fabricated. Such MEMS devices may be employed in a variety of applications including, for example, electrical, fluidic, mechanical, and optical applications, and combinations thereof.

Summary [002] Disclosed herein are MEMS devices that include electromagnetic and other actuators for selectively generating displacement forces. The MEMS devices disclosed herein may be employed in many applications, including, for example, to generate displacement forces in electrical, fluidic, mechanical, and optical systems, and combinations thereof. In particular, the MEMS devices disclosed herein may be utilized to control electrical, optical, and electro-optical components in an optical fiber based system, such as an optical fiber communication system or an optical fiber computer system.

[003] According to one exemplary embodiment disclosed herein, a MEMS device may include a substrate having a surface, an actuable element at least partially formed from the substrate, and an electromagnetic actuator disposed on the substrate for selectively applying a first force to the actuable element to displace the actuable element along a path. The actuable element may have a base and an arm coupled to the base. The base may include a portion comprised of a magnetic material. The electromagnetic actuator may comprise an electrically conductive coil, and the path of the actuable element may pass through a gap in the coil. The electromagnetic actuator may also comprise a magnetic core about which the electrically conductive coil may be wound.

[004] In one aspect, the actuable element may include an optical element for attenuating an optical beam lying in the path. The optical element may include a shutter for selectively blocking an optical beam. The shutter may include at least one of an opaque, a semi- transparent, a semi-reflective, and a reflective surface.

[005] According to another exemplary embodiment disclosed herein, a MEMS device may include a control mechanism operable to selectively apply a second force to the actuable element in a direction opposite to the first force. The control mechanism may be

formed from the substrate. The control mechanism may be coupled to the actuable element at one end and to the substrate at another end. The control mechanism may include at least one of one or more stops, one or more clamps, one or more cantilevers, one or more springs, and one or more MEMS actuators.

[006] According to another exemplary embodiment disclosed herein, a MEMS device may include a suspension mechanism for selectively controlling the location of the actuable element in a direction substantially perpendicular to the surface of the substrate. The suspension mechanism may include at least one of one or more clamps, one or more springs, one or more cantilevers, one or more permanent magnets, and one or more electromagnets.

[007] According to another exemplary embodiment disclosed herein, a method for fabricating a MEMS device on a substrate may include constructing an electromagnetic MEMS actuator on the surface of the substrate by building an electrically conductive coil on the surface of the substrate, the coil being arranged to form a gap between two ends of the coil, and forming an actuable element from a layer of the substrate at a position on the substrate to facilitate displacement of at least a portion of the actuable element relative to the gap upon activation of the electromagnetic MEMS actuator.

[008] In one aspect, a method for fabricating a MEMS device on a substrate may further include constructing a cantilever on the substrate, the cantilever being coupled at one end to the substrate and at another end to the actuable element.

[009] In one aspect, forming an actuable element may include applying and patterning a magnetic layer of the actuable element on the substrate, applying and patterning a mask layer on the substrate to define a shape of the actuable element, etching a top layer of the substrate in accordance with the mask to form the actuable element, and releasing the actuable element from the substrate to permit displacement of the actuable element relative to the substrate.

[010] In one aspect, applying and patterning a magnetic layer of the actuable element on the substrate may comprise applying a layer of magnetic material to a top layer of the substrate, applying a layer of a first material to the layer of magnetic material to form a first mask layer over the layer of magnetic material, applying a layer of a second material to the first mask layer to form a second mask layer, patterning the second mask layer to form a first window through the second mask layer to the first mask layer, patterning the first mask layer through the first window to form a second window through the first layer to the layer

of magnetic material, removing the second mask, patterning the layer of magnetic material through the second window to remove magnetic material exposed through the second window, and removing the first mask.

[0111 In one aspect, constructing an electromagnetic MEMS actuator may include applying a lower layer of electrically conductive material to the substrate, patterning the lower layer to form a lower set of winding elements, applying a second layer of electrically conductive material over the lower set of winding elements, patterning the second layer to form connectors over the winding elements of the lower set of winding elements, and applying a third layer of electrically conductive material over the connectors, and patterning the third layer to form an upper set of winding elements that are electrically connected to the winding elements of the lower set of winding elements by the connectors.

[012] Also disclosed herein are MEMS devices and methods of manufacturing MEMS devices that can minimize errant motion of the components of a MEMS device, particularly during fabrication of the MEMS device.

[013] In one exemplary embodiment, the MEMS device may include a substrate, an actuable element, an actuator disposed on the surface of the substrate for selectively applying a first force to the actuable element to displace the actuable element along a displacement path adjacent at least one sidewall of a channel formed in the substrate, and a stop located on the at least one sidewall of the channel to restrict displacement of the actuable element along the path during fabrication. In one embodiment having a second stop, the stops may be located on opposing faces of an indented portion of the at least one sidewall and the actuable element can comprise a stop arm extending from the actuable element in a direction towards the indented portion to a point between the stops on the opposing faces of the indented portion. At least one cantilever, for controlling the displacement of the actuable element along the displacement path during operation of the MEMS device, can couple to the substrate at a first end and couple to the actuable element at a second end adjacent the stop arm. A distance between the actuable element and the at least one sidewall of the channel can be maintained generally constant along the displacement path.

[014] In a further exemplary embodiment, a MEMS device can have stops located on opposing sidewalls of the channel and spaced apart in a direction along the displacement path. The actuable element can comprise an offset portion between the stops and the offset portion can have offset faces on opposite ends thereof, such that each offset face may be in

an opposed relationship with one of the stops. A distance between the actuable element and the sidewalls of the channel can be maintained generally constant along the displacement path.

[015] In embodiments including a pair of stops, the stops can be located on opposing sidewalls of the channel and can extend from the sidewalls into the channel to reduce a width of the channel to less than a width of a first portion of the actuable element.

[016] The stops may extend equally from each side of the channel and may be located adjacent a gap formed by the ends of the magnetic core of the MEMS actuator. The first portion can comprise magnetic material, to which the actuator can apply a magnetic field, or the first portion can comprise tabs extending from the actuable element towards the sidewalls of the channel. The tabs can extend equally from opposing sides of the actuable element. The stops and the first portion of the actuable element can have contact surfaces configured to reduce stiction between the stops and the first portion of the actuable element, such as by dimpling or anti-stiction coatings.

[017] A method of restricting displacement along a displacement path of an actuable element of a MEMS device so as to minimize derailing of the actuable element from a channel formed by at least one sidewall along the displacement path can comprise indenting the at least one sidewall to form opposed stop faces thereon, extending an arm portion of the actuable element to a position between the opposed stop faces and maintaining a width between the at least one sidewall and the actuable element generally constant along the displacement path.

[018] Also disclosed herein are MEMS devices and methods of manufacturing MEMS devices that minimize the effect of fabrication tolerances on the properties of the devices.

[019] In one exemplary embodiment, the MEMS device may include an actuable element having a magnetic portion comprising a magnetic material, and an electromagnetic MEMS actuator that includes an electrically conductive coil wound about a magnetic core. The coil and the magnetic core can be arranged to generate a magnetic field within a gap formed by spaced apart ends of the magnetic core upon application of a current to the coil. The actuable element may be displaceable along a displacement axis relative to the gap upon application of the magnetic field on the magnetic portion of the actuable element. The gap and/or the magnetic portion of the actuable element can vary along the displacement axis so as to vary the force applied to the actuable element along the displacement axis.

[020] In another exemplary embodiment, a MEMS device can include an actuable element having a magnetic portion comprising a magnetic material, and an electromagnetic MEMS actuator comprising an electrically conductive coil wrapped around a core. The core can include a first end spaced apart from a second end by a gap distance to form a gap. The actuable element may be displaceable along a displacement axis that passes through the gap upon application of a magnetic field generated by the electromagnetic MEMS actuator on the magnetic portion of the actuable element. One or both of the core ends can be angled with reference to the displacement axis to vary a gap distance between the core ends. The angle with respect to the displacement axis can be greater than 45° and may be greater than 70°.

[021] In a further exemplary embodiment, a MEMS device can include an actuable element having a magnetic portion comprising a magnetic material, and an electromagnetic MEMS actuator comprising an electrically conductive coil wrapped about a magnetic core.

The actuator can be arranged to generate a magnetic field within a gap formed by spaced apart ends of the magnetic core upon application of a current to the coil. The actuable element may be positioned proximate the gap and may be displaceable along a displacement axis relative to the gap upon application of a magnetic field on the magnetic portion of the actuable element. The magnetic portion can have a first angled surface oriented at an angle to the displacement axis.

[022] In certain exemplary embodiments, the magnetic portion can have a second angled surface oriented at an angle to the displacement axis and the first and second angled surfaces can be symmetrically oriented about the displacement axis. The angle with respect to the displacement axis can be greater than 45° and may be greater than 70°. The magnet portion can include a surface oriented parallel to the displacement axis.

[023] In another embodiment, a MEMS device can include a substrate, an actuable element, an actuator disposed on the surface of the substrate for selectively applying a first force to the actuable element to displace the actuable element along a displacement path, at least one control mechanism, such as, for example, a cantilever, coupled to the actuable element at a first end and coupled to the substrate at a second end to control displacement of the actuable element along the displacement path and a pin coupled to the substrate and positioned in a deflection path of the control mechanism, the pin altering the deflection path of at least a portion of the control mechanism upon contact with the control mechanism.

[024] The pin can have a contact surface configured to minimize stiction between the pin and the control mechanism, e. g. , the cantilever. The contact surface can be coated with an anti-stiction coating and may be textured. The contact surface can be arcuate or cylindrical in shape or may have a polygonal cross-section. The pin can be positioned a distance D from the first end of the cantilever that is selected to increase the spring constant of the cantilever. The pin can be positioned a distance Dx from the displacement path and a distance Dy from a longitudinal axis of the cantilever. The MEMS device can further include a second pin coupled to the substrate and positioned in a deflection path of the cantilever. The cross-section of the cantilever, defined by a cantilever width W and a cantilever height H, can vary along the longitudinal axis of the cantilever by varying the width W and/or the height H along the longitudinal axis of the cantilever.

Brief Description of the Drawings [025] These and other features of the MEMS devices disclosed herein will be more fully understood by reference to the following detailed description in conjunction with the attached drawings. In the attached drawings, like reference numerals refer to like parts throughout the different views, and reference numerals that differ by increments of 100 refer to similar parts in different embodiments. While the drawings illustrate principles of the MEMS devices disclosed herein, they are not drawn to scale, but show only relative dimensions.

[026] Fig. 1A is a top view of an embodiment of a MEMS device disclosed herein.

[027] Fig. 1B is a side elevation view of the MEMS device shown in Fig. 1A along the line lA-lA'.

[028] Fig. 1C is a side elevation view of the MEMS device shown in Fig. 1A along the line 1B-1B'.

[029] Fig. 2A is a top view of an embodiment of a MEMS actuator disclosed herein.

[030] Fig. 2B is a side elevation view of the MEMS actuator shown in Fig. 2A along the line 2A-2A'.

[031] Fig. 2C is a side elevation view of the MEMS actuator shown in Fig. 2A along the line 2B-2B'.

[032] Fig. 2D is a partial top view of the MEMS actuator shown in Fig. 2A, enlarged to show detail.

[033] Fig. 3A is a top view of the MEMS actuator shown in Fig. 2A, illustrating a first state of operation.

[034] Fig. 3B is a top view of the MEMS actuator shown in Fig. 2A, illustrating a second state of operation.

[035] Fig. 4A is a partial top view of the MEMS actuator shown in Fig. 2A, enlarged to show an exemplary embodiment of the MEMS actuable element disclosed herein.

[036] Fig. 4B is a partial top view of the MEMS actuator shown in Fig. 2A, enlarged to show another exemplary embodiment of the MEMS actuable element disclosed herein.

[037] Fig. 5 is a top view of the MEMS device shown in Fig. 1A, illustrating an embodiment of a stop disposed in the path of an actuable element.

[038] Fig. 6 is a top view of the MEMS device shown in Fig. 1A, illustrating an embodiment of a clamp.

[039] Fig. 7A is a top view of another embodiment of a MEMS device disclosed herein, illustrating an embodiment of a double-cantilever structure.

[040] Fig. 7B is a top view of another embodiment of a MEMS device disclosed herein, illustrating another embodiment of a double-cantilever structure.

[041] Fig. 8A is a top view of another embodiment of a MEMS device disclosed herein, illustrating an embodiment of a spring.

[042] Fig. 8B is a top view of another embodiment of a MEMS device disclosed herein, illustrating another embodiment of a spring.

[043] Fig. 9A is a schematic view of a magnetic suspension system.

[044] Fig. 9B is a schematic view of an electromagnetic suspension system.

[045] Fig. 10 is a top view of another embodiment of a MEMS device disclosed herein, the embodiment having two MEMS actuators and a single actuable element.

[046] Fig. 11 is a top view of another embodiment of a MEMS device disclosed herein, the embodiment having two MEMS actuators and two actuable elements.

[047] Fig. 12 is a top view of another embodiment of a MEMS device disclosed herein, illustrating an optical attenuator.

[048] Fig. 13 is a top view of another embodiment of a MEMS device disclosed herein, illustrating another embodiment of an optical attenuator.

[049] Fig. 14 is a top view of another embodiment of a MEMS device disclosed herein, illustrating an electrical switch.

[050] Fig. 15A is a top view of another embodiment of a MEMS device disclosed herein, illustrating an optical switch.

[051] Fig. 15B is a side elevation view of the actuable element of a MEMS device shown in Fig. 15A along the line 15A-15A'.

[052] Fig. 16A is a top view of another embodiment of a MEMS device disclosed herein, illustrating a fluid valve.

[053] Fig. 16B is a top view of another embodiment of a MEMS device disclosed herein, illustrating a fluid pump.

[054] Figs. 17A-E are flowcharts illustrating an exemplary method of fabricating the MEMS devices disclosed herein.

[055] Fig. 18A is a top view of an exemplary embodiment of a MEMS device disclosed herein.

[056] Fig. 18B is a side elevation view of the MEMS device shown in Fig. 18A along the line 18A-18A'.

[057] Fig. 18C is a side elevation view of the MEMS device shown in Fig. 18A along the line 18B-18B'.

[058] Fig. 19 is a top view of an embodiment of a MEMS device disclosed herein.

[059] Fig. 20 is an enlarged top partial view of an embodiment of a MEMS device disclosed herein.

[060] Fig. 21 is an enlarged top partial view of an embodiment of a MEMS device disclosed herein.

[061] Fig. 22A is a top view of an embodiment of a MEMS device disclosed herein.

[062] Fig. 22B is a side elevation view of the MEMS device shown in Fig. 22A along the line 22A-22A'.

[063] Fig. 22C is a side elevation view of the MEMS device shown in Fig. 22A along the line 22B-22B'.

[064] Figs. 23A-23B are plots of the relationship of forces acting on an actuable element of a MEMS device to the displacement of the actuable element.

[065] Fig. 24A is a partial top view of the MEMS device of Fig. 22A, enlarged to show detail.

[066] Fig. 24B is a partial top view of an exemplary embodiment of the MEMS device of Fig. 22A, enlarged to show detail.

[067] Fig. 25 is a partial top view of the MEMS device of Fig. 22A, enlarged to show detail.

[068] Fig. 26A is a top view of an embodiment of a MEMS device disclosed herein.

[069] Fig. 26B is a side elevation view of the MEMS device shown in Fig. 26A along the line 26B-26B'.

[070] Fig. 26C is a top view of another embodiment of the MEMS device.

[071] Fig. 27A is a schematic representation of a cantilever of the MEMS device subjected to a first force.

[072] Fig. 27B is a schematic representation of the cantilever subjected to an increased force.

[073] Fig. 27C is a flow chart of a method for determining the placement of a means for altering a spring constant of the cantilever.

Detailed Description of Certain Exemplary Embodiments [074] Certain exemplary embodiments will now be described to provide an overall understanding of the principles of the MEMS devices and methods of fabrication disclosed herein. One or more examples of these embodiments are illustrated in the drawings. Those of ordinary skill in the art will understand that the MEMS devices and methods of fabrication disclosed herein can be adapted and modified to provide instruments and methods for other applications and that other additions and modifications can be made without departing from the scope of the present disclosure. For example, the features illustrated or described as part of one embodiment or one drawing can be used on another embodiment or another drawing to yield yet another embodiment. Such modifications and variations are intended to be included within the scope of the present disclosure.

[075] An exemplary embodiment of a MEMS device disposed on a substrate is shown in Figs. lA, 1B, and 1C. The MEMS device 10 includes a MEMS actuator 12 and an actuable element 14. In the exemplary embodiment, the MEMS actuator 12 is an electromagnetic actuator, as shown in Figs. 2A, 2B, 2C, and 2D and described below. Alternatively, the MEMS actuator may be another type of MEMS actuator such as, for example, an electrostatic, thermal, or piezoelectric actuator. The illustrated MEMS actuator includes an electrically conductive coil arranged to generate a magnetic field with a gap 8 formed by the spaced-apart ends of the actuator 12 upon application of a current to the coil. The MEMS electromagnetic actuator 12 is coupled to a first electrically conducting lead 18 and a second electrically conducting lead 19 that terminate in a first electrode pad 44 and a second electrode pad 46, respectively.

[076] The actuable element 14 is shaped to include a base 20 and an elongated arm 22 coupled to and extending from the base 20. As discussed below, the base 20 or other

portion of the actuable element 14 may include a layer or portion of magnetic material. The base 20 of the actuable element 14 may be positioned proximate the gap 8 such that the base 20 can be displaced relative to the gap 8 upon application of a magnetic field on the actuable element 14. Optionally, a cantilever 24 or other mechanism for controlling the displacement of the actuable element may be coupled to the actuable element 14 at one end and to the substrate 54 at another end.

[077] In the exemplary embodiment illustrated in Figs. 1A, 1B, and 1C, the MEMS actuator 12 is constructed on the surface of the substrate 54 and the actuable element 14 and the cantilever 24 is constructed from the substrate 54, preferably from a top device layer 52 of the substrate 54. In this manner, the actuable element 14 and the cantilever 24 may be suspended over a cavity 16 (denoted by crosshatching in Fig. 1A) in the substrate 54 and are free to be displaced relative to the substrate 52. Using the manufacturing processes described below, all or at least a portion of the components of the MEMS device, e. g. , the actuator 12, the actuable element 14 and/or the cantilever 24 (or other control mechanism), may be constructed from one or more layers of the substrate 54 to reduce the extent of the MEMS device 10 in a direction perpendicular from the substrate surface. Alternatively, all or portions of the components of the MEMS device 10 may be constructed on the surface of the substrate 54 through deposition processes or other additive processes.

[078] The MEMS device 10 provides precise electronic control of the position of the actuable element 14. Operation of the MEMS device 10 can be understood in the following manner. An external power source (not shown) provides an operating current to the first and second electrode pads 44,46. The first and second electrode pads 44,46 communicate the operating current to the actuator 12 through the first and second leads 18, 19, respectively. After receiving an operating current, the actuator 12 may selectively apply a first force to the actuable element 14. The first force tends to displace the actuable element 14 relative to the gap 8 along a path 6 that is substantially parallel to the axis of actuation 26. The cantilever 24 selectively applies a second force to the actuable element 14. Like the first force, the second force also acts along the axis of actuation 26. Unlike the first force, however, the second force acts in an opposite direction, and tends to restore the actuable element 14 to its original position on the path 6 and limit the displacement of actuable element 14 along the path 6. To effect displacement of the actuable element 14, an operating current is applied to the MEMS actuator 12 to generate a first force that has a magnitude greater than that of the second force. Upon removal of the operating current, the

second force restores the actuable element 14 to its original position on the path 6. By varying the operating current to the MEMS actuator 12, the MEMS device 10 provides precise electronic control of the displacement of the actuable element 14 along the path 6.

[079] As shown in Figs. 1A, 1B, and 1C, each of the path 6, the actuator 12, and the actuable element 14 lies in a plane that is substantially parallel to the substantially planar surface of the substrate 54. Alternately, any combination of the path 6, the actuator 12, and the actuable element 14 may lie in a plane that is inclined at an angle to the substantially planar surface of the substrate 54. Also alternately, the path 6 may include constituent portions that are not coplanar with each other, the actuator 12 may include constituent portions that are not coplanar with each other, and the actuable element 14 may include constituent portions that are not coplanar with each other. Additionally, the path 6, the actuator 12, and the actuable element 14 may not be coplanar with each other.

[080] An exemplary embodiment of the actuator 12 is shown in Figs. 2A, 2B, 2C, and 2D.

The actuator 12 includes a magnetic core 30 and an electrically conductive coil 32 wrapped around the core 30 in a manner analogous to a solenoid or a toroid. Alternately, the actuator 12 may include an electrically conductive coil 32 without a magnetic core 30.

Preferably, however, the actuator 12 includes a magnetic core 32 to enhance the strength of the magnetic field generated at the core gap 38 for a given operating current applied to the actuator 12. Preferably, the actuator 12 is substantially planar, and lies in a plane that is substantially parallel to the substantially planar surface of the substrate.

[081] As shown in Figs. 2A, 2B, 2C, and 2D, the core 30 has a substantially circular shape when viewed from above and a substantially rectangular cross-section that has an extent that is substantially constant across the circumference of the core 30. Alternately, the core 30 may have a variety of shapes when viewed from above and a variety of cross-section shapes. For example, these shapes may include the shape of any type of oval, including an ellipse and a circle, any type of polygon, including a rectangle, a square, and a triangle, and the shape of any type of semi-oval, including a semi-circle. Also alternately, the cross- section may have an extent that varies across the circumference of the core 30.

[082] The core 30 is constructed from magnetic material. The core may be constructed from a material having a magnetic permeability of at least 4 x 103. Preferably, the magnetic permeability of the core material is between approximately 8 x 103 and approximately 1 x 106. Preferably, the core is constructed from a soft magnetic material, including, but not limited to, NiFe (permalloy), CoFe (permendur), CoZr, FeN, AlSiFe (sendust) and alloys

thereof, such as NiFeMo (supermalloy), NiFeCuCr (mumetal), NiFeCo, CoFeB, CoFeV (supermendur), CoFeCr (hiperco), CoZrTa, FeAIN, and FeTaN.

[083] As shown in Figs. 2A and 2D, the core 30 has a first core end 34 and a second core end 36 separated from each other by a core gap 38. The core gap 38 has a size and shape that permits the actuable element 14 to be displaced therethrough. The size and shape of the core gap 38 depend on a variety of core parameters including the core shape and the core composition. The size and shape of the core gap 38 also depend upon the limitations of the fabrication process. Preferably, the core gap 38 has an extent in a direction substantially perpendicular to the path of at least approximately 5 um.

[084] As shown in Figs. 2A and 2D, the electrically conductive coil 32 has a number of coil windings 33, a first coil end 40 located near the first core end 34, and a second coil end 42 located near the second core end 36. As shown in Figs. 2B and 2C and explained in greater detail below, each winding 33 of the coil 32 includes an upper winding element 32a, a lower winding element 32b, a first connector 32c, and a second connector 32d. As shown in Fig. 2D, the first coil end 40 is connected to first lead 18 (which terminates in first electrode pad 44 as shown in Fig. 1A), and the second coil end 42 is connected to second lead 19 (which terminates in second electrode pad 46 as shown in Fig. 1A). The first and second electrode pads 44,46 are connected to a source of external power (not shown) for providing an operating current to the actuator 12. The source of external power may be an analog DC power source, which may be used to provide fine control of the position of the actuable element 14. The source of external power may also include any other source of external power known by one of ordinary skill in the art to be suitable for a MEMS device.

[085] As shown in Figs. 2A, the first coil end 40 is separated from the second coil end 42 by a coil gap 50. The coil gap 50 may have an extent that is greater than an extent of the core gap 38. Alternately, the coil gap 50 may have an extent that is at least equal to an extent of the core gap 38.

[086] In the actuator 12, the gap 8 through which the actuable element 14 can be displaced may be defined by the core gap 38 and/or the coil gap 50. In the embodiments illustrated in Figs. 2A-2D, the gap 8 in the actuator 12 is defined by core gap 38.

[087] As shown in Figs. 2A, 2B, 2C, and 2D, the coil 32 has a substantially circular shape when viewed from above and a substantially rectangular cross-section that has an extent that is substantially constant across the circumference of the coil 32. Alternately, the coil 32 may have a variety of shapes when viewed from above and a variety of cross-section

shapes, provided that the shapes of the coil 32 are compatible with the shapes of the core 30. For example, these shapes may include the shape of any type of oval, including an ellipse and a circle, any type of polygon, including a rectangle, a square, and a triangle, and the shape of any type of semi-oval, including a semi-circle. Also alternately, the cross- section may have an extent that varies across the circumference of the coil 32. The coil 32 may have an inductance of at least approximately 50 nH, and preferably between approximately 100 nH and approximately 1 MH.

[088] The coil 32 is constructed from an electrically conductive material. Preferably, the coil 32 is constructed from an electrically conductive material having a resistivity of less than approximately 1 x 10-7 Qm at 20° C. Preferably, the coil 32 is constructed from copper or an alloy thereof. Alternately, the coil 32 may be constructed from aluminum, gold, silver, and alloys thereof.

[089] The physical dimensions of the core 30 and the coil 32 depend on a variety of parameters. Generally, the shape of the core 30 may be optimized by finite-element modeling to reduce saturation and flux leakage. The number of windings 33 in the coil 32, the width of the core gap 38, and the cross-sectional areas of the first and second core ends 34,36 can be varied to adjust the strength of the magnetic field generated at the core gap 38. The coil 32 may have at least ten windings 33, and preferably greater than fifty windings 33.

[090] An exemplary embodiment of the actuable element 14 is shown in Fig. 2D. In the illustrated exemplary embodiment, the base 20 of the actuable element 14 includes a front portion 70 constructed from a magnetic material. The front portion 70 is positioned to provide an interface between the base 20 and the arm 22. The front portion 70 may, however, be positioned elsewhere on the base 20. Additionally, the entire base 20 may be constructed from magnetic material, rather than just the front portion 70. The front portion 70 may be constructed from a permanent magnetic material, including, but not limited to, ferrites, remalloy, vicalloy, AlNiCo, Co, CoPt, a rare earth metal, and combinations thereof.

In this embodiment, the actuable element 14 has a permanent magnetic dipole moment illustrated by arrow 84. The front portion 70 may also be comprised of a soft magnetic material, including, but not limited to, NiFe (permalloy), CoFe (permendur), CoZr, FeN, AlSiFe (sendust) and alloys thereof, such as NiFeMo (supermalloy), NiFeCuCr (mumetal), NiFeCo, CoFeB, CoFeV (supermendur), CoFeCr (hiperco), CoZrTa, FeAlN, and FeTaN.

In the case of soft magnetic materials, the actuable element 14 does not have a permanent magnetic dipole moment.

[091] Displacement of the actuable element 14 along the path 6 through the core gap 38 depends, at least in part, upon the relationship between the magnetic field generated in the core gap 38 by the actuator 12 and the magnetic dipole properties of the front portion 70.

An exemplary mode of operation of the MEMS device 10 may be understood in the following manner. Referring to Figs. lA, 2A, and 2D, an external power source (not shown) provides an operating current between the first and second coil ends 40,42 via the first and second leads 18,19 and the first and second electrode pads 44,46. The current 80 passing through the coil 32 generates a magnetic field 82 at the core gap 38. Depending on the relationship between the magnetic field 82 and the magnetic dipole properties of the front portion 70, the actuable element 14 may experience an attractive or repulsive force that tends to pull or push the actuable element 14 along the path 6 towards or away from the core gap 38, respectively.

[092] The displacement of an actuable element 14 in which the front portion 70 has a permanent magnetic dipole moment 84 can be controlled by controlling the direction of the magnetic field 82 in the core gap 38. The operating principles of controlling the direction of the magnetic field 82 in the core gap 38 are shown in Figs. 3A and 3B. The current may have a first state and a second state, as indicated by arrow 80 in Figs. 3A and 3B. As shown in Fig. 3A, in the first state, current 80 passes clockwise through the coil 32 from the first coil end 40 to the second coil end 42. Passing current 80 through the coil 32 in the first state generates a magnetic field 82 having a first direction 82a that is substantially parallel to the direction of the magnetic moment 84 in the front portion 70. As a result, in the first state, the actuable element 14 experiences an attractive electromagnetic force tending to pull the actuable element toward the core gap 38. As shown in Fig. 3B, in the second state, current 80 passes counterclockwise through the coil 32 from the second coil end 42 to the first coil end 40. Passing current 80 through the coil 32 in the second state generates a magnetic field 82 having a second direction 82b that is opposite to the first direction 82a and that is also substantially antiparallel to the direction of the magnetic moment 84 in the front portion 70. As a result, in the second state, the actuable element 14 experiences a repulsive electromagnetic force tending to push the actuable element 14 away from the core gap 38.

[093] Based on the foregoing principles, displacement of the actuable element 14 having a front portion 70 with a permanent magnetic dipole moment 84 may be controlled in accordance with the following procedure. First, current 80 may be applied in the first state to the actuator 12. Second, current 80 may be maintained in the first state until the actuable element 14 is displaced to a desired position on the path 6. Third, current 80 may be removed from the first state, and current 80 may be applied in the second state to restore the actuable element 14 to its original position on the path 6. Accordingly, displacement of an actuable element 14 having a portion 70 with a permanent magnetic dipole moment 84 may be precisely controlled without any of the optional control mechanisms, including a clamp, a cantilever, a spring, a suspension system, or a second MEMS actuator, described below.

[094] This procedure does not require that the front portion 70 have a permanent magnetic dipole moment 84 aligned or antialigned with the magnetic field 82 in the core gap 38.

Rather, this procedure minimally requires a permanent magnetic dipole moment 84 having a component that lies substantially along the direction of the magnetic field 82 and that is sufficient in magnitude to generate a useful electromagnetic force for displacing the actuable element 14.

[095] As provided previously herein, a front portion 70 comprised of soft magnetic material does not have a permanent magnetic dipole moment 84. As a result, passing current through the coil 32 in both the first and second states tends to produce an attractive force, rather than an attractive force in the first state and a repulsive force in the second state. Displacement of a front portion 70 comprised of soft magnetic material thus tends to require additional control mechanisms. A variety of suitable control mechanisms are provided below.

[096] As suggested in Figs. 1A, 2A, and 2D, the front portion 70 may have a variety of shapes. Preferably, the front portion 70 has a shape and a material composition that tends to generate a desired relationship between the displacement of the actuable element 14 and the current 80 passing through the coil 32. This relationship may be linear or non-linear depending on the desired application of the MEMS device 10. A front portion 70 constructed from a permanent magnetic tends to generate an approximately linear relationship, and a portion 70 constructed from a soft magnetic material tends to generate a non-linear relationship.

[097] An exemplary preferred shape of a front portion 70 comprised of a permanent magnetic material for generating an approximately linear relationship is shown in Fig. 4A.

The front portion 70 may have a length 90 in a direction substantially parallel to the path 6 and a cross-section that has an extent that is substantially constant along the length 90. A front portion 70 constructed from a permanent magnetic material and having this shape tends to generate an approximately linear relationship between displacement and current 80.

Increasing the length 90 of magnetic material in the front portion 70 may result in a more exactly linear relationship between displacement and current 80.

[098] An exemplary shape of a front portion 70 comprised of a permanent magnetic material for generating a non-linear relationship is shown in Fig. 4B. The front portion 70 may have a length 90 in a direction substantially parallel to the path 6 and a cross-section that has an extent that varies along the length 90. A front portion 70 constructed from a permanent magnetic material and having this tapered shape tends to generate a non-linear relationship between displacement and current 80. A front portion 70 constructed from a soft magnetic material may be preferred, however, for generating a non-linear relationship.

[099] A variety of control mechanisms are available for controlling the displacement of the actuable element 14 along the path 6 by restoring the actuable element 14 to a starting position, limiting the displacement of the actuable element 14 along the path, and/or inhibiting deviation of the actuable element 14 in one or more directions from the path 6.

One or more of the exemplary control mechanisms may be used with an actuable element 14 that includes a front portion 70 comprised of a permanent magnetic material and also with an actuable element 14 that includes a front portion 70 comprised of a soft magnetic material. Displacement of the actuable element 14 may be controlled by providing a stop, a clamp, a cantilever, a spring, a second MEMS actuator, and any combination of the foregoing items. The control mechanism may be substantially planar, and may be disposed in a plane that is substantially parallel to the substantially planar surface of the substrate 54.

Alternately, the control mechanism may be disposed in a plane inclined at an angle to the substantially planar surface of the substrate 54. Also alternately, the control mechanism may include constituent portions that are not coplanar with each other.

[0100] The control mechanisms may control the location of the actuable element 14 in a direction substantially perpendicular to the substantially planar surface of the substrate 54.

Suitable control mechanisms for limiting displacement from the surface of the substrate 54 include a clamp, a cantilever, a spring, a suspension system, and any combination of the foregoing items. Each of these mechanisms may be substantially planar, and each of these mechanisms may be disposed in a plane that is substantially parallel to the substantially

planar surface of the substrate 54. Alternately, each of these mechanisms may be disposed in a plane inclined at an angle to the substantially planar surface of the substrate 54. Also alternately, each of these mechanisms may include constituent portions that are not coplanar with each other.

[0101] An exemplary embodiment of a control mechanism in the form of a stop is shown in Fig. 5. The MEMS device 10a includes a stop 92 in the form of a wall disposed on the substrate 54 to selectively inhibit displacement of the actuable element 14 beyond a desired position on the path 6. The MEMS device 10a also includes a suitable secondary control mechanism (not shown) for controlling the location of the actuable element 14 in the direction substantially perpendicular to the substantially planar surface of the substrate.

The secondary control mechanism may be a clamp, a cantilever, a spring, or a suspension system, as described below. The stop 92 may intersect the path 6 and the axis of actuation 26 and have a surface 94 that is oriented at an angle 96 to the direction of the path 6. In this embodiment, the stop 92 may inhibit displacement of the actuable element 14 by providing a rigid barrier to further displacement and may also deflect the actuable element 14 onto another path. For this latter purpose, the surface 94 of the stop 92 may be at least partially comprised of or coated with a resilient material for deflecting the actuable element 14.

Additionally, the surface 94 may be at least partially comprised of or coated with a wear- protective material for protecting the surface 94 from being deformed or otherwise altered by contact with the actuable element 14. The stop 92 may have a surface 94 oriented at any angle 96 with respect to the path 6. For example, the stop 92 may have a surface 94 oriented at a substantially perpendicular angle to the path 6.

[0102] An exemplary embodiment of a control mechanism in the form of a clamp is shown in Fig. 6. A variety of clamps may be utilized in the MEMS device lOb. For example, the MEMS device 1 Ob may include, but is not limited to, a mechanical clamp, an electrostatic clamp, and an electromagnetic clamp. An exemplary electromagnetic clamp is shown in Fig. 6. The MEMS device 1 Ob includes an electromagnetic clamp 85 that may be disposed on the substrate 54 for selectively clamping the actuable element 14 at a desired position on the path 6. The clamp 85 may reduce the amount of power consumed by the MEMS actuator 12, thereby possibly increasing the useful life of the MEMS actuator 12. Operation of the clamp 85 may be understood in the following manner. First, the actuable element 14 may be displaced to a desired position by energizing the actuator 12. Second, the clamp 85 may be energized to maintain the displacement of the actuable element 14 at a desired

location, and the actuator 12 may be de-energized. The clamp 85 may include first and second leads 86,88 and first and second electrode pads 87, 89 for connection to an external power source (not shown). The clamp 85 may be coupled to the actuator 12, the actuable element 14, or both. The clamp 85 may be constructed in a manner analogous to the that for the electromagnetic actuator described below.

[0103] An exemplary embodiment of a cantilever control mechanism is shown in Figs. 1A, 1B, and 1C. A cantilever 24 may be suspended within the cavity 16 formed in the substrate 54 and coupled at one end to the arm 22 of the actuable element 14 and at another end to the substrate 54 for selectively applying a second force to the actuable element 14 in a direction opposite to the first force applied by the actuator 12. As provided previously herein, this second force tends to restore the actuable element 14 to its original position on the path 6. The magnitude of the second force selectively applied by the cantilever 24 depends on a variety of factors, including the physical dimensions of the cantilever 24 and the Young's modulus of the material from which the cantilever 24 is constructed.

[0104] In certain embodiments, the cantilever 24 may be constructed directly from the substrate material. For example, as shown in Figs. 1B and 1C and explained in greater detail below, the substrate 54 may include a base layer 54a and a buried oxide layer 54b formed thereon. A device layer 52 of single crystal Si may be formed on the buried oxide layer 54b. The cantilever 24 in one exemplary embodiment may be constructed from the single crystal Si device layer 52 of the substrate 54 and may be suspended above the base layer 54a of the substrate 54. The cantilever 24 may be fabricated by a release etching process or other process as described in detail below. Fabricating the cantilever 24 from the substrate material, i. e. , the single crystal Si device layer of the present example, may generate a durable and robust control mechanism suitable for mass production. The desired degree of elasticity of the cantilever 24 can be selected for the particular application of the MEMS device. For example, in certain applications, a rigid cantilever 24 may be desirable, while in other applications, a softer cantilever may be preferable. Generally, the Young's modulus of the cantilever 24 depends on the crystal orientation of the Si device layer 52.

As such, a cantilever 24 having a desired degree of elasticity can be fabricated by selecting a Si device layer 52 that has an appropriate crystal orientation. Exemplary crystal orientations in a plane substantially perpendicular to the base layer 54a include, in conventional notation, <111>, <110>, and <100>.

[0105] As shown in Figs. 1A, 1B, and 1C, the cantilever 24 has a substantially rectangular shape when viewed from above and a substantially rectangular cross-section that has an extent that is substantially constant across a length 72 of the cantilever. Alternately, the cantilever 24 may have a variety of shapes when viewed from above and a variety of cross- section shapes, provided that the shapes of the cantilever 24 are suitable for selectively applying a control force to the actuable element 14. For example, these shapes may include the shape of any type of oval, including an ellipse and a circle, any type of polygon, including a rectangle, a square, and a triangle, and the shape of any type of semi-oval, including a semi-circle. Also alternately, the cross-section may have an extent that varies across the length of the cantilever 24.

[0106] Other exemplary embodiments of cantilever control mechanisms are shown in Figs.

7A and 7B. The MEMS device 10c includes a first cantilever 24a and a second cantilever 24b that may be formed from the device layer 52 of the substrate 54 and may be suspended over the base layer 54a and coupled to the arm 22 of the actuable element 14. As shown in Fig. 7A, the first and second cantilevers 24a, 24b may both be coupled to a first side 64 of the arm 22. Alternately, as shown in Fig. 7B, the MEMS device 1 Od includes a first cantilever 24a that may be coupled to the first side 64 of the arm 22 and a second cantilever 24b that may be coupled to a second side 66 of the arm 22. Also alternately, as shown in Fig. 7B, the first and second cantilevers 24a, 24b may be coupled to the first and second sides 64,66 of the arm 22 so that the first cantilever 24a is substantially coaxial with the second cantilever 24b.

[0107] Various other embodiments of cantilever control mechanisms are possible. For example, any number of cantilevers may be formed from or disposed on the substrate and coupled to the arm 22. In one embodiment, all of the cantilevers may be coupled to the first side 64 of the arm. In another embodiment, a first number of cantilevers may be coupled to the first side 64, and a second number of cantilevers may be coupled to the second side 66.

In yet another embodiment, at least one cantilever coupled to the first side 64 may be substantially coaxial with at least one cantilever coupled to the second side 66.

[0108] An exemplary embodiment of a spring control mechanism is shown in Fig. 8A. The MEMS device 10e includes a spring 60 that may be disposed on the substrate and attached to one end of a mounting structure 62 for providing a restoring force to the actuable element 14. The mounting structure 62 may have another end attached to the substrate. The spring 60 may be disposed at a variety of locations and orientations on the substrate. For example,

as shown in Fig. 8A, the spring 60 may be mounted along the axis of actuation and attached to the end of the arm of the actuable element 14 located furthest from the base 20.

Alternately, the spring 60 may be mounted along the axis of actuation 26 and attached to the base 20 of the actuable element 14. Operation of the spring 60 can be understood in the following manner. As provided previously herein, the actuator 12 selectively generates a first force that tends to displace the actuable element 14 along a path 6 that extends at least partially into the gap 8. The spring 60 selectively applies a second force to the actuable element 14. Like the first force, the second force also acts along the axis of actuation 26.

Unlike the first force, however, the second force acts in an opposite direction, and tends to restore the actuable element to its original position on the path. A procedure similar to that previously provided herein may be followed to effect displacement of the actuable element 14. The spring 60 may be comprised of any material known to one of ordinary skill in the art to be suitable for a MEMS spring, for example, Si and Ti. The magnitude of the restoring force applied by the spring 60 depends on a variety of factors, including the spring constant of the spring 60, which itself depends on the type of material comprising the spring 60 and the physical dimensions of the spring 60, such as the number of spring coil turns and the length, width, and thickness of the spring.

[0109] Another exemplary embodiment of the spring control mechanism is shown in Fig.

8B. The MEMS device 1 Of includes two springs 60A and 60B attached to mounting structures 62A, 62B, respectively, for providing a restoring force to the actuable element 14. The mounting structures 62A, 62B are also attached to the substrate. A procedure similar to that previously provided herein may be followed to control the displacement of the actual element 14 using the springs 60A and 60B.

[0110] Various other embodiments of spring control mechanisms are possible. For example, any number of springs may be disposed on the substrate and coupled to the arm 22. In one embodiment, all of the springs may be coupled to the first side 64 of the arm 22.

In another embodiment, a first number of springs may be coupled to the first side 64, and a second number of springs may be coupled to the second side 66. Additionally, any number of springs may be coupled to the arm 22 and the base 20 and mounted along the axis of actuation 26.

[0111] As described below, the actuable element 14, including the base 20 and the arm 22, may be formed directly from the substrate material. In the exemplary embodiments illustrated in Figs. 1A, 1B, 7A, and 7B, for example, the actuable element 14 is formed

directly from the device layer 52 and is, thus, suspended over the base layer 54a. Each of the control mechanisms, including the cantilever 24, the spring 60, the clamp 85, and the stop 92, may also be formed directly from the substrate material. Alternately, the actuable element 14, as well as each of the control mechanisms, may be deposited, built upon, or otherwise added to the substrate. Forming both the control mechanism, including cantilever (s) 24, spring (s) 60, clamp (s) 85, and stop (s) 92, and the actuable element 14 from the substrate may result in a more robust connection between the control mechanism and the actuable element 14 and may reduce manufacturing steps.

[0112] The MEMS devices disclosed herein may optionally include a suspension system that suspends the actuable element above the substrate to minimize friction between the actuable element and the surface of the substrate during operation. In one exemplary embodiment, the suspension system may include a mechanical support, such as one or more clamps, cantilevers, or springs, as previously provided herein. In another exemplary embodiment, the suspension system may use two or more permanent magnets to provide a magnetic suspending force on the actuable element. In another exemplary embodiment, the suspension system may use one or more electrically activated magnetic coils that selectively provide a magnetic suspending force on the actuable element.

[0113] An exemplary embodiment of a magnetic suspension system is shown in Fig. 9A.

The magnetic suspension system 73 provides a magnetic force for suspending the actuable element 14 above the base layer 54a. The magnetic suspension system 73 may include a first magnet 74 coupled to the bottom surface 76 of the actuable element 14. The first magnet 74 may extend the entire length of the actuable element 14, or may be discretely positioned on the actuable element 14. The magnetic suspension system 73 may also include a second magnet 77 embedded in or positioned on the base layer 54a. Preferably, the first and second magnets 74,77 are permanent magnets. For example, the first and second magnets 74,77 may be cobalt alloy magnets. Preferably, the poles of the first and second magnets 74,77 are oriented such that a repulsive magnetic force acts on the actuable element 14 to suspend the actuable element 14 above the base layer 54a. The repulsive force is indicated schematically by arrows in Fig. 9A.

[0114] An exemplary embodiment of an electromagnetic suspension system is shown in Fig. 9B. The electromagnetic suspension system 61 provides a suspension force to the actuable element 14. Preferably, the electromagnetic suspension system 61 is positioned on the surface of the base layer 54a. As shown in Fig. 9B, the electromagnetic suspension

system 61 may include a coil 63 of electrically conductive material connected to leads 71.

An electromagnetic suspension system including a coil of electrically conductive material may be constructed in a manner known by one of ordinary skill in the art. The coil of electrically conductive material may be constructed in the manner shown in U. S. Patent No.

5,778, 513 to Miu et al. Alternately, the electromagnetic suspension system 61 may include coils of electrically conductive material wrapped about a core of soft magnetic material in a manner analogous to a solenoid. The coil 63 is connected to an external power source (not shown). In this embodiment, the actuable element 14 may include a core 67 of non- magnetic material to which a layer of permanent magnet material 69 may be attached. The magnetic layer 69 may extend the entire length of the actuable element 14, or may be discretely positioned on the actuable element 14. Upon the application of a current to the electromagnetic suspension system 61, the coils 63 and the core 65 generate a force that tends to repel the layer of permanent magnet material 69, thereby suspending the actuable element 14 above the base layer 54a. The repulsive force is indicated schematically by arrows in Fig. 9B. Removing current from the electromagnetic system 61 causes the layer of permanent magnet material 69 to be attracted to the coils 63, thereby latching the actuable element 14 in place on the base layer 54a. The electromagnetic suspension system 61 thus allows selective suspension and latching of the actuable element 14. The electromagnetic suspension system 61 may be constructed in a manner analogous to that for the electromagnetic actuator described below.

[0115] Another exemplary embodiment of the MEMS device disclosed herein is shown in Fig. 10. The MEMS device 210 includes a first MEMS actuator 212, a second MEMS actuator 312, and an actuable element 214. The first and second MEMS actuators 212,312 may include magnetic cores 230,330, electrically conductive coils 232,332 wrapped around the cores 230,330, leads 218,219, 318, and 319, and electrode pads 244,246, 344, and 346. The actuable element 214 includes a first base 220, a second base 320, and an arm 222 coupled to the first and second bases 220,320. The first actuator 212 includes a gap 238 through which the actuable element 214 can be displaced. The second actuator 312 also includes a gap 338 through which the actuable element 214 can be displaced. The MEMS device 210 and its constituent parts are disposed on a substrate having a substantially planar surface.

[0116] In an exemplary embodiment, the first and second MEMS actuators 212,312 may jointly provide a control mechanism for the actuable element 214. Operation of this

embodiment of the MEMS device 210 may be understood in the following manner. First, the first actuator may be energized in a manner similar to that previously provided herein.

Second, the first actuator 212 may selectively generate a first force that tends to displace the actuable element 214 along a path 206 that extends at least partially into the gaps 238, 338.

Third, after the actuable element 214 reaches a desirable position on the path 206, the second actuator 312 may be energized. Fourth, the first actuator 212 may be de-energized, and the second actuator 312 may be energized to provide a second force. Like the first force, the second force is directed along the path 206. Unlike the first force, however, the second force is directed opposite to the first force, and tends to restore the actuable element 214 to its original position on the path 206.

[0117] Other modes of operating the first and second actuators 212,312 to control the displacement of the actuable element 214 are possible. For example, the first and second actuators 212,312 may be designed to continuously provide first and second forces to the actuable element 212. In this embodiment, displacement of the actuable element 214 may be controlled by controlling the relative strengths of the first and second forces.

Additionally, the first and second actuators 212,312 may be selectively energized to permit oscillation of the actuable element 214 along a path 206 extending through the gaps 238, 338.

[0118] As shown in Fig. 10, the second MEMS actuator 312 may be an electromagnetic actuator. Alternately, the second MEMS actuator 312 may be any other type of MEMS actuator known to one of ordinary skill in the art, including, but not limited to, an electrostatic actuator, a thermal actuator, and a piezoelectric actuator. In these embodiments, the actuable element 214 may be modified to include a portion comprised of an electrically conductive material, a thermally sensitive material, and a piezoelectric material, respectively.

[0119] A control mechanism for limiting the displacement of the actuable element 214 in a direction away from the substrate may also optionally be provided. Suitable control mechanisms may include one or more cantilevers, springs, or clamps.

[0120] As shown in Fig. 10, each of the path 206, the first and second actuators 212,312, and the actuable element 214 lies in a plane that is substantially parallel to the substantially planar surface of the substrate. Alternately, any combination of the path 206, the first and second actuators 212,312, and the actuable element 214 may lie in a plane that is inclined at an angle to the substantially planar surface of the substrate. Also alternately, any

combination of the path 206, first and second actuators 212,312, and the actuable element 214 may include constituent portions that are not coplanar with each other. Additionally, any combination of the path 206, the first and second actuators 212,312, and the actuable element 214 may be formed from or disposed on the substrate.

[0121] Another exemplary embodiment of the MEMS device disclosed herein is shown in Fig. 11. The MEMS device 410 includes a first MEMS actuator 412, a first actuable element 414, a second MEMS actuator 512, and a second actuable element 514. The first and second MEMS actuators 412,512 may include magnetic cores 430,530, electrically conductive coils 432,532 wrapped around the cores 430,530, leads 418,419, 518, and 519, and electrode pads 444,446, 544, and 546. The first actuable element 414 includes a first base 420 and a first arm 422 coupled to the first base 420, and the second actuable element 514 includes a second base 520 and a second arm 522 coupled to the second base. The first actuator 412 includes a gap 438 through which the first actuable element 414 can be displaced, and the second actuator 512 includes a gap 538 through which the second actuable element 514 can be displaced. The MEMS device 410 and its constituent parts may be formed from or disposed on a substrate having a substantially planar surface.

[0122] Operation of the MEMS device 410 is based on principles previously provided herein. In one embodiment, the MEMS device 410 may be operated to provide simultaneous or separate actuation of the first and second actuable elements 414,514. For example, the first MEMS actuator 412 may selectively apply a first force to the first actuable element 414 to displace the first actuable element 414 along a first path 406 that extends at least partially into the gap 438. Also in this embodiment, the second MEMS actuator 512 may selectively apply a second force to the second actuable element 514 to displace the actuable element 514 along a second path 506 that extends at least partially into the gap 538.

[0123] In a variety of exemplary embodiments, the MEMS device 410 may be modified to include one or more of the exemplary control mechanisms previously provided herein, including, but not limited to, a stop, a clamp, a cantilever, a spring, a suspension system, and an additional MEMS actuator. For example, one or more stops may be disposed on the substrate to selectively inhibit displacement of the first and/or second actuable elements 414,514. Also, one or more clamps may be disposed on the substrate for selectively clamping and/or limiting displacement of the first and/or second actuable elements 414,514 in a direction away from the substrate. Additionally, one or more cantilevers may be

suspended over the substrate and coupled to the first and/or second arms 422,522 of the first and second actuable elements 414,514. Further, one or more springs may be disposed on the substrate and coupled to the first and/or second arms 422,522 of the first and second actuable elements 414,514. Further still, one or more additional MEMS actuators may be disposed on the substrate to provide a joint control mechanism for the first and/or second actuable elements 414,514. Further yet still, one or more suspension mechanisms may be disposed on the substrate to provide a suspending forces to the first and/or second actuable elements 414,514.

[0124] In another variety of exemplary embodiments, the MEMS device 410 may be modified to include any combination of the control mechanisms previously provided herein.

For example, one or more cantilevers may be disposed on the substrate and coupled to the first actuable element 414, and one or more springs may be disposed on the substrate and coupled to the second actuable element 514.

[0125] As shown in Fig. 11, each of the first and second paths 406,506, first and second actuators 412,512, and first and second actuable elements 414,514 is substantially planar, and each of the first and second paths 406,506, the first and second actuators 412,512, and the first and second actuable elements 414,514 lies in a plane that is substantially parallel to the substantially planar surface of the substrate. Alternately, any combination of the first and second paths 406,506, the first and second actuators 412,512, and the first and second actuable elements 414,514 may lie in a plane that is inclined at an angle to the substantially planar surface of the substrate. For example, the first actuator 412 and first actuable element 414 may be substantially planar and lie in a first plane, and the second actuator 512 and second actuable element 514 may also be substantially planar and lie in a second plane oriented at an angle with respect to the first plane. Also alternately, any combination of the first and second paths 406,506, the first and second actuators 412,512, and the first and second actuable elements 414,514 may include constituent portions that are not coplanar with each other.

[0126] Another exemplary embodiment of the MEMS device disclosed herein includes an array having any number of MEMS actuators, any number of actuable elements, and any number of control mechanisms disposed on a substrate. In one exemplary embodiment, all of the MEMS actuators, actuable elements, and control mechanisms may be disposed in the same plane and may be oriented at a variety of angles with respect to each other. In another exemplary embodiment, one or more of the MEMS actuators and/or actuable elements

and/or control mechanisms may be disposed in planes that are oriented at a variety of angles with respect to each other and with respect to the substrate.

[0127] Advantageously, the MEMS devices disclosed herein may be used to generate forces sufficient to displace an actuable element along paths having lengths between approximately 1, um and approximately 3,000 jjm. The actual amount of displacement of the actuable element depends on a variety of design factors, including the elasticity of the control mechanism and the physical parameters of the electromagnetic actuator, such as the amount of current applied to the actuator, the number of coil turns, and the permeability and saturation of the material from which the magnetic core is constructed.

[0128] The MEMS devices disclosed herein may be incorporated into a variety of structures, including, but not limited to, electrical switches, fluid pumps, fluid valves, optical attenuators, optical switches, tunable filters, and gratings. Details of fabrication of these devices are provided below.

[0129] An exemplary embodiment of the MEMS device disclosed herein incorporated into an optical attenuator is shown in Fig. 12. The MEMS device 610 includes a MEMS actuator 612 and an actuable element 614. The MEMS actuator 612 may be constructed in a manner similar to that previously provided herein to include a magnetic core 630 and electrically conductive coil 632 wrapped around the core 630. The MEMS actuator 612 also includes a core gap 638 through which the actuable element 614 can be displaced. The actuable element 614 includes a base 620 and an arm 622 coupled to the base 620. The actuable element 614 also includes a shutter 640 coupled to the arm 622. Optionally, the actuable element 614 may include first and second cantilevers 624a, 624b coupled to the arm 622 of the actuable element 614. The MEMS device 610 and its constituent parts are disposed on a substrate having a substantially planar surface.

[0130] Operation of the MEMS device 610 is based on principles previously provided herein. In this embodiment, the actuable element 614 includes an optical element that may be displaced along an optical path 606 intersecting an input optical fiber 602a and an output optical fiber 602b. By selectively applying an operating current to the actuator 612, the actuable element 614 and the shutter 640 can be selectively positioned within the optical path 606 to attenuate the optical beam 601 between the optical fibers 602a and 602b. In this manner, the MEMS device 610 can function as a variable optical attenuator. The MEMS device 610 provides precise electronic control of the optical signal level between

the optical fibers 602a, 602b by selectively moving the shutter 640 into and out of the optical path 606 between the optical fibers 602a, 602b.

[0131] A variety of constructions are possible for the shutter 640. For example, the shutter 640 may be constructed of opaque material or may include an opaque surface for blocking the optical beam 601 when positioned in the optical path 606. Alternately, the shutter 640 may be constructed from a reflective material or may include a reflective surface. Suitable reflective materials include gold or any other reflective material selected based on the wavelength of the optical signal. Also alternately, the shutter 640 may be constructed from a semi-transparent or a semi-reflective material or may include a semi-transparent or a semi-reflective surface. Preferably, the shutter 640 is positioned on the arm 622 of the actuable element 614 so that the surface of the shutter 640 facing the input optical fiber 602a is oriented at an angle to the optical path 606. In this manner, the shutter 640 may operate to reflect at least a portion of the optical beam 601 away from the output optical fiber 602b when the shutter 640 is positioned in the optical path 606. Additionally, in this manner, the shutter 640 may operate to inhibit reflection of the optical beam 601 onto the input optical fiber 602a. Preferably, however, the shutter 640 is constructed from a non- opaque material, to reduce the absorption of the optical beam 601 by the shutter 640 and consequential heating.

[0132] The shutter 640 may be a separate component disposed on the substrate and coupled to the actuable element 614. Alternately, the shutter 640 may be formed from the actuable element 614 by, for example, coating a section of the actuable element 614 with an opaque or reflective coating.

[0133] By selecting the desired optical element, the MEMS device 610 may be used with any optical component known to one of ordinary skill in the art to facilitate control or manipulation of an optical beam 601. For example, instead of a shutter 640, the actuable element 614 may include an optical filter, a tunable optical grating, or a mirror. A variety of other optical components may also be constructed using the MEMS device 610. For example, the actuable element 614 may include a mirror, and the MEMS device 610 may function to selectively switch the optical beam between two optical fibers. Also, the MEMS device 610 may be used to control or modulate the intensity of a beam of particles, for example, a beam of atoms or subatomic particles.

[0134] As shown in Fig. 12, the input and output optical fibers 602a, 602b and the optical beam 601 are substantially planar, and lie in a plane substantially parallel to the

substantially planar surface of the substrate. Alternately, the input and output optical fibers 602a, 602b and the optical beam 601 may lie in a plane that is inclined at an angle to the substantially planar surface of the substrate. For example, the optical fibers 602a, 602b and the optical beam 601 may lie in a plane that is perpendicular to the substantially planar surface of the substrate. Also alternately, any combination of the optical fibers 602a, 602b and the optical beam 601 may include constituent portions that are not coplanar with each other.

[0135] Another exemplary embodiment of the MEMS device disclosed herein incorporated into an optical attenuator is shown in Fig. 13. The MEMS device 710 includes first and second MEMS actuators 712,812 and first and second actuable elements 714,814. The first and second actuable elements 714,814 include first and second shutters 740,840 coupled to the arms 722,822. Optionally, the first and second actuable elements 714, 814 may include cantilevers 724a, 724b, 724c, 724d coupled to the arms 722,822 of the actuable elements 714,814. The MEMS device 710 and its constituent parts are disposed on a substrate having a substantially planar surface.

[0136] Operation of the MEMS device 710 is based on principles previously provided herein. In this embodiment, the actuable elements 714,814 comprise two optical elements having two shutters 740,840. The first and second shutters 740,840 may have a variety of relative shapes and orientations. For example, the first and second shutters 740,840 may be shaped and arranged to permit the first shutter 740 to overlap the second shutter 840 upon joint actuation. In this embodiment, the first and second shutters 740, 840 may have mated or interlocking shapes. Also, the first and second shutters 740,840 may be shaped and arranged to permit the first shutter 740 to contact the second shutter 840 upon actuation. Additionally, the first and second shutters 740,840 may be disposed at different locations on the substrate to provide staggered attenuation of the optical beam 601.

[0137] An exemplary embodiment of the MEMS device disclosed herein incorporated into an electrical switch is shown in Fig. 14. The MEMS device 910 includes a MEMS actuator 912 and an actuable element 914. The MEMS actuator 912 may be constructed in a manner similar to that previously provided herein to include a gap 908 through which the actuable element 914 can be displaced. The actuable element 914 includes a base 920 and an arm 922 coupled to the base 914. The actuable element 914 also includes a contact bar 909 coupled to the arm 922. The MEMS device 910 also includes a first pair of electrical contacts 939a, 939b and a second pair of electrical contacts 941a, 941b. The electrical

contacts 939a, 939b, 941a, 941b may be connected to other electrical circuit elements (not shown). The MEMS device 910 and its constituent parts are disposed on a substrate having a substantially planar surface. The actuable element 914 may be suspended over a cavity 916 in the substrate.

[0138] Operation of the MEMS device 910 is based on principles previously provided herein. In this embodiment, the actuable element 914 comprises an electrical element that may be displaced along a path 906 contacting the first pair of electrical contacts 939a, 939b and the second pair of electrical contacts 941a, 941b. By selectively applying an operating current to the actuator 912, the actuable element 914 and the contact bar 909 can be selectively displaced along the path 906 to complete an electrical connection between either the first pair of electrical contacts 939a, 939b or the second pair of electrical contacts 941a, 941b. The electrical contacts 939a, 939b, 941a, and 941b and the contact bar 909 form two switches that may be selectively closed. In this manner, the MEMS device 910 can function as an electrical switch.

[0139] As shown in Fig. 14, each of the MEMS actuator 912, the actuable element 914, the contact bar 909, and the electrical contacts 939a, 939b, 941a, and 941b lies in a plane that is substantially parallel to the substantially planar surface of the substrate. Alternately, any combination of the MEMS actuator 912, the actuable element 914, the contact bar 909, and the electrical contacts 939a, 939b, 941a, and 941b may lie in a plane that is inclined at an angle to the substantially planar surface of the substrate. Also alternately, any combination of the MEMS actuator 912, the actuable element 914, the contact bar 909, and the electrical contacts 939a, 939b, 941 a, and 941b may include constituent portions that are not coplanar with each other.

[0140] An exemplary embodiment of the MEMS device disclosed herein incorporated into an optical switch is shown in Figs. 15A and 15B. As shown in Fig. 15A, the MEMS device 1010 includes a MEMS actuator 1012 and an actuable element 1014. The MEMS actuator 1012 may include a gap 1008 through which the actuable element 1014 can be displaced.

The actuable element 1014 includes a base 1020 and an arm 1022 coupled to the base 1020.

The actuable element may also include a cantilever 1024. The MEMS device 1010 and its constituent parts are disposed on a substrate having a substantially planar surface. The actuable element 1014 may be disposed over a cavity 1016 in the substrate.

[0141] As shown in Figs. 15A and 15B, the actuable element 1014 also includes one or more slots 1003 shaped and arranged to receive a moveable optical fiber 1002a. The

moveable optical fiber 1002a may be attached to the slot 1003 using any securing means known to one of ordinary skill in the art, including, but not limited to, an adhesive.

[0142] Operation of the MEMS device 1010 is based on principles previously provided herein. In this embodiment, the actuable element 1014 comprises an optical element that may be displaced along a path 1006 intersecting one or more stationary output optical fibers 1002b, 1002bb. By selectively applying an operating current to the actuator 1012, the actuable element 1014 and the optical fiber 1002a can be selectively displaced along the optical path 1006 to align the optical fiber 1002a with one of the stationary fibers 1002b, 1002bb. Alignment of the optical fiber 1002a with one of the optical fibers 1002b, 1002bb results in the transmission of an optical beam between the optical fiber 1002a and one of the optical fibers 1002b, 1002bb. In this manner, the MEMS device 1010 can function as an optical switch.

[0143] As shown in Fig. 15A, the MEMS device 1010 also includes walls 1092a and 1092b. The walls 1092a, 1092b are disposed in the path of the actuable element 1014 to facilitate precise alignment of the optical fiber 1002 and one of the optical fibers 1002b, 1002bb. The wall 1092a is disposed such that the optical fiber 1002a will be precisely aligned with the optical fiber 1002b when the actuable element 1014 contacts the wall 1092a. Similarly, the wall 1092b is disposed such that the optical fiber 1002a will be precisely aligned with the optical fiber 1002bb when the actuable element contacts the wall 1092b.

[0144] The walls 1092a, 1092b may also be disposed at a variety of other locations on the substrate. For example, the walls 1092a, 1092b may be disposed on opposite sides of the cantilever 1024.

[0145] As shown in Fig. 15A, the optical fibers 1002a, 1002b lie in a plane substantially parallel to the substantially planar surface of the substrate. Alternately, the optical fibers 1002a, 1002b may lie in a plane that is inclined at an angle to the substantially planar surface of the substrate. For example, the optical fibers 1002a, 1002b may lie in a plane that is perpendicular to the substantially planar surface of the substrate. Also alternately, any combination of the optical fibers 1002a, 1002b may include constituent portions that are not coplanar with each other.

[0146] A variety of other components may also be constructed using the MEMS device 1010. For example, the actuable element 1014 may include one or more slots 1003 shaped and arranged to transport any object along the path 1006.

[0147] An exemplary embodiment of the MEMS device disclosed herein incorporated into a fluid valve is shown in Fig. 16A. The MEMS device 1110 includes a MEMS actuator 1112 and an actuable element 1114. The MEMS actuator 1112 may include a gap 1108 through which the actuable element 1114 can be displaced. The actuable element 1114 includes a base 1120 and an arm 1122 coupled to the base 1120. The actuable element 1114 may be disposed over a cavity 1116 in the substrate. The MEMS device 1110 includes a valve 1193 coupled to a source of fluid (not shown). The valve 1193 includes a valve seal 1195, inlets 1197, and a valve element 1191 coupled to the arm 1122 of the actuable element 1114. The MEMS device 1110 and its constituent parts are disposed on a substrate having a substantially planar surface. The MEMS device 1110 also includes a suitable control mechanism (not shown) for controlling the location of the actuable element 1214 in a direction substantially perpendicular to the substantially planar surface of the substrate. This control mechanism may be a clamp, a cantilever, a spring, or a suspension system, as previously provided herein.

[0148] Operation of the MEMS device 1110 is based on principles previously provided herein. In this embodiment, the actuable element 1114 comprises a plunger that may be displaced along a path 1106 opening and closing the valve seal 1195. In Fig. 16A, the valve seal 1195 is open. In this state, fluid from a source of fluid (not shown) may flow into the valve 1193 through the inlets 1197, around the valve element 1191, and out of the valve through the valve seal 1195. By selectively applying an operating current to the actuator 1112, the actuable element 1114 and the valve element 1191 can be selectively displaced along the path 1106 to close the valve seal 1195, thereby preventing fluid from flowing through the valve 1193. In this manner, the MEMS device 1110 can function as a fluid valve.

[0149] Various other configurations of the MEMS device 1110 are possible. For example, the valve 1193 may be constructed in any manner known to one of ordinary skill the art for constructing a valve that provides controlled fluid flow.

[0150] As shown in Fig. 16A, each of the MEMS actuator 1112, the actuable element 1114, and the valve 1193 lies in a plane that is substantially parallel to the substantially planar surface of the substrate. Alternately, any combination of the MEMS actuator 1112, the actuable element 1114, and the valve 1193 may lie in a plane that is inclined at an angle to the substantially planar surface of the substrate. Also alternately, any combination of the

MEMS actuator 1112, the actuable element 1114, and the valve 1193 may include constituent portions that are not coplanar with each other.

[0151] An exemplary embodiment of the MEMS device disclosed herein incorporated into a fluid pump is shown in Fig. 16B. The MEMS device 1210 includes a MEMS actuator 1212 and an actuable element 1214. The MEMS actuator 1212 may include a gap 1208 through which the actuable element 1214 can be displaced. The actuable element 1214 includes a base 1220 and an arm 1222 coupled to the base 1220. The actuable element 1214 may be disposed over a cavity 1216 in the substrate. The MEMS device 1210 includes a pump 1243 coupled to a source of fluid (not shown). The pump 1243 includes an outlet 1241, an inlet 1245, and a pump element 1242 coupled to the arm 1222 of the actuable element 1214. The MEMS device 1210 and its constituent parts are disposed on a substrate having a substantially planar surface. The MEMS device 1210 also includes a suitable control mechanism (not shown) for controlling the location of the actuable element 1214 in a direction substantially perpendicular to the substantially planar surface of the substrate. This control mechanism may be a clamp, a cantilever, a spring, or a suspension system, as previously provided herein.

[0152] Operation of the MEMS device 1210 is based on principles previously provided herein. In this embodiment, the actuable element 1214 comprises a piston that may be displaced along a path 1206 pumping fluid from the inlet 1245 to the outlet 1241. In Fig.

16B, the pump 1243 pushes fluid disposed between the pump element 1242 and the outlet 1241 toward the outlet 1241. By selectively applying an operating current to the actuator 1212, the actuable element 1214 and the pump element 1242 can be selectively displaced along the path 1206 to permit the pump 1243 to intake fluid through the inlet 1245 and push that fluid to the outlet 1241. In this manner, the MEMS device can function as a fluid pump.

[0153] Various other configurations of the MEMS device 1210 are possible. For example, the pump 1243 may be constructed in any manner known to one of ordinary skill in the art for constructing a pump that provides controlled fluid flow.

[0154] As shown in Fig. 16B, each of the MEMS actuator 1212, the actuable element 1214, and the pump 1243 lies in a plane that is substantially parallel to the substantially planar surface of the substrate. Alternately, any combination of the MEMS actuator 1212, the actuable element 1214, and the pump 1243 may lie in a plane that is inclined at an angle to the substantially planar surface of the substrate. Also alternately, any combination of the

MEMS actuator 1212, the actuable element 1214, and the pump 1243 may include constituent portions that are not coplanar with each other.

[0155] The MEMS devices disclosed herein can be constructed from conventional micromachining processes known to one of ordinary skill in the art including, but not limited to, electroplating, etching, photolithography, sputtering, and vacuum deposition.

Advantageously, the MEMS devices disclosed can be fabricated in batches. An exemplary method of constructing the MEMS device 10 shown in Figs. 1A, 1B, and 1C, including the MEMS electromagnetic actuator 12 shown in Figs. 2A, 2B, 2C, and 2D is provided in Figs.

17A-E. This exemplary method may also be modified in a manner familiar to one of ordinary skill in the art to fabricate a batch of MEMS devices 10 from a single wafer.

[0156] The component parts of MEMS device 10 are micromachined in successive layers upon the substrate 54. The substrate 54 includes a base layer 54a and a buried oxide layer 54b formed thereon. Preferably, the base layer 54a is an Si wafer. Initially, a buried oxide layer 54b is applied to the surface of the base layer 54a, step 1500. Preferably, the buried oxide layer 54b is a layer of thermal oxide, for example, Si02. Next, a device layer 52 is applied to the surface of the buried oxide layer 54b, step 1510. Preferably, the device layer 52 is single crystal Si having a crystal orientation perpendicular to the plane of the insulating substrate 54 sufficient to fabricate a cantilever 24 having a desired elasticity, as previously indicated herein. Alternately, a single MEMS device 10 or a batch of MEMS devices 10 may be fabricated from a commercially available, pre-fabricated, composite silicon-on-insulator (SOI) wafer or a batch of SOI wafers, respectively.

[0157] The components of the MEMS device 10 may be fabricated at least partially from the material of the device layer 52. For example, as explained in greater detail below, the actuable element 14 includes an arm 22 that may be formed at least partially or entirely from the material of the device layer 52 and a base 20 that includes at least a portion 70 comprising magnetic material. Additionally, other components of the MEMS device 10 may be fabricated at least partially or entirely from the material of the device layer 52. For example, as explained in greater detail below, the cantilever 24 may be formed entirely from the device layer 52. Since the components of the MEMS device 10 may be fabricated at least partially from the material of the device layer 54, the MEMS device 10 has a compact size in a direction substantially perpendicular to the substantially planar surface of the substrate 54. The MEMS actuator 12 has an extent in a direction substantially perpendicular to the substantially planar surface of the device layer 52 of less than

approximately 100 J. m, and preferably less than approximately 50 um. Additionally, the base 20 of the actuable element 12, specifically, the magnetic portion 70 of the base 20 and any optional spacer layer 20a (described below), has an extent in a direction substantially perpendicular to the substantially planar surface of the device layer 52 of less than approximately 10, um, and preferably less than approximately 4, um.

[0158] After step 1510, a layer of magnetic material is applied to the device layer 52 to form the base 20 of the actuable element 14, step 1520. The layer of magnetic material may include multiple alternating layers of chrome and a permanent magnetic material, for example, CoPt. In this embodiment, alternating layers of chrome may be applied to induce the desired epitaxial orientation. The multiple alternating layers may be applied by sputtering or vacuum deposition. Alternately, the layer of magnetic material may include alternating layers of chrome and a soft magnetic material, for example, NiFe. In this embodiment, alternating layers of chrome may be applied to facilitate adhesion of the soft magnetic material. Optionally, before applying the layer of magnetic material, a spacer layer 20a may be applied to the device layer 52 to adjust the location of the base 20 in the z- direction (i. e. in a direction perpendicular to the surface of the substrate) and separate the base 20 from the device layer 52, step 1515. The spacer layer 20a may include a layer of any material that will not be affected by subsequent fabrication processes. For example, the spacer layer 20a may include a layer of a polymeric material, a layer of a metallic material, and/or a layer of non-magnetic material. Preferably, the spacer layer 20a is constructed from a non-oxide material that is different from the materials out of which the masks described below are constructed.

[0159] Fabricating the magnetic portion 70 of base 20 from a permanent magnetic material continues by depositing a layer of a first mask material on the layer of magnetic material to define the base 20 of the actuable element 14, step 1530. As provided previously herein, the magnetic portion 70 need not have the same footprint as the base 20. The first mask material may include a layer of material resistant to high-powered non-reactive ion beam etching. For example, the first mask material may include a layer of aluminum oxide or silica. A layer of a second mask material is then deposited and patterned to form windows through the second mask layer to the first mask layer, step 1535. Depositing and patterning the second mask may include applying a layer of photoresist on the first mask material, soft baking the layer of photoresist, patterning a mask defining the base 20, exposing the layer of photoresist through the mask, and developing the layer of photoresist to remove areas

exposed by ultraviolet light and open windows in the photoresist layer to the first mask layer. Next, the first mask layer is patterned through the windows of the second mask layer to form windows through the first mask layer to the layer of magnetic material, step 1538.

Patterning the first mask layer may include reactive ion beam etching the first mask layer.

Then, the layer of photoresist may be stripped, and the magnetic layer may be patterned through the windows in the first mask layer to define the base 20 of the actuable element 14, step 1540. In step 1540, all of the magnetic layer may be etched away except that portion of the magnetic layer defining the magnetic portion 70 of the base 20 of the actuable element 14. Etching away the magnetic layer may include high-powered non- reactive ion beam etching the magnetic layer in accordance with the first mask. The first mask may then be removed by wet etching, step 1542. Wet etching may include chemically etching away the first mask using a concentrated acid.

[0160] In fabricating the magnetic portion 70 of the base 20, depositing a first mask of material resistant to high-powered non-reactive ion beam etching may not be necessary, depending on whether high-powered non-reactive ion-beam etching is used to pattern the magnetic material. For example, fabricating the base 20 from a soft magnetic material may continue by following a method similar to that above, except that the first mask of material resistant to high-powered non-reactive ion beam etching may not need to be applied.

[0161] After step 1540, an insulation layer is applied to the device layer 52 and cured, step 1550. The insulation layer defines the location of the cavity 16 in the substrate 54, within which the actuable element 14 and the cantilever 24 may be suspended. The insulation layer may include a layer of aluminum oxide or silica applied by sputtering or vacuum deposition. A mask is then patterned from the insulation layer to define the cavity 16, the location of the arm 22, and the location of the cantilever 24, step 1560. The insulation layer is then etched away in accordance with the mask to define the cavity 16, the location of the arm 22, and the location of the cantilever 24, step 1570.

[0162] Next, a protective layer may be applied to protect the insulation layer, step 1580.

The protective layer may include any material than can be selectively etched without affecting the structures fabricated from or built upon the device layer 52. For example, the protective layer may be a layer of metal, such as TiW, applied by sputtering or vacuum deposition. The protective layer may also be a layer of silicon nitride or a layer of a polymeric material. Preferably, the protective layer is not an oxide layer.

[0163] A mask is then patterned from the protective layer to protect the cavity 16, the location of the arm 22, and the location of the cantilever 24, step 1590. The protective layer is then etched away in accordance with the mask to protect the cavity 16, including the location of the arm 22, and the location of the cantilever 24, step 1600.

[0164] As suggested in Figs. 2A, 2B, and 2C, the MEMS actuator 12 may be fabricated from the following process. A lower layer of electrically conductive material may be applied to the insulating substrate 54. The lower layer may then be etched to form a set of lower winding elements 32b in coil 32. Next, connectors 32c, 32d may be formed over the set of lower winding elements 32b. (Optionally, a magnetic core 30 may then be formed over the set of lower winding elements 32b and between the connectors 32c, 32d. ) An upper layer of electrically conductive material may then be applied over the connectors 32c, 32d (and, optionally, the magnetic core 30). The upper layer of electrically conductive material may then be etched to form a set of the upper winding elements 32a in coil 32.

The upper layer of electrically conductive material may be etched so that the upper winding elements 32a are connected to the lower winding elements 32b via connectors 32c, 32d to form the coil 32. Fabrication of the MEMS actuator 12 in this way permits compact packing of the coil 32 around the core 30. For example, in the preferred embodiment, the coil 32 continuously covers approximately 80% of the surface area of the core 30.

[0165] More specifically, after step 1600, a metal seed layer is applied to the device layer 52 to form a lower layer of electrically conductive material in the coil 32, step 1610. The metal seed layer may include multiple alternating layers of chrome and an electrically conductive material. Preferably, the electrically conductive material is copper or a copper alloy. Alternately, the electrically conductive material may include, but is not limited to, aluminum, gold, silver, and alloys thereof. A mask defining the winding elements of the lower set of winding elements 32b is then deposited and patterned to open windows to the metal seed layer, step 1620. Depositing and patterning the mask may include applying a layer of photoresist on the layer of electrically conductive material, soft baking the layer of photoresist, patterning a mask defining the winding elements of the lower set of winding elements 32b, exposing the layer of photoresist through the mask, and developing the layer of photoresist to remove areas exposed by ultraviolet light and open windows in the photoresist layer to the metal seed layer. An electrically conductive material, preferably copper or a copper alloy, and alternately gold, silver, and alloys thereof, is then electroplated onto the metal seed layer in the windows defined by the photoresist layer to

form the winding elements of the lower set of winding elements, step 1630. The layer of photoresist is then stripped, and the metal seed layer is removed, step 1640. Removing the metal seed layer may include etching away the metal seed layer in accordance with the mask to facilitate electrical isolation of the winding elements of the lower set of winding elements 32b. Optionally, a polymeric material may be applied between the lower winding elements 32b and cured to electrically isolate the lower winding elements 32b from each other, step 1660. The polymeric material may be any non-conductive polymer or other non-magnetic material. The polymeric material may also be a cross-linked polymer, for example, baked photoresist, polyamide, and BCB (known by the trade name"Cyclotene").

[0166] Preferably, fabrication of the leads 18,19 and the electrode pads 44,46 follows fabrication of the lower winding elements 32b. After step 1660, a metal seed layer is applied to the device layer 52, step 1670. The metal seed layer may include multiple alternating layers of chrome and an electrically conductive material. Preferably, the electrically conductive material is gold or a gold alloy. Alternately, the electrically conductive material may include, but is not limited to, copper, silver, and alloys thereof. A mask is then patterned to define the leads 18,19 and the electrode pads 44,46, step 1680.

An electrically conductive material, preferably gold or a gold alloy, and alternately copper, silver, and alloys thereof, is then electroplated onto the metal seed layer, step 1690. The metal seed layer is then etched away in accordance with the mask to define the electrode pads 44,46 and the leads 18,19, step 1700. Optionally, a polymeric material may be applied and cured over the leads 18,19 and the edges of the electrode pads 44,46 to protect them from contamination during the remainder of the fabrication process, step 1710. The polymeric material may be the polymeric material identified in step 1660.

[0167] After step 1710, an insulation layer is applied and cured over the lower winding elements 32b to insulate the lower winding elements 32b from the magnetic core 30, step 1720. The insulation layer may include a layer of the polymeric material mentioned in step 1660.

[0168] The connectors 32c, 32d may be constructed by following a technique analogous to that outlined in steps 1610-1640 and 1720 for forming and insulating the set of lower windings, step 1730.

[0169] The magnetic core 30 may be constructed by following a technique analogous to that outlined in steps 1610-1640 for forming the set of lower windings. After step 1730, a metal seed layer is applied to form the magnetic core, step 1740. The metal seed layer may

include alternating layers of chrome and a magnetic material. In this embodiment, alternating layers of chrome may be applied to facilitate adhesion of the magnetic material.

Preferably, the magnetic material is NiFe (permalloy). The magnetic material may, however, be selected from any soft magnetic material, including, but not limited to, CoFe (permendur), CoZr, FeN, AlSiFe (sendust) and alloys thereof, such as NiFeMo (supermalloy), NiFeCuCr (mumetal), NiFeCo, CoFeB, CoFeV (supermendur), CoFeCr (hiperco), CoZrTa, FeAIN, and FeTaN. A mask is then patterned to define the magnetic core, step 1750. A magnetic material, preferably NiFe, is then electroplated onto the metal seed layer in accordance with the mask, step 1760. The metal seed layer is then etched away in accordance with the mask to define the magnetic core 30, step 1770. Optionally, an insulation layer may be applied over the magnetic core 30 and cured to insulate the magnetic core 30 from the upper layer of electrically conductive material in the coil 32, step 1780. The insulation layer may include a layer of the polymeric material mentioned in step 1660.

[0170] The upper layer of electrically conductive material in the coil 32 and the upper winding elements 32a may be constructed by following a technique analogous to that outlined in steps 1610-1660 for forming the lower layer of electrically conductive material in the coil 32 and the lower winding elements 32b, step 1790. Additionally, an insulation layer may be applied and cured over the upper winding elements 32a, step 1795. The insulation layer may include a layer of the polymeric material mentioned in step 1660.

[0171] After step 1795, a gross mask is patterned, step 1800. Next, the gross mask is repatterned using the mask of step 1590, step 1810. The protective layer is then etched away in accordance with the repatterned gross mask to reveal the arm 22, the cantilever 24, and the location of the cavity 16, step 1820. Subsequently, the device layer 52 is removed by utilizing deep Si etching, step 1830. A chemical dry etch based on the conventional Bosch process may also be used to remove the device layer 52. Thereafter, the cavity 16 is formed by removing the buried oxide layer 54b, step 1840. The buried oxide layer 54b may be removed by first removing the repatterned gross mask and then utilizing a release etch or other conventional process, for example, a wet chemical etch or a dry etch. Cleaning and drying the MEMS device 10 completes processing, step 1850. Preferably, cleaning and drying is performed with supercritical carbon dioxide. Alternately, cleaning and drying may be performed with any other solvent known to one of ordinary skill in the art to be suitable for cleaning and drying a MEMS device.

[0172] In a batch fabrication process, the MEMS devices 10 may be singulated, e. g. die cut, after removing the device layer 52. Alternatively, the MEMS devices 10 may be singulated after removing the buried oxide layer, or after cleaning and drying.

[0173] An exemplary embodiment of a MEMS device disposed on a substrate is shown in Figs. 18A, 18B, and 18C. As previously described, the MEMS device 2010 can include a MEMS actuator 2012 and an actuable element 2014. In the exemplary embodiment, the MEMS actuator 2012 can be an electromagnetic actuator, as previously described in greater detail. Alternatively, the MEMS actuator may be another type of MEMS actuator such as, for example, an electrostatic, thermal, or piezoelectric actuator. The illustrated MEMS actuator 2012 can include an electrically conductive coil arranged to generate a magnetic field within a gap 2008 formed by the spaced-apart ends of the actuator 2012 upon application of a current to the coil, as by first electrically conducting lead 2018 and a second electrically conducting lead 2019.

[0174] As previously described, the actuable element 2014 may be shaped to include a base 2020 and an elongated arm 2022 that can be coupled to and extend from the base 2020. As discussed below, the base 2020 or other portion of the actuable element 2014 may include a layer or portion of magnetic material. The base 2020 of the actuable element 2014 may be positioned proximate the gap 2008 such that the base 2020 can be displaced relative to the gap 2008 upon application of a magnetic field on the magnetic material of the actuable element 2014. Optionally, a cantilever 2024 or more than one cantilever 2024, or other mechanism for controlling the displacement of the actuable element may be coupled to the actuable element 2014 at one end and to the substrate 2054 at another end. Exemplary mechanisms for controlling the displacement of the actuable element have been previously described.

[0175] In the exemplary embodiment illustrated in Figs. 18A, 18B, and 18C, the MEMS actuator 2012 can be constructed on the surface of the substrate 2054 and the actuable element 2014 and the cantilever 2024 can be constructed from the substrate 2054, preferably from a top device layer 2052 of the substrate 2054 in the case of a Silicon-on- Insulator (SOI) wafer. In this manner, the actuable element 2014 and the cantilever 2024 may be suspended over a cavity 2016 (denoted by crosshatching in Fig. 18A) in the substrate 2054 and thus may be free to be displaced relative to the substrate 2054 along a displacement axis 2026. Using the deep etching and wet etching manufacturing processes previously described, all or at least a portion of the components of the MEMS device, e. g.,

the actuator 2012, the actuable element 2014 and/or the cantilever 2024 (or other control mechanism), may be constructed from one or more layers of the substrate 2054 to reduce the extent of the MEMS device 2010 in a direction perpendicular from the substrate surface.

[0176] During the wet etching process, and particularly during removal of the MEMS device from the etching bath and during drying of the MEMS device, forces may act on the actuable element 2014 or other components of the MEMS device that can result in substantial movement of the actuable element 2014. For example, vibration and/or agitation of the MEMS device can cause errant motion of one or more components of the MEMS device. For example, the actuable element 2014 may derail from the displacement axis 2026, or the base 2020 may be dislodged from its position adjacent gap 2008. To resist errant motion of the actuable element 2014, stops 2028 may be constructed from one or more layers of substrate 2054 so as to limit the movement of actuable element 2014, particularly during and following wet etching.

[0177] In the exemplary embodiment of Figs. 18A, 18B and 18C, actuable element 2014 can be seen to extend along displacement axis 2026 within a trench, or channel, 2030 formed by sidewalls 2032,2034, which can extend the depth of SOI device layer 2052.

Sidewalls 2032,2034 may be spaced a distance w from actuable element 2014, such that actuable element 2014 may be generally centered in channel 2030. Sidewall 2032 can have an opening 2036 therein to permit the cantilever 2024 to connect to the actuable element 2014. Sidewall 2034 can have an indented portion 2038 extending a distance y along the displacement axis 2026, where sidewall 2034 can be displaced a distance 1 further from displacement axis 2026 than other portions of sidewall 2034.

[0178] One or more stops 2028 can be formed by the opposing faces of the indented portion 2038, the opposing faces being oriented generally perpendicular to the displacement axis 2026. Actuable element 2014 may have a stop arm 2040 that can extend from actuable element 2014 toward sidewall 2034 so as to be between stops 2028. Thus, movement of actuable element 2014 along displacement axis 2026 may be restricted to the movement of stop arm 2040 between stops 2028. The distance y can be chosen such that the movement of actuable element 2014 between stops 2028 is sufficient for normal operation of MEMS device 2010, yet can constrain movement of actuable element, particularly during the wet etching process, to restrict derailing or dislodging of actuable element 2014 from channel 2030.

[0179] It is known in the art that strong interfacial adhesion, commonly referred to as stiction, may be present between contacting microstructure surfaces, such as contacting surfaces of stop arm 2040 and stops 2028. In certain exemplary embodiments, one, or both stops 2028 and/or stop arm 2040 can have a dimple 2042 formed thereon, which can minimize stiction between actuable element 2014 and stop 2028 when actuable element 2014 can contact stop 2028. A dimple 2042 may have a curved or angled contact surface that is configured to limit the amount of surface area contact between the contact surface of the dimple 2042 and a surface of a component of a MEMS device. Alternatively, or in addition, contact surfaces of stop arm 2040 and/or stops 2028 may be textured and/or an anti-stiction coating may be applied thereto to minimize stiction.

[0180] It may also be known in the art that deep etching, e. g. , deep reactive ion etching (DRIE), of features such as channel 2030 can be better controlled when the widths of such features, e. g. , the width w, may be nearly constant. The more uniform that the width can be maintained, the more uniform the etching rates may be across the feature, providing for tighter tolerances. For the exemplary embodiment of Figs. 18A, 18B and 18C, stop arm 2040 and indented portion 2038 can be correspondingly shaped to maintain the width w generally constant. Other shapes that may serve to maintain a nearly constant width or distance between sidewalls 2032,2034 and actuable element 2014 may also be used.

[0181] In the embodiment of Fig. 18A, sidewall 2032 may be shown a distance w from actuable element 2014. However, it can be understood that the operation of stops 2028 can require only that one sidewall be indented to form the stops 2028. Thus, in one embodiment, MEMS device 2010 may include only sidewall 2034 and sidewall 2032 may not be present, or may be located other than the distance w from actuable element 2014.

Also, in the embodiment of Fig. 18A, stop arm 2040 may be located on actuable element 2014 opposite to where cantilever 2024 couples to actuable element 2014, though other locations of stop arm 2040 and stops 2028 along displacement axis 2026 may provide satisfactory control of movements during fabrication of MEMS device 2010.

[0182] Fig. 19 can show a top view of another exemplary embodiment of a MEMS device 2100. In the embodiment of Fig. 19, actuable element 2114 can include an offset portion 2144 in which the displacement axis 2126 and elongated arm 2122 can be offset, as indicated by offset displacement axis 2126a and offset elongated arm portion 2122a. Offset portion 2144 can have a width x that can encompass elongated arm portion 2122 and its offset 2122a. Offset portion 2144 can form offset faces 2144a and 2144b at opposite ends

of offset portion 2144. Sidewalls 2132,2134 may be similarly offset such that channel 2130 can be offset, as indicated by offset channel 2130a, to form stops 2128a and 2128b that may be opposed to respective offset faces 2144a, 2144b. Thus, movement of actuable element 2114 can be limited by faces 2144a and 2144b contacting respective stops 2128a and 2128b.

[0183] In a manner similar to that described for the embodiment of Figs. 18A, 18B and 18C, measures can be taken to minimize stiction between offset faces 2144a, 2114b and respective stops 2128a, 2128b. One or both of stops 2128a, 2128b and/or offset faces 2144a, 2144b can have dimples 2142 formed thereon to minimize stiction between stops 2128a, 2128b and offset faces 2144a, 2144b when the offset faces 2144a, 2144b can contact stops 2128a, 2128b. Alternatively, or in addition, offset faces 2144a, 2144b and/or stops 2128a, 2128b may be textured and/or an anti-stiction coating may be applied thereto. Other methods, as previously described, may also be used to minimize stiction. Also in a manner similar to that described for the embodiment of Figs. 18A, 18B and 18C, offset portion 2144 and sidewalls 2132,2134 can be correspondingly shaped to maintain a width w between actuable element 2114 and sidewalls 2132,2134 generally constant. Other shapes that may serve to maintain a nearly constant width or distance between sidewalls 2132, 2134 and actuable element 2114 may also be used. Further, cantilever 2124 can be coupled to offset portion 2144.

[0184] Fig. 20 shows an enlarged top partial view of another exemplary embodiment of a MEMS device 2210. The base portion 2220 of actuable element 2214 can have a width wu greater than the width w2 of elongated arm 2222. Channel 2230, which may pass through gap 2208, can have a first extent dl, which can be greater than width wl. Stops 2228 can be located at gap 2208 on both sides of displacement axis 2226 and can reduce the extent of channel 2230 at stops 2228 to an extent of d2, where d2 may be less than wl but greater than w2. Thus, movement of actuable element 2214 along displacement axis 2226 in a first direction corresponding to arrow 2246 can be limited by contact between base portion 2220 and stops 2228. Movement in the opposite direction may be limited by contact between base portion 2220 and end 2230a of channel 2230, or between base portion 2220 and actuator 2212 when channel 2230 may extend to actuator 2212. Preferably, for ease of fabrication and to increase the opportunity for base portion 2220 to contact both stops 2228, the stops 2228 may be symmetric about displacement axis 2226. Other modes of fabrication, operation and/or design preference may have the stops 2228 extend unequally

into channel 2230, may have only a single stop 2228 to one side of displacement axis 2226, or may have stops 2228 that may not be located opposite one another along axis 2226.

[0185] Fig. 21 illustrates an enlarged top partial view of another exemplary embodiment of a MEMS device 2310. Stops 2328 may be located along channel 2330, which may pass through gap 2308. Stops 2328 extend into channel 2330 from opposing sidewalls 2332, 2334 to reduce an extent of the channel 2330 to an extent of d3 between stops 2328, where d3 can be greater than a width w2 of actuable element 2314. Tabs 2348 can extend from actuable element 2314 towards sidewalls 2332,2334 to increase the width of actuable element 2314 from w2 to a width of w3 at tabs 2348, where w3 may be greater than d3.

Thus, movement of actuable element 2314 along displacement axis 2326 in a first direction corresponding to arrow 2346 can be limited by contact between tabs 2348 and stops 2328.

Movement in the opposite direction may be limited by contact between base portion 2320 and end 2330a of channel 2330, or between base portion 2320 and actuator 2312 when channel 2330 may extend to actuator 2312. Preferably, for ease of fabrication and to increase the opportunity for tabs 2348 to contact both stops 2328, the tabs 2348 and/or the stops 2328 may be symmetric about displacement axis 2326. Other modes of fabrication, operation and/or design preference may have the tabs 2348 and/or stops 2328 extend unequally and/or unsymmetrically into channel 2330, provided that contact may still be made between tabs 2348 and stops 2328.

[0186] For the embodiments shown in Figs. 20 and 21, measures can be taken to minimize stiction between contacting surfaces of stops 2228 and base portion 2220, of stops 2328 and tabs 2348 and of base portion 2220,2320 and actuator 2212,2312. Such measures can include providing dimples 2242,2342 on, one or both of the contacting surfaces, texturing one or both of the contacting surfaces, applying an anti-stiction coating to one or both of the contacting surfaces and/or other methods, as previously described.

[0187] An exemplary embodiment of a MEMS device disposed on a substrate is shown in Figs. 22A, 22B, and 22C. The MEMS device 3010 can include a MEMS actuator 3012 and an actuable element 3014. In the exemplary embodiment, the MEMS actuator 3012 can be an electromagnetic actuator, as previously described in greater detail. Alternatively, the MEMS actuator may be another type of MEMS actuator such as, for example, an electrostatic, thermal, magnetostrictive, or piezoelectric actuator or any other type of MEMS actuator. The illustrated MEMS actuator 3012 can include an electrically conductive coil arranged to generate a magnetic field with a gap 3008 formed by the

spaced-apart ends of the actuator 3012 upon application of a current to the coil, as by first electrically conducting lead 3018 and a second electrically conducting lead 3019.

[0188] As previously described, the actuable element 3014 may be shaped to include a base 3020 and an elongated arm 3022 that can be coupled to and extend from the base 3020. As discussed below, the base 3020 or other portion of the actuable element 3014 may include a layer or portion of magnetic material. The base 3020 of the actuable element 3014 may be positioned proximate the gap 3008 such that the base 3020 can be displaced relative to the gap 3008 upon application of a magnetic field on the magnetic material of the actuable element 3014. Optionally, a cantilever 3024 or other mechanism for controlling the displacement of the actuable element may be coupled to the actuable element 3014 at one end and to the substrate 3054 at another end. Exemplary control mechanisms are previously described.

[0189] In the exemplary embodiment illustrated in Figs. 22A, 22B, and 22C, the MEMS actuator 3012 can be constructed on the surface of the substrate 3054 and the actuable element 3014 and the cantilever 3024 can be constructed from the substrate 3054, preferably from a top device layer 3052 of the substrate 3054. In this manner, the actuable element 3014 and the cantilever 3024 may be suspended over a cavity 3016 (denoted by crosshatching in Fig. 22A) in the substrate 3054 and thus may be free to be displaced relative to the substrate 3054. Using the manufacturing processes previously described, all or at least a portion of the components of the MEMS device, e. g. , the actuator 3012, the actuable element 3014 and/or the cantilever 3024 (or other control mechanism), may be constructed from one or more layers of the substrate 3054 to reduce the extent of the MEMS device 3010 in a direction perpendicular from the substrate surface. Alternatively, all or portions of the components of the MEMS device 3010 may be constructed on the surface of the substrate 3054 through deposition processes or a combination of additive/subtractive processes.

[0190] The MEMS device 3010 can provide precise electronic control of the position of the actuable element 3014. Operation of the MEMS device 3010 can be understood in the following manner. An external power source (not shown) can provide an operating current that can be communicated to the actuator 3012 through the first and second leads 3018, 3019, respectively. After receiving an operating current, the actuator 3012 may selectively apply a first force to the actuable element 3014. The first force can tend to displace the actuable element 3014 relative to the gap 3008 along a path 3006 that may be substantially

parallel to the axis of actuation 3026. The displacement of actuable element 3014 can result in bending of the cantilever 3024, which can effectively apply a second force to the actuable element 3014. Like the first force, the second force also may act along the axis of actuation 3026. Unlike the first force, however, the second force may act in an opposite direction, tending to restore the actuable element 3014 to its original position on the path 3006 and tending to limit the displacement of actuable element 3014 along the path 3006.

Based on known beam theory analysis of the cantilever 3024, the second force may increase as the displacement of actuable element 3014 increases along the path 3006. Thus, for a given first force, there can be a displacement of actuable element 3014 where the second force balances the first force.

[0191] Upon removal of the first force, as by removal of the operating current, the second force restores the actuable element 3014 to its original position on the path 3006, i. e. , the cantilever 3024 may return to its original unbent or relaxed, position. By varying the operating current to the MEMS actuator 3012, the MEMS device 3010 provides precise electronic control of the displacement of the actuable element 3014 along the path 3006. In the embodiment of Figs. 22A, 22B, and 22C, the second force can be provided by the cantilever 3024. Other embodiments may include other means, including a clamp, spring, suspension system, or a second MEMS actuator to control the displacement of actuable element 3014 along the path 3006, as previously described.

[0192] As shown in Figs. 22A, 22B, and 22C, each of the path 3006, the actuator 3012, and the actuable element 3014 can lie in a plane that can be substantially parallel to the substantially planar surface of the substrate 3054. Alternately, any combination of the path 3006, the actuator 3012, and the actuable element 3014 may lie in a plane that can be inclined at an angle to the substantially planar surface of the substrate 3054. Also alternately, the path 3006 may include constituent portions that may not be coplanar with each other, the actuator 3012 may include constituent portions that may not be coplanar with each other, and the actuable element 3014 may include constituent portions that may not be coplanar with each other. Additionally, the path 3006, the actuator 3012, and the actuable element 3014 may not be coplanar with each other.

[0193] The actuator 3012 can include a magnetic core 3030 and an electrically conductive coil 3032 (shown partially in Fig. 22A for clarity) wrapped around the core 3030 in a manner analogous to a solenoid or a toroid. Preferably, the actuator 3012 is substantially planar, and lies in a plane that is substantially parallel to the substantially planar surface of

the substrate. The core 3030 can have a substantially circular shape when viewed from above and may have a substantially rectangular cross-section that has an extent that is substantially constant across the circumference of the core 3030. Alternately, the core 3030 may have a variety of shapes when viewed from above and a variety of cross-section shapes. For example, these shapes may include the shape of any type of oval, including an ellipse and a circle, any type of polygon, including a rectangle, a square, and a triangle, and the shape of any type of semi-oval, including a semi-circle. Also alternately, the cross- section may have an extent that varies across the circumference of the core 3030.

[0194] The core 3030 can have a first core end 3034 and a second core end 3036 separated from each other by the gap 3008. The gap 3008 can have a size and shape that can permit the displacement of actuable element 3014 therethrough. The size and shape of gap 3008 may depend on a variety of core parameters including the core shape and the core composition. The width of the gap 3008, or gap distance d between core ends 3034,3036, and the cross-sectional areas of the first and second core ends 3034,3036 can be varied to adjust the strength of the magnetic field generated at the gap 3008. The magnetic field strength will decrease with increasing gap distance d, and/or decreasing cross-sectional areas of core ends 3034,3036. The size and shape of the gap 3008 may also depend upon the limitations of the fabrication process.

[0195] As previously described, the base 3020 of actuable element 3014, or a portion of the base 3020 may be constructed from a magnetic material (e. g. , permanent or soft magnetic material), or may have a layer of magnetic material thereon. Displacement of the actuable element 3014 along the path 3006 through the gap 3008 may depend, at least in part, upon the relationship between the magnetic field generated in the gap 3008 by the actuator 3012, as previously described, and the magnetic dipole properties of the magnetic material of base 3020. The magnitude of the activation force that tends to pull or push the actuable element 3014 along the path 3006, and hence the displacement of the actuable element 3014 along the path 3006 may depend on the volume of magnetic material within the gap 3008. The volume of the magnetic material provided on base 3020 can be varied to adjust the magnitude of the force acting on the magnetic material. Varying at least one of or a combination of the width, height, or length of magnetic material may vary the volume of magnetic material on the base. For example, as the width of magnetic material in the gap 3008 becomes smaller, the force acting on the magnetic material may also become smaller.

[0196] In the embodiment of Figs. 22A, 22B and 22C, the displacement of actuable element 3014 along the path 3006 may be controlled by the action of cantilever 3024. As described previously, bending of cantilever 3024 under the influence of the force on base 3020 resulting from the magnetic field applied at gap 3008 can control the amount of displacement of actuable element 3014 along the path 3006. From well-known beam theory analysis, the resultant displacement in the plane of Figs. 22A, 22B, and 22C perpendicular to a longitudinal axis 3038 of cantilever 3024 may vary as the cube of the width W of the cantilever 3024. Thus, small variations in the width of cantilever 3024, as may result from limitations in the fabrication, can result in large variations in the displacement of actuable element 3014 along the path 3006.

[0197] The relationships of the first force, i. e. , the force resulting from the magnetic field acting on the magnetic material of the base 3020 at gap 3008, and the second force, i. e. , the force due to bending of the cantilever 3024, to the displacement of the actuable element 3014 along the path 3006 can be shown in Figs. 23A and 23B. In the illustrated embodiment, the first force is constant as the actuable element is advanced through the gap.

To achieve a desired displacement Xo of the actuable element 3014, a constant first force can be applied to actuable element 3014, as indicated by line 3060. Upon applying the first force, displacement of the actuable element 3014 may increase until the second force may equal the first force, as indicated by line 3062.

[0198] Changes in the width of cantilever 3024 can result in changes in the stiffness, or bending resistance, of cantilever 3024. In Fig. 23A, the change in stiffness can be reflected in the slope of line 3062. By increasing the width of cantilever 3024 by an amount AW, the stiffness of cantilever 3024 may be increased, as indicated by the increased slope of line 3064 with respect to line 3062. The resultant displacement may be decreased, as indicated by the intersection of line 3064 and line 3060. In a similar fashion, decreasing the width of cantilever 3024 by an amount AW may decrease the stiffness of cantilever 3024 and the resultant displacement may be increased, as indicated by line 3066. Thus, the AW variation in the width of cantilever beam 3024 can result in a total variation of AX in the displacement of actuable element 3014.

[0199] It can be seen from Fig. 23A, that, for the same iaW, if the first force can be made to be decreasing near the desired displacement, as indicated by line 3060a, the variation in the displacement can be lessened, as indicated by AX', where AX'< AX. As previously described, the actuation force exerted on actuable element 3014 can depend on the strength

of the magnetic field and also on the volume of magnetic material within gap 3008. Fig.

24A provides a detailed view of the base 3120 of actuable element 3114, first and second core ends 3134,3136 and gap 3108. In the embodiment of Fig. 24A, base 3120 can be rectangular and may be fabricated from a magnetic material, or may have a layer of magnetic material thereon. Thus, for the embodiment of Fig. 24A, the amount of magnetic material within gap 3108, as actuable element 3114 travels through gap 3108, can remain constant.

[0200] However, the strength of the magnetic field in the gap 3108 can be varied by changing a gap distance d between first and second core ends 3134,3136. The gap distance d can be varied by angling all or a portion of one or both core ends 3134,3136 with respect to the axis of actuation 3126. In Fig. 24A, core end 3134 can form an angle a with axis 3126. Core end 3136 can have a first portion 3136a parallel with axis 3126 and a second portion 3136b can form and angle ß with axis 3126. Thus, core ends 3134,3136 may have differing shapes and the angles a and ß may differ. Angling of core ends 3134,3136, as shown in Fig. 24A, can result in the gap distance d increasing in the direction corresponding with the increasing displacement shown in Figs. 23A and 23B and illustrated by arrow 3180 in Fig. 24A. It can be understood that the angling of core ends 3134,3136 may be such as to increase or decrease the gap distance d in the direction 3180. Also, the gap distance d can be made to increase over a portion, remain constant over a portion and/or decrease over a portion, in any desired order and/or combination. It can also be understood that core ends 3134,3136 may be curved, such that the gap distance d may vary non-linearly.

[0201] In a preferred embodiment shown in Fig. 24B, core ends 3234,3236 can be angled symmetrically about axis 3226, i. e. , the angle a may be equal to the angle ß. For the preferred embodiment of Fig. 24B, the angles a and ß may be chosen such that tan (a) = tan (p) z 3.0, or a = p ? ; 71. 6°. In the embodiment of Fig. 24B, dl may be approximately 63 p. m, xi may be approximately 21 llm and the gap 3208 may have a first width wl of approximately 60 Fm. The magnetic material on base 3220 may have an extent between core ends 3234,3236 of approximately 50 um and may have an extent along the axis 3226 of approximately 100 ju. m. In this embodiment, the magnetic field strength between the core ends 3234,3236 will decrease as the gap distance d increases. As base 3220 of actuable element 3214 moves through gap 3208 in the direction of arrow 3280, the force

exerted on the magnetic material of base 3220 will decrease. Thus, the force exerted may generally follow the shape of line 3060a in Figs. 23A and 23B and the fabrication variations in the width of cantilever 3224 may be have a diminished effect on the variation of the displacement, AX. For the embodiment of Fig. 24B, the decreasing force may be realized at a displacement of around 35 jum to provide a desired displacement Xo of about 40, um, with a 5% accuracy.

[0202] As previously noted, the force exerted on the magnetic material of base 3020 may depend on the volume of magnetic material within gap 3008. Fig. 25 provides a detailed view of the base 3320 of actuable element 3314, first and second core ends 3334,3336 and gap 3308 of an embodiment of the MEMS device 3310. As in Fig. 24A, base 3320 of Fig.

25 may be fabricated of magnetic material, or may have a layer of magnetic material thereon. In the embodiment of Fig. 25, core ends 3334,3336 may be parallel with respect to axis 3326. Thus, for the embodiment of Fig. 25, the gap distance d may remain constant.

However, in the embodiment of Fig. 25, the volume of magnetic material provided on base 3320 may be varied to effect a change in the magnetic moment of the base 3320. In the illustrated embodiment of Fig. 25, a layer of magnetic material is deposited over the entire top surface of the base 3320. However, it can be understood that, the layer of magnetic material need not conform to the shape of the top surface of the base 3320. For example, the layer of magnetic material may be applied to a portion of the top surface of the base 3320 or another portion of the base 3320.

[0203] In the illustrated embodiment of Fig. 25, base 3320, and thus the corresponding layer of magnetic material, may have a first portion 3320a having a constant width and a tapered portion 3320b having a width that can vary along the length of base 3320, i. e., along the axis 3326. In the embodiment shown in Fig. 25, the width of tapered portion 3320b can be seen to taper symmetrically about the axis 3326, with the width of base portion 3320 within the gap 3308 decreasing as actuable element 3314 may move in the direction of arrow 3380. By decreasing the width of base portion 3320 and, thus, the corresponding layer of magnetic material deposited on the base 3320, the volume of magnetic material provided on the base 3320 is reduced, resulting in a decrease in the actuation force applied to the actuable element 3314 as the actuable element advances through the gap. In a manner similar to that described in relation to Fig. 24B, the force exerted may generally follow the shape of line 3060a in Figs. 23A and 23B and the

fabrication variations in the width of cantilever 3324 may be have a diminished effect on the variation of the displacement, aX.

[0204] It will be understood that in other embodiments the width, length, and/or height (i. e., thickness) of magnetic material provided on the base portion 3320 can be varied to vary the volume of magnetic material provided on the base. For example, the width of tapered portion 3320b may taper from one side or the other of base portion 3320, or the taper may be only a partial taper, i. e. , the taper may not extend over the full width of base portion 3320. Additionally, the taper need not be linear, but may vary along the length of tapered portion 3320b. Further, tapered portion 3320b may extend the full length of base 3320, i. e., there may be no first, or constant width, portion 3320a, or the tapered portion 3320b may be interchanged with the first portion 3320a, such that the tapered portion 3320b may be adjacent elongated arm 3322 of actuable element 3314 in Fig. 25. It can also be understood that the width of tapered portion may increase or decrease in the direction of arrow 3380, or portions may have an increasing width, while other portions may have a decreasing width, or a constant width. In certain exemplary embodiments, the height (i. e. , thickness) of all or a portion of the magnetic layer may be varied.

[0205] In certain exemplary embodiments, both the volume of magnetic material provided on base 3320 and the gap distance d may be varied to increase and/or decrease the activation force on the actuable element 3314 as it advances through the gap 3308.

[0206] It can also be understood that the stiffness of cantilever 3024 or other control mechanism may be varied such that the same AW can have less effect on the variation in the displacement, aX. Referring to Figs. 26A and 26B, there can be shown top and side elevation views of an exemplary embodiment of a MEMS device 3410, corresponding to Figs. 22A and 22C, respectively, of MEMS device 3010, that can be fabricated to have an effective stiffness that increases after a known displacement may be reached. A pin 3472 can be constructed from the substrate 3454, and can extend the depth of substrate 3454. Pin 3472, or other suitable structure, may be positioned in the deflection path of cantilever 3424 to alter the stiffness of the cantilever upon contact with the cantilever. When the magnetic field at gap 3408 between core ends 3434,3436 acts on the magnetic material of base 3420 of actuable element 3414 to move actuable element 3414 in the direction of arrow 3480, cantilever 3424 can deflect in the manner of cantilever 3024 of Figs. 22A, 22B, and 22C, until cantilever 3424 contacts pin 3472.

[0207] When contacted by cantilever 3424, the pin 3472 can provide additional support for cantilever 3424, thus increasing the force required to obtain the desired displacement Xo compared to a similar cantilever without the pin 3472, e. g. , effectively increasing the spring constant Kl of the cantilever. Referring Fig. 23B, the increased resistance to bending can be reflected in steeper slopes of lines 3062,3064 and 3066, with consequent reduction of the variation in the displacement, AX. Line 3062a can illustrate the relationship of the bending force of a cantilever 3424 to the displacement of actuable element 3414. As can be seen, the slope of line 3062a can increase, i. e. , the resistance to bending can increase (spring constant increases from Kl to K2), when cantilever 3424 contacts pin 3472.

[0208] Fig. 26C is a top view, corresponding to Fig. 26A, of one embodiment of cantilever 3524, wherein the width of cantilever 3524 can be increased along a length of cantilever 3524 from the point at which the cantilever 3524 contacts the pin 3572 to the end of cantilever 3524 coupled to actuable element 3514. The increased width can provide additional stiffness to further increase the slope of line 3062a. It can be understood that increasing a depth of cantilever 3524 may also increase the stiffness or bending resistance of cantilever 3524. Conversely, decreasing a depth of the cantilever 3524 may also decrease the stiffness or bending resistance of cantilever 3524. While the embodiments of Figs. 26A, 26B, and 26C may be described in relation to an electromagnetic actuator, it can be understood that actuator 3412,3512 may be another type of MEMS actuator such as, for example, an electrostatic, thermal, magnetostrictive, or piezoelectric actuator or any other type of MEMS actuator.

[0209] In cross-section, pin 3472 may be round, as may be shown in Fig. 26A, oval, oblong, or other arcuate shape as may be desired. Additionally, in one embodiment illustrated in Fig. 26C, pin 3572 may have a polygonal cross-section. A polygonal cross- section can provide a plurality of points where cantilever 3524 can contact pin 3572, with bending resistance increasing as cantilever 3524 contacts the successive contact points.

Fig. 26C illustrates an embodiment having two contact points designated a and b, respectively. Cantilever 3524 may first make contact at point a of pin 3572, with a concomitant increase in bending resistance. As the displacement of actuable element 3514 and bending of cantilever can continue to increase, cantilever 3524 may then make contact at point b of pin 3572, with a further increase in bending resistance. It can be understood that pin 3572 may be shaped to provide additional contact points as may be needed to attain the desired displacement. In one embodiment, multiple pins may be provided to provide

multiple contact points. It can be understood from the above description that other configurations of pin 3472,3572 may be used. As an example, pin 3472,3572, or other motion limiting structure, need not be placed within cavity 3416,3516, but may be a portion of an end wall 3480, 3580 of the MEMS device 3410,3510, as indicated in phantom in Figs. 26A, 26C.

[0210] The location of pin 3472 may depend on the forces to applied to the actuable element and the desired displacement of the actuable element. Figs. 27A, 27B, and 27C illustrate an exemplary methodology for determining the location of the pin 3672. Fig. 27A illustrates a cantilever 3624 that is coupled at a distal end 3614 to an actuable element, with pin 3672 located a distance D from the rest position of distal end 3614 of cantilever 3624, i. e., a distance Dx from the displacement path 3606 of the actuable element and a distance Dy from a longitudinal axis 3638 of the cantilever 3624. A displacement 81 of the distal end 3614, resulting from a force F1, can be such that cantilever 3624 can contact pin 3672.

Using known beam formulas, the following relationships can be determined: F, = 6EIDy/{(2L + DX) (t-Dx)} ; , = 2L3Dy/{(2L+Dx)(L-Dx)2}; and [1] K1 = F1/#1 = 3EI/L3, where E is Young's modulus, I is the second moment of inertia of the cantilever 3624, L is the length of cantilever 3624 and Kl is the spring constant of cantilever 3624 as defined in relation with Fig. 23B.

[0211] The force on actuable element 3614 may be increased to a force Fo, corresponding with line 3060 of Fig. 23B, i. e., Fo = Fl + AF, with a subsequent increase in the deflection of cantilever 3624, to achieve the desired displacement, i. e., Xo = (51 + AS, as illustrated in Figs. 23B and 27B. Again using known beam formulas, the following additional relationships can be obtained: X0 = F0Dx2(3L+Dx)/12EI+(2L+Dx)Dy/2(L-Dx)=(F0-F1)Dx2(3L+Dx)/12EI+# 1 ; K2 = #F/##=12EI/Dx2(3L+Dx); and α = K2/K1 = 4L3/Dx2(3L + Dx), [2] where K2 can be the spring constant after cantilever 3624 contacts pin 3672 and a is the ratio of the spring constant K2 to the spring constant Kl.

[0212] Using the above relationships, the following method 3700, as illustrated in Fig. 27C, can be used to determine the placement of pin 3672 in terms of Dx and Dy, such that the displacement of the actuable element 3614 can be controlled to within a desired displacement accuracy. Given the ideal spring constant Kl for the cantilever 3624, a known force Fo and a fabrication tolerance AW in width W of cantilever 3624, and since Kl oc I oc W3, the variations in the spring constant Kl can be determined (3702) from: K, + = K, (1 + AW/W) 3 and K,-= K, (1-AW/W) 3.

Using the above relationships for a and Kl, K2, the following can be obtained (3704): aKI (X0-#1)=F0-K1#1; αK1+(X0+-#1)=F0-K1+#1; and aK, (X0-#1) = Fo-Kl 61 For a given desired displacement accuracy of Yo, i. e., (Xo--Xo +)/Xo < Yo, where X0+ can be a minimum tolerable displacement and Xo-can be a maximum tolerable displacement, oc can thus be determined (3706): α > F0(1/K1--1/K1+)/(Y0X0).

Dx can then be determined from Equation 2 (3708) and Dy can then be determined from Equations 1 and 3 (3710). It can be seen from Fig. 23B that the variation in the displacement, as indicated by Aux', may be lessened from what may be expected without pin 3672.

[0213] Referring back to Fig. 26A, it can be understood that the closer pin 3472 may be placed to the axis of actuation 3426, the greater the increase in bending resistance when cantilever 3424 contacts pin 3472. Also, the closer pin 3472 may be placed to the longitudinal axis 3438 of cantilever 3424, the smaller the overall displacement of actuable element 3414 may be. In one embodiment, multiple cantilevers 3524 may be used and one or more pins 3572 may be placed in the displacement path of one or more of the multiple cantilevers 3524. Such multiple cantilevers are previously described in more detail. It can further be understood that other mechanisms for controlling the displacement of actuable element 3414,3514 may be used, such as previously described in further detail. In embodiments employing a spring control mechanism, pin 3472,3572 may be positioned to contact the spring control mechanism to effectively increase the spring constant of the

spring control mechanism in a manner analogous to increasing the spring constant Kl of cantilever 3424,3524.

[0214] It is known in the art that strong interfacial adhesion, commonly referred to as stiction, may be present between contacting microstructure surfaces, such as contacting surfaces of cantilever 3524 and pin 3572. When stiction can occur between contacting surfaces of cantilever 3524 and pin 3572, then actuable element 3514 may not return to its rest, or relaxed, position when the magnetic field force is turned off. Known methods can be used to diminish stiction between cantilever 3524 and pin 3572, including texturing of one or both of the surfaces of cantilever 3524 and pin 3572 and applying an anti-stiction coating to one or both of the surfaces of cantilever 3524 and pin 3572.

[0215] While the MEMS devices disclosed herein have been particularly shown and described with reference to the exemplary embodiments thereof, those of ordinary skill in the art will understand that various changes may be made in the form and details herein without departing from the spirit and scope of the disclosure. As an example, one or more of the corners formed in the actuable elements, the sidewalls, and/or the stops illustrated in Figs. 18A, 18B, 18C, and 19-21 may be replaced with radius corners for ease of fabrication or by design choice. Those of ordinary skill in the art will recognize or be able to ascertain many equivalents to the exemplary embodiments described specifically herein by using no more than routine experimentation. Such equivalents are intended to be encompassed by the scope of the present disclosure and the appended claims.

[0216] We claim: