CN1567018A - Small piece with micro lens array - Google Patents

Abstract

The invention provides a single chip made of a wafer on which a micro-mirror array is formed, the micro-mirrors constituting the micro-mirror array being arranged above the circuits and electrodes for driving them, the area of which is from 1 square centimeter to 1 square inch; the micro-mirrors are quadrilateral and have no side parallel to the side of the chip, the horizontal rows of micro-mirrors extend landing from one corner to the other following the horizontal side of the rectangular array, and a vertical line corresponding to the addressed column in each row extends from each micro-mirror to connect to the micro-mirrors of each other row; and vertical columns of the microlenses extending generally along the vertical edges of the rectangular array from one corner to the other, a horizontal line in each column corresponding to the addressed row extending from each microlens and connecting to the microlenses in the other columns; wherein a number of row lines and column lines are provided for addressing the micromirror and the number of row lines multiplied by the number of column lines is larger than the number of pixels.

Description

Small piece with micro lens array

The present application is a divisional application, and the original application is international patent application PCT/US01/24332 with the international application date of 2001, 8 and 3, which is a phase of entering china at 30/1/2003, and the chinese national application number of the original application is 01813608.7.

Technical Field

The present invention relates to a tablet, and more particularly to a tablet having an array of microlenses.

Background

U.S. patent nos. 5,835,256 and 6,046,840 to Huibers, and U.S. patent application No. 09/617,419 to Huibers et al, the subject matter of each of which is incorporated herein by reference, disclose microelectromechanical devices (MEMS) for controlling light beams, such as in optical switches, and/or for displays (e.g., projection displays). A common feature is the movable micro-mirror to deflect the light beam through different angles depending on the tilt angle of the micro-mirror elements. In one type of conventional direct view display system or projection display system, reflective micromirror elements are used to generate the image. Typical micromirror elements are square and have a single tilt angle for the "on" state, flat for the "off" state, or the same tilt angle for the "on" state and the "off" state, but opposite signs.

Disclosure of Invention

The present invention provides an individual chip made of a wafer and having formed thereon an array of micro-mirrors arranged in a two-dimensional form; in the chip, the micro-mirrors constituting the micro-mirror array are disposed above the circuit and electrodes for electrostatically driving the micro-mirrors; wherein the micro-mirror is quadrilateral and no side is parallel to a side of the platelet; wherein the micro-mirror array is rectangular and has an area of from 1 square centimeter to 1 square inch; wherein the horizontal rows of micro-mirrors extend floor-to-floor from one corner to the other in rows parallel to one side of the rectangular image, and wherein a vertical line in each row corresponding to an addressed column extends from each micro-mirror to be connected to the micro-mirrors of each other row; and wherein vertical columns of the micro-mirrors extend landing from one corner to the other in columns parallel to one side of the rectangular image, and wherein horizontal lines in each column corresponding to an addressed row extend from each micro-mirror and are connected to the micro-mirrors of each other column; wherein a number of row lines and column lines are provided for addressing the micromirror, and wherein the number of row lines multiplied by the number of column lines is larger than the number of pixels.

In order to minimize diffraction of lightalong the switching direction, and in particular to minimize diffraction of light reaching the acceptance cone of the collection optics, non-rectangular microlenses (as used herein "rectangular" includes square microlenses) are used in the present invention. Diffraction as used herein refers to scattering of light off of a periodic structure, where the light is not necessarily monochromatic or phase coherent. Also, to minimize the cost of the illumination optics and the size of the display unit of the present invention, the light sources are placed orthogonal to the rows (or columns) of the array and/or the light sources are placed orthogonal to one side of a frame that defines the active area of the array. The incident beam, while orthogonal to the sides of the rows (or columns) and/or active area, is not necessary for the sides of the individual micro-mirrors in the array to be orthogonal. The orthogonal sides cause the incident beam to diffract in the direction of the micromirror switch and cause light to "leak" into the "on" state, even if the micromirror is in the "off" state. This diffraction of light reduces the contrast ratio of the micromirror.

The present invention optimizes the contrast ratio of the micromirror array such that when the micromirrors are in their "off" state, they send the least light to the spatial region where the light is controlled to reach when the micromirrors are in their "on" state. More specifically, the present invention includes a specially positioned light source and incident light beam, and specially designed micro-mirrors in the array that minimize diffraction of light reaching the acceptance cone of the projection (or viewing) optical system to provide improved contrast ratios. By taking into account the tight fit of the micro-mirrors and the large fill factor (fill factor) with low diffraction from the "off" state to the "on" state, the arrangement and structure of the present invention minimizes the non-reflective area in the array even when the array is illuminated along the optical axis of the micro-mirror period. That is, by the angled edges not being parallel to the axis of rotation of the micromirror, the structure optimizes contrast ratio, and optimizes fill factor by the hinge (hinge) which requires relatively small area and allows adjacent micromirrors to tilt together with a very small amount of wasted non-reflective area. The configuration and shape of the microlenses of the various examples of the present invention also reduces cross-talk between adjacent microlenses when the microlenses are electrostatically deflected.

Another aspect of the invention is a micromirror array in which individual micromirrors are tilted asymmetrically about a flat or non-deflected state. By having the "off" state of the micromirror at an angle smaller than the opposite angle of the micromirror in the "on" state, a) diffracted light from the edge of the micromirror and entering the collection optics is minimized; b) light scattered from underneath the micromirror and entering the collection optics is also minimized; c) the travel of the micro-mirrors is reduced, which minimizes the possibility of adjacent micro-mirrors colliding with each other, which in turn reduces the gaps between the micro-mirrors and increases the fill factor of the micro-mirrors; d) the angle of deflection of the micromirror can be increased to a larger range than that of the micromirror array arrangement having the same deflection angle for the on-state and the off-state.

Another aspect of the invention is a package for a micromirror array, the package having a portion capable of transmitting light that is not parallel to the substrate on which the micromirror is formed. The light-transmitting portion may be any suitable material, such as a glass plate, quartz, or polymer, and is contemplated to direct specular reflection from the light-transmitting substrate rather than from a parallel light-transmitting plate in the package. Preferably the specular reflection is directed sufficiently away from the collection optics so that an increase in the size of the illumination cone will keep the specular reflection from entering the collection optics.

In yet another aspect, the invention is a projection system comprising an array of active micro-mirrors arranged in a rectangle, the micro-mirrors being capable of rotation about a switching axis between an off-state and an on-state, the micro-mirrors corresponding to pixels in a viewed image; comprising a light source for directing light to the array of micro-mirrors, the light source being arranged so as to direct light non-perpendicular to at least two sides of each micro-mirror and parallel to at least two other sides of each micro-mirror when each micro-mirror is viewed as a top view; comprising a collection optical system arranged to receive light from the micro-mirror in the on-state.

Another aspect of the invention is a projection system comprising an array of micro-mirrors, each micro-mirror corresponding to a pixel in a viewed image and having the shape of a concave polygon or one or more non-rectangular parallelograms; a light source for directing light to a collection optics of the micromirror array, the collection optics being arranged to receive light reflected from the micromirror.

Yet another aspect of the invention is a projection system comprising a light source for providing an incident light beam, an array of movable reflective elements, and a collection optics for projecting the light from the array, wherein a projected image from the projection system is to be displayed on a target in a rectangular image comprised of thousands to millions of pixels, each pixel being shaped as a concave polygon, a single non-rectangular parallelogram, or a combination of non-rectangular parallelograms.

Yet another aspect of the invention is a projection system comprising a light source, an array of movable micro-mirror elements, and collection optics, wherein each micro-mirror element in the array has a switching axis that is substantially parallel to at least one edge of the array's active area and at an angle of 35 to 60 degrees to one or more edges of the micro-mirror element.

Another aspect of the invention is a projection system comprising a light source and an array of movable micro-mirror elements, each micro-mirror element having a leading edge that is not perpendicular to the incident light beam and not perpendicular to any edge of the active area, such that the contrast ratio is increased by a factor of 2 to 10 compared to a micro-mirror element having an edge that is perpendicular to the incident light beam.

Another aspect of the invention is a projection system comprising a light source, collection optics and an array of movable micro-mirror elements, the projection system having a diffraction pattern (pattern) substantially as shown in FIG. 21C.

Yet another aspect of the invention is a projection system comprising a light source and a rectangular array of movable micro-mirrors, the micro-mirrors being movable between an on-state and an off-state and being capable of reflecting light in the on-state to a predetermined spatial area, wherein the light source is arranged to direct light at an angle of substantially 90 degrees to at least one side of the rectangle defined by the array, and wherein substantially no diffracted light enters the predetermined spatial area when the micro-mirrors are in the off-state.

Another aspect of the invention is a method of projecting an image onto a target, comprising: directing the light beam to a rectangular array of micro-mirrors, the light beam being directed at an angle to the side in front of the rectangular array, the angle being in the range of plusor minus 40 degrees plus 90 degrees, and wherein the micro-mirrors in the array are polygonal in shape and positioned so that the light beam is incident on all sides of the polygon at an angle other than 90 degrees; and projecting light from the micro-mirror onto a target to form an image thereon.

Another aspect of the invention is a projection system comprising a light source, a light collection system, and an array of micro-mirrors arranged to spatially modulate light from the light source, the array being formed on a substrate and configured such that each micro-mirror is capable of being in a first position when not actuated, each micro-mirror being capable of being moved to an on position to direct light to the light collection system of the array and being capable of being moved in an opposite direction to an off position to direct light away from the light collection system, both the on position and the off position being different from the first position, and wherein the on position is at an angle relative to the first position that is different from the angle of the off position.

Another aspect of the invention is a method for spatially modulating a light beam comprising directing a light beam from a light source to a light collection system through an array of micro-mirrors arranged to spatially modulate the light beam from the light source, the array being formed on a substrate and each micro-mirror being in a first position when not modulated, modulating the micro-mirrors in the array causing each micro-mirror to move to an on position in which it directs light to the light collection system of the array and to an off position to direct light away from the light collection system, both the on position and the off position being different from the first position, and wherein the on position is at an angle with respect to the first position that is different from the angle at the off position.

Another aspect of the invention is also an optical micromechanical element formed on a substrate, the element having an open position at a first angular value relative to the substrate, and a third position substantially parallel to the substrate, the first and second values being different at respective positions at a second angle relative to the substrate, the open and closed positions being defined by the optical micromechanical element abutting the substrate or abutting a structure formed on said substrate.

Yet another aspect of the invention is a method for modulating light comprising reflecting light from an array of deflectable micro-mirrors disposed on a planar substrate, said micro-mirrors being tilted to a first position or to a second position; wherein the angle formed between said first position and the substrate is substantially different from the angle formed between said second position and the substrate.

Another aspect of the invention is a method for modulating light comprising a light source, a planar array of light modulators, the array comprising deflectable elements and collection optics, wherein the elements in the array are selectively configured into at least two states, wherein in a first state the elements direct light from the light source into the collection optics through a first angle, in a second state the elements direct light from the light source into the collection optics through a second angle, and a third angle represents light reflected from the array as if the array were a micro-mirrored surface, wherein a difference between the first angle and the third angle is substantially different from a difference between the second angle and the third angle.

Another aspect of the invention is a projection system comprising a light source for providing a light beam; a micro-mirror array comprising a plurality of micro-mirrors disposed in the beam path; and acollection optical system disposed in the optical path of the light beam after the light beam is incident on the array of micro-mirrors and reflected off a plurality of micro-mirrors in the array as on and off micro-mirror modes; wherein the micromirror array comprises a substrate on which the micromirror array is fixed, each micromirror on the substrate being movable from a non-deflected position to an on position and an off position, wherein the on position and the off position are at different angles with respect to the non-deflected position.

Another part of the invention is also a method for projecting an image onto a target comprising directing a light beam from a light source onto a micro-mirror array; modulating each of the micro-mirrors to an on position or an off position, wherein in the on position the micro-mirror directs light to a collection optical system arranged to receive light from the micro-mirror when the micro-mirror is in the on position, wherein the pattern of opening and closing the micro-mirror forms an image; and wherein the value of the angle subtended by the position of the micromirror in the micromirror open position is different from the value of the angle subtended by the position of the micromirror in the micromirror closed position.

Still another part of the present invention is a method for spatially modulating a light beam comprising directing a light beam onto an array of micro-mirrors, the array of micro-mirrors being movable to a first or second position, wherein in the first position the micro-mirrors direct a portion of the light beam incident thereon into a collection optical system, and wherein the minimum distance between adjacent micro-mirrors of each micro-mirror in the second position is less than the minimum distance between adjacent micro-mirrors of each micro-mirror in the first position.

Another aspect of the invention is an apparatus comprising a substrate on which areflective or diffractive micromechanical device is formed; a package for a fixed substrate and a movable micromechanical device; wherein the package includes an optically transmissive window that is not parallel to the substrate.

Still another aspect of the invention is a projection system comprising a light source; a light collection optical system; a substrate on which a reflective or diffractive micromechanical device is formed; a package for a fixed substrate and a movable micromechanical device; wherein the package comprises an optically transmissive window that is non-parallel to the substrate; an encapsulated micromechanical device disposed in the path of a light beam from a light source is used to modulate light from the light beam, and a collection optical system collects the modulated light.

Still another part of the invention is a projector (projector) comprising a light source, a packaged microelectromechanical (MEMS) device having a substrate with a micromechanical device thereon and a window in the package, the window being disposed at an angle to the substrate, and collection optics disposed to receive light modulated by the packaged microelectromechanical device (MEMS) from the light source.

Another aspect of the invention is a method for manufacturing a micro-lens comprising providing a substrate;

depositing a first sacrificial (sacrificial) layer on the substrate and patterning the first sacrificial layer; depositing at least one hinge (hinge) layer on the sacrificial layer and patterning the at least one hinge layer to form at least one flexible hinge; depositing and patterning a second sacrificial layer, depositing at least one lens layer over the second sacrificial layer and patterning the at least one lens layer to form a lens element, and removing the first and second sacrificial layers to release the micro-lenses.

Still another aspect of the invention is an optical micromechanical device, comprising a substrate; a first pillar on the substrate; a flexible hinge, wherein a proximal end of the flexible hinge is on the post; a second post connected to the distal end of the flexible hinge; and a plate attached to the second post.

Drawings

FIG. 1 is a top view of one embodiment of a micro-lens of the present invention;

FIGS. 2A through 2E are cross-sectional views of a microlens, taken along line 2-2 of FIG. 1, for illustrating a method of manufacturing a microlens of the present invention;

FIGS. 3A through 3D are cross-sectional views of the same microlens shown in FIGS. 2A through 2E, but taken along line 3-3 of FIG. 1;

fig. 4A to 4J are cross-sectional views of a micro-lens for explaining another method of manufacturing a micro-lens according to the present invention;

fig. 5A to 5G are cross-sectional views of a micro-lens for explaining still another method of manufacturing a micro-lens according to the present invention;

fig. 6A to 6C are plan views of differently shaped micromirror plate and hinge combinations;

FIG. 7 is a plan view of a portion of a micromirror array having a plurality of micromirrors identical to those of FIG. 6A;

FIG. 8 is a partially exploded perspective view of one embodiment of the present invention;

FIGS. 9A to 9C are cross-sectional views illustrating the action of the micro-lenses of the embodiment of FIG. 8;

FIGS. 10A through 10D are cross-sectional views of a process according to yet another embodiment of the invention;

fig. 11A to 11C are cross-sectional views illustrating actions of a micro-lens manufactured according to the method illustrated in fig. 10A to 10D.

Fig. 12 is a plan view of a plurality of micro-mirrors in a micro-mirror array constructed in accordance with the method of fig. 11A through 11C.

Fig. 13 is a partially exploded perspective view of the micro-optic of fig. 12.

Fig. 14A to 14C show a micromirror having a flat non-deflected "off" state.

Fig. 15A to 15C show the micro-mirrors in the "on" and "off" states with equal angular deflection.

Fig. 16A to 16C show the micro-lenses having the angle of the "on" state larger than that of the "off" state.

Fig. 17A to 17E show an encapsulation of a micromirror plate with an angled window.

Fig. 18 is an illustration of an illumination system for a micromirror array of the present invention.

Fig. 19A to 19E show the relationship between incident light, the side of the micromirror plate, and the side of the active area.

Fig. 20 is a diagram of an example of a prior art micromirror array.

Fig. 21 and 22 are diagrams of embodiments of the invention in which square microlenses are at an angle to the sides of the active area.

Fig. 23 to 25 show micro-mirrors in which the "leading edge" and "trailing edge" of the micro-mirror are not perpendicular to the incident light beam.

Fig. 26A to 26F and 27A to 27F are illustrations of micro-lenses having one or more parallelogram shapes.

Figure 28 is an illustration of a single microlens.

Fig. 29 is an illustration of a micromirror array having front and back sides perpendicular to an incident light beam and another part at an angle of 45 degrees to the incident light beam.

Fig. 30 and 31 are illustrations of a micromirror array in which the micromirror has no edge parallel or perpendicular to the incident light beam and the edge of the active area of the array.

Fig. 32A to 32J are illustrations of micro-mirrors with corresponding hinge structures.

Fig. 33A to 33C are illustrations of diffraction patterns having diffraction lines passing through the acceptance cone of the collection optical system (33A) and avoiding the acceptance cones (33B and 33C).

Detailed Description

Methods for microfabricating movable microlenses or microlens arrays are disclosed in U.S. Pat. Nos. 5,835,256 and 6,046,840 to Huiber, the subject matter of each of which is incorporated herein by reference. A similar method for constructing the micro-lenses of the invention is illustrated in fig. 1-3. Fig. 1 is a top view of one embodiment of a micro-lens of the present invention. As can be seen in fig. 1, posts 21a and 21b support the mirror plate 24 by hinges 120a and 120b on a substrate having electrodes (not shown) thereon for causing deflection of the mirror plate 24. Although not shown in fig. 1, as will be discussed further herein, thousands or even millions of micro-mirrors 24 can be provided in an array for reflecting an incident beam thereon and projecting the image to a viewer or target/screen.

The micro-mirrors 24 and other micro-mirrors in the array can be manufactured in a number of different ways. One such method is shown in fig. 2A through 2E (taken along cross-section 2-2 of fig. 1), in which the microlenses are preferably fabricated on a light-transmissive substrate, which is then bonded to a circuit substrate. This method is further disclosed in U.S. patent provisional application No. 60/2,292,46 to lkov et al, filed on 30/8/2000, and in U.S. patent application No. 09/7,324,445 to lkov et al, filed on 7/12/2000. Although the method will be described in connection with a substrate that transmits light, any other suitable substrate can be used, such as a semiconductor substrate with circuitry. If a semiconductor substrate such as monocrystalline silicon is used, it may be preferable to electrically connect the posts of the micromirror plate to 3 layers of metal in the integrated circuit IC process, and to use a conductive material as at least a portion of the micromirror plate. The method of forming the micro-mirror directly on the circuit substrate (instead of on a separate, light-transmissive substrate) will be discussed in more detail herein.

As can be seen in FIG. 2A, a light-transmissive substrate 13 (at least before adding a layer thereon) such as glass (e.g., Corning1737F or Eagle2000), quartz, Pyrex^TMSapphire, etc. are provided. The optically transparent substrate may have a selective light barrier layer added to its underside to aid in handling the substrate during processing. Such a light blocking layer may be a TiN layer deposited by reactive sputtering to a thickness of 2000 angstroms behind the light transmissive substrate, which TiN layer is removed once processing is complete. The substrate may be of any shape and size, although the shape of a standard wafer used in integrated circuit fabrication is preferred.

As can be seen in fig. 2A, a sacrificial layer 14, such as amorphous silicon, is deposited. The sacrificial layer may be another suitable material that can later be removed from the micromechanical structure material (e.g., SiO)₂Polysilicon, polyimide, novolac, etc.). The thickness of the sacrificial layer can vary widely depending on the size of the movable element/micromirror and the desired tilt angle, and is preferably about 5000 angstroms, although the thickness can range from 500 angstroms to 50000 angstroms. As an alternative to amorphous silicon, the sacrificial layer can be any of a number of polymers, photoresist or other organic materials (or even polysilicon, silicon nitride, silicon dioxide)Silicon, etc., depending on the material selected to be resistant to the etchant, and the etchant selected). Selected adhesion promoters (e.g., SiO)₂Or SiN) may be applied prior to depositing the sacrificial material.

In order to provide a contact area between the substrate 13 and the later deposited micromechanical structure layer, a hole 6 having a width "d" is formed in the sacrificial layer. The holes are formed by spinning on the photoresist and directing light through a mask to increase or decrease the solubility of the resist, depending on whether the resist is a positive or negative resist. The dimension "d" can range in size from 0.2 to 2 microns (preferably about 0.7 microns), depending on the size of the final micromirror and the micromirror array. After developing the resist to remove the resist in the region of the holes, the holes are etched in the amorphous silicon of the sacrificial layer by chlorine or other suitable etchant (depending on the sacrificial material). The remaining photoresist is then removed, for example, by a plasma of oxygen. The pores in the sacrificial layer may be of any suitable size, although preferably have a diameter of from 0.1 to 1.5 μm, more preferably about 0.7+/-0.25 μm. The etching proceeds to the glass/quartz substrate or to any intermediate layers, such as adhesion promoter layers, preferably in an amount less than 2000 angstroms if the light transmissive substrate is completely etched. If the sacrificial layer 14 is a material that can be directly patterned, such as a novolac (novolak) or other photosensitive photoresist, then an additional photoresist layer deposited and developed over the sacrificial layer 14 is not required. In such a case, the sacrificial layer of photoresist is patterned to remove material in the region of the holes 6 and then selectively hardened prior to deposition of additional layers.

At this point, as can be seen in fig. 2B, the first structural layer 7 is deposited by, for example, chemical vapor (vapor) deposition. Preferably the material is silicon nitride or silicon oxide deposited by LPCVD (low pressure chemical vapour deposition) or PECVD (plasma enhanced chemical vapour deposition), however any suitable thin film material such as polysilicon, a metal or metal alloy, silicon carbide or an organic compound can be deposited at this point (of course the sacrificial layer and etchant should be appropriate for the structural material used). The thickness of the first layer may vary depending on the size of the movable element and the desired stiffness of the element, however in one embodiment the layer has a thickness of from 100 to 3200 angstroms, more preferably between 900 and 1100 angstroms. As can be seen in fig. 2B, the layer 7 extends into holes etched in the sacrificial layer.

As can be seen in fig. 2C, a second layer 8 is deposited. The material may be the same as the first layer (e.g., silicon nitride) or different (silicon oxide, silicon carbide, polysilicon, etc.) and can be deposited as for the first layer by chemical vapor deposition. The thickness of the second layer may be larger or smaller than the first layer, depending on the desired stiffness of the movable element, the desired flexibility of the hinge, the material used, etc. In one embodiment, the second layer has a thickness of from 50 angstroms to 2100 angstroms, and preferably about 900 angstroms. In another embodiment, the first layer is PECVD deposited and the second layer is LPCVD deposited.

In the embodiment shown in fig. 2A to 2E, both the first and second layers are deposited in the areas defining the movable (micromirror) elements and the pillars. Depending on the desired hardness for the micromirror element, it is also possible to deposit only either the first layer or the second layer in the area of the micromirror element. Likewise, instead of 7,8 two layers, a single layer can be used for the entire microstructure area, although this can involve a compromise between the stiffness of the plate and the flexibility of the hinge. Likewise, if a separate layer is used, the area constituting the hinge can be partially etched to reduce the thickness in this area and increase the resulting flexibility of the hinge. It is also possible to use more than two layers to create a layered (laminated) movable element, which is particularly desirable when the size of the movable element is increased, for example for switching optical beams in an optical switch. The materials for such a layer or layers may also include metal alloys and dielectric materials or compounds of metals and nitrogen, oxygen, carbon (particularly transition metals). Some of these alternative materials are disclosed in U.S. patent application No. 60/228,007, the subject matter of which is herein incorporated by reference.

As can be seen in fig. 2D, a reflective layer 9 is deposited. The reflective material may be gold, silver, titanium, aluminum or other metal, or an alloy of more than one metal, although aluminum deposited by PVD is preferred. The thickness of the metal layer can be from 50 angstroms to 2000 angstroms, preferably about 500 angstroms. An optional metal passivation layer may be added, such as a 10 to 1100 angstrom silicon oxide layer deposited by PECVD over the 9 layers. Other metal deposition techniques may also be used to deposit the metal layer 9, such as chemical liquid deposition and electroplating. After deposition of the 9 layers, the photoresist is spun and patterned, followed by etching the metal layer with a suitable metal etchant. In thatIn the case of an aluminum layer, a chlorine (or bromine) chemistry can be used (e.g., with Cl such as Ar and/or He optionally diluted, preferably inert, with₂And/or BCl₃(or Cl)₂、CCl₄、Br₂、CBr₄Etc.) plasma/RIE etching). It should be noted that the reflective layer need not be deposited last, but rather can be deposited directly on the sacrificial layer 14, between other layers forming the micromirror element, or as the only layer forming the micromirror element. However, in some approaches, since the dielectric is deposited at higher temperatures, it may be desirable to deposit the dielectric layer followed by the metal layer.

Referring to fig. 2E, the first and second layers 7,8 can be etched after the reflective layer with a known etchant or combination of etchants (depending on the metal used and the desired degree of isotropy). For example, the first and second layers can be chlorinated or fluorinated (or other halides) (e.g., with F)₂、CF₄、CHF₃、C₃F₈、CH₂F₂、C₂F₆、SF₆plasma/RIE etching, etc., or more combinations of gases like those described above, or additional gases, e.g., CF₄/H₂、SF₆/Cl₂Or using more than one etching species, e.g. CF₂Cl₂All possibly with one or more optional inert dilutants). Of course, if different species are used for the first and second layers, then a different etchant is used to etch each layer (depending on the materials used, plasma etch chemistries known in the art). If the reflective layer is deposited before the first and second layers, the etch chemistry used should be reversed. Alternatively, all layers can be etched together, depending on the materials used. Slits 20a and 20b having a width "E" shown in FIG. 2E are used to separate posts 21 from the micromirror plate body 22.

Figures 3A to 3D show the same process taken along a different cross-section (cross-section 3-3 in figure 1) and show a light-transmitting substrate 13 on which a sacrificial layer 14 is deposited. A structural layer 7 is deposited on the sacrificial layer 14. As can be seen in fig. 3B to 3C, a portion of layer 7 is removed before additional layers 8 and 9. This removed portion is in the region constituting the hinge to increase the flexibility of the hinge region. This manner of "thinning" the hinge region is proposed in U.S. provisional patent application 60/178,902 to True et al, filed on 28/1/2000 and also in U.S. patent application 09/767,632 to True et al, filed on 22/1/2001, the subject matter of each of which is incorporated herein by reference. After portions of layers 7 and 8 are removed and layer 9 is added, formation of the pattern of layers 7,8 and 9 follows as set forth above. As can be seen in FIG. 3D, the width "a" of the hinge 23 is from 0.1 μm to 10 μm, preferably about 0.7 μm. The hinges 23 are separated from each other by a gap "b" and from the adjacent micromirror plate by a gap "c", which may also have a width "a" of from 0.1 μm to 10 μm, preferably about 0.7 μm.

The processing steps generally mentioned above can be implemented in a variety of ways. For example, a glass wafer (e.g., Corning1737F, Eagle2000, quartz or sapphire wafer) can be provided and coated on its back with an opaque coating of 2000 angstroms in thickness, such as Cr, Ti, Al, TaN, polysilicon, or TiN, or other opaque coating, which is used for processing in order to make the transparent substrate temporarily opaque. Then, according to FIGS. 1-4, an optional adhesion layer is deposited (e.g., a material with silicon unsaturation, such as SiNx or SiOx, or a conductive material, such as glassy graphite, or indium tin oxide), the hydrogenated amorphous silicon sacrificial material is then subjected to a plasma enhanced chemical vapor deposition system, deposited at 5000 angstrom thickness (gas SiH4(200sccm), 1500sccm Ar, 100W power, 3.5T pressure, 380 ℃ temperature, 350mil electrode spacing, or 150sccm SiHy gas, 100sccm Ar, 55W power, 3Torr pressure, 380 ℃ temperature, 350mil electrode spacing, or 200sccm SiH4, 1500sccm Ar, 100W power, 300 ℃ temperature, 3.5T pressure, or at other processing points between these settings) on a transparent wafer, for example, using material company (Applied Materials) P5000. Alternatively, the sacrificial material can be LPCVD deposited at 560 ℃ using a wire as set forth in U.S. patent No. 5,835,256 to Huiber et al, the subject matter of which is herein incorporated by reference. Alternatively, the sacrificial material can be deposited by sputtering, or can be a non-silicon containing material that contains organic species (to be later removed, such as plasma oxidized ash). a-SiN is patterned (photoresist and chemically etched with chlorine, e.g., Cl2, BCl3, and N2) to form holes for holding the glass substrate and the micromirror plate hole together. A first silicon nitride layer for producing hardness in the micromirror and for connecting the micromirror to glass, deposited by PECVD (RF power 150W, pressure 3Torr, temperature 360 ℃, electrode spacing 570mil, gas N2/SiH4/NH3(1500/25/10), or RF power 127W, pressure 2.5Torr, temperature 380 ℃, gas N2/SiH4/NH3(1500/25/10sccm), electrode spacing 550mil, or other process parameters may be used, for example, power 175W and pressure 3.5Torr) to 900 angstroms thickness, and patterned (RF power 100 to 200W, pressure 800mT, RF power 0.8 to 1.1mm, CHF 4/3/ Ar 60 or 70/40 to 70/600 (CHF 6760 to 36800 seem), so as to remove the silicon nitride in the area where the micromirror hinge is formed. Next, a second silicon nitride layer was deposited by PECVD at a thickness of 900 angstroms (RF power 127W, pressure 2.5T, temperature 380 ℃, gas N2/SiH4/NH3(1500/25/10sccm), electrode spacing 550 mil). Then, aluminum was sputtered to the second silicon nitride layer at a thickness of 500 angstroms and at a temperature of 140 to 180 ℃, with a power of 2000W, Ar of 135 sccm. Alternatively, the metal in place of aluminum can be an aluminum alloy (Al-Si (1%), Al-Cu (0.5%) or AlSiCu or AlTi) and either implanted aluminum or target doped aluminum. Aluminum was patterned in P5000 using chlorine chemistry (pressure 40mT, power 550W, gas BCl3/Cl2/N2 50/15/30 sccm). Then, the SiN layer was etched (pressure 100mT, power 460W, gas CF4/N2(9/20sccm)), followed by ashing in H2O + O2+ N2 chemistry in plasma. The remaining structure was then ACT cleaned (acetone + DI wafer solution) and spin dried. (this cleaning may be with EKS265 photoresist residue remover of EKC technology or other cleaner-based solutions). After the resist was coated on the wafer with the microstructure thereon, the TiN on the backside was chemically etched in a plasma by BCl3/Cl2/CF4 (or other metal etchants from the CRC manual for metal etchants), either polished or ground by CMP, or cleaned using an acidic vapor such as HF followed by a second ACT clean (acetone + DI wafer solution) and a second spin dry. The wafer was separated into individual dies and each die was exposed to 300W CF4 plasma (pressure 150Torr, 85sccm for 60 seconds followed by 300 seconds of etching in He, XeF2 and N2 (etch pressure 158 Torr)). The etching was accomplished by providing the dice in a gas cell of N2 at about 400 Torr. The second region/chamber had 3.5Torr XeF2 and 38.5Torr He therein. The barrier between the two regions/plenums is removed resulting in an etching mixture of combined XeF2, He, and N2.

Alternatively, a transparent wafer (e.g., Corning1737F) was coated with TiN to a thickness of 2000 angstroms on the back side of the glass wafer. Then, in accordance with fig. 1-4, a sacrificial layer of hydrogenated amorphous silicon was deposited at a thickness of 5300 angstroms (100W power, 3.5T pressure, 300 ℃ temperature, 200sccm SiH4, 1500sccm Ar, or 2.5Torr power, 50W temperature, 360 ℃ electrode spacing, 350mil SH4 flow 200sccm Ar flow 2000sccm) on a glass wafer in P5000, applied materials, inc. The a-Si is patterned (photoresist and chemically etched with chlorine, e.g., Cl2, BCl3 and N2-500W) to form holes for securing the micro-lenses to the glass substrate. A first silicon nitride layer for production in a micromirror and for connecting the micromirror to glass is deposited by PECVD (pressure of 3Torr, 1500W, 360 ℃, gap of 570, SiH4 of 25 seem, NH3 of 10 seem, N2 of 1500 seem) to a thickness of 900 angstroms and patterned (CF4/CHF3) in order to remove silicon nitride in the area where the micromirror hinge is to be formed. Next, a second silicon nitride layer was deposited by PECVD at a thickness of 900 angstroms (same conditions as for the deposition of the first layer). Aluminum was then sputtered to the second silicon nitride layer at a thickness of 500 angstroms (150C). Alternatively, the metal in place of aluminum can be an aluminum alloy (Al-Si (1%), Al-Cu (0.5%) or AlSiCu or AlTi) and either implanted aluminum or target doped aluminum. Aluminum was patterned in P5000 using chlorine chemistry (BCl3, Cl2, N2). The SiN layer is then etched (CHF3, CF4), followed by ashing in a separate asher (O2, CH3OH at 250C). The remaining structure is then cleaned with EKS265 photoresist residue remover using EKC technology. After the resist was coated on the wafer with the microstructure thereon, the TiN on the backside was etched in a plasma with SF6/Ar, followed by a second clean and a second spin dry.

After the sacrificial and structural layers are deposited on the wafer substrate, the wafer is separated into individual dies, and each die is then placed in a Dryrec parallel plate RF plasma reactor. 100sccm of CF4 and 30sccm of O2 flowed to the plasma chamber, which was operated at about 200mtorr for 80 seconds. The dice were then etched at 143Torr etch pressure (combined XeF2, He, and N2) for 300 seconds. Etching was achieved by providing the die in a gas cell of N2 at about 400 Torr. The second region/gas cell had XeF2 with 5.5Torr and He with 20Torr in it. The barrier between the two regions/plenums was removedresulting in an etching mixture of combined XeF2, He and N2. The above steps can also be accomplished in a parallel plate plasma etcher with 300W power, CF4(150Torr, 85sccm), operating for 120 seconds. Additional characteristics of the second etch (chemical, non-plasma) are disclosed in U.S. patent application 09/427,841 to Patel et al, filed on 26.10.1999, and in U.S. patent application 09/649,569 to Patel et al, filed on 28.8.2000, the subject matter of each of which is incorporated herein by reference.

Although the hinges of each micromirror plate can be formed substantially in the same plane as the micromirror elements ( layers 7,8 and 9 for the micromirror body versus layers 8 and 9 for the micromirror hinges in fig. 3) as proposed above, they can also be constructed separately from and parallel to the micromirror elements in a different plane and as part of a separate processing step (after deposition of the second sacrificial layer material). This overlying type of hinge is disclosed in fig. 8 and 9 of the aforementioned U.S. patent 6,046,840, and in more detail in U.S. patent application 09/631,536 to Huibers et al, filed on 3.8.2000, the subject matter of which is incorporated herein by reference. Whether formed with one sacrificial layer as in the figures or two (or more) sacrificial layers as in the overlying hinges, such sacrificial layers are removed with an isotropic etchant as will be discussed below. The "release" of the micro-mirror can be done immediately following the steps described above, or can be done immediately prior to combining with the circuitry on the second substrate. If the circuitry, electrodes and micro-mirrors are not formed on the same substrate, then a second substrate is provided which comprises a large array of electrodes on a metal layer (e.g. metal 3) on top of a substrate (e.g. silicon wafer) after forming the micro-mirrors on a light transmissive substrate as proposed above. As can be seen in fig. 11A, a light transmissive substrate 40, on which an array of micro-mirrors 44 is formed as discussed above, is bonded to a second substrate 60 having circuits and electrodes at voltages Vo, Va, Vb formed on the substrate as a final layer. (a separate electrode for each micromirror can also be used in embodiments with micromirrors with a single direction of movement, such as illustrated in fig. 1.) the micromirror 44 is kept separated from the electrodes on the substrate 60 by spacers (spacers) 41 (e.g., photoresist spacers adjacent to each micromirror and/or spacers deposited within the epoxy when bonding the substrate 40 to the substrate 60). One or more electrodes on the circuit substrate electrostatically control a microdisplay pixel (pixol) (a micromirror on the overlying light transmissive transport substrate). The voltage on each electrode on the back surface determines whether the pixel of its corresponding microdisplay is optically "on" or "off". Thereby forming a visible image on the microdisplay. Details of the backside and method for producing pulse width modulated gray scale or color images are disclosed in U.S. patent No. 09/564069 to Richards, the subject matter of which is incorporated herein by reference. The combination of the first substrate and the second substrate is shown in detail in the aforementioned Llkov et al patent application, the bonding of many different types of wafers being known in the art. Such as adhesion, anodization, fusion of optical crystals, microwave solder, and thermocompression bonding.

The release of the present microlens can be a single step or a multiple step process, depending on the type of process used for the type of sacrificial material. In one embodiment of the invention, the first etch is completed with a relatively low selectivity (e.g., less than 200: 1, preferably less than 100: 1, more preferably less than 10: 1), while the subsequent second etch has a higher selectivity (e.g., greater than 100: 1, preferably greater than200: 1, more preferably greater than 1000: 1). Such double etching is further illustrated in U.S. patent application No. 60/293,032 to Patol et al, filed 5, month 22, 2001, and incorporated herein by reference. Of course, other release methods can be used, depending on the sacrificial material. For example, if the photoresist or other organic material is a sacrificial material, oxygen plasma ashing or supercritical fluid release can be used. The plasma containing pure oxygen can generate a substance that corrodes organic matter to form H2O, CO, and CO2 as products, and does not etch SiO2, Al, or Si. Or if the sacrificial material is SiO2, for example an isotropic dry etchant (CHF3+ O2, HF3 or SF6) can be used. If the sacrificial material is silicon nitride, fluorine atoms can be used to isotropically etch the silicon nitride (e.g., CF4/O2, CHF3/O2, CH2F2, or CH3F plasma). If the sacrificial material is amorphous silicon, then fluorine atoms in the form of XeF2, BrF3 or BrCl3 can be used, and if the sacrificial material is aluminum, then fluorine atoms (BCL3, CCl4, sicl4) can be used. Of course any etchant (and sacrificial material) will be selected, at least in part, according to the amount of recess desired for etching.

Another process for forming a micro-lens is shown in fig. 4A to 4J. As can be seen in fig. 4A, a substrate 30 (which can be any suitable substrate, such as a glass/quartz substrate or a semiconductor circuit substrate) has deposited thereon a sacrificial material 31. Any suitable sacrificial material can be used, preferably a material with a large etch selectivity between the material being etched and the sacrificial material. One possible sacrificial material is an organic sacrificial material such as photoresist or other organic material such as that set forth in U.S. patent application 60/298,529 to Reid et al, filed 6/15 2001. Other known microelectromechanical (MEMS) sacrificial layers, such as amorphous silicon or PSG, can be used depending on the exact composition of the structural layer. If the sacrificial material is not directly patternable, a photoresist layer 32 is added and developed to form one or more apertures (FIG. 4B). Then as seen in fig. 4C, apertures 34 are etched into the sacrificial material 31 and the photoresist 32 is removed. As can be seen in fig. 4D, a (preferably electrically conductive) layer 35 is deposited, which will eventually constitute at least 1 flexible part for the MEMS device (in this example a micro-mirror structure). The layer 35 can also form posts 36 for securing the micromirror to the substrate, or even all or part of the micromirror plate body. As will be discussed further herein, the conductive layer 35 in the preferred embodiment of the present invention comprises metal-silicon, aluminum, boron-nitrogen, with the preferred metal being a transition metal, particularly the latter transition metal. Layer 35 may also be a multi (preferably conductive) layer. Or a conductive layer between multiple other types of layers. (structured dielectric layers, reflective layers, anti-stiction layers, etc.). Layer 35 need not be electrically conductive, and layer 35 can also be insulating depending on the appropriate method, target material, and gases used in the deposition process.

Fig. 4E shows an additional portion of photoresist 37 (patterned) followed by an etch of a portion of nitride layer 35 and removal of the photoresist (fig. 4F). Then, as can be seen in fig. 4G, a layer of micro-lens structure material 38 is deposited. The material may be conductive or insulating, and may be multi-layered. If the material is a single layer, it is preferably reflective (e.g. an aluminum layer or a gold or metal alloy layer). Then, as can be seen in FIG. 4H, photoresist 39 is added and developed, followed by etching to remove portions of layer 38 (e.g., in the areas where the bending operation is to be performed). Finally, as can be seen in FIG. 4J, the sacrificial layer is removed to release the micro-electromechanical device so that the MEMS is free standing on the substrate. The circuitry formed on or in the substrate 30 (if the substrate is a circuit substrate) and the light blocking barrier on the substrate 30 to improve the automated processing of the substrate (if the substrate is a light transmissive substrate, such as glass, quartz, sapphire, etc.) are not shown in fig. 4.

As can be seen in fig. 4A to 4J, a free standing MEMS microelectromechanical structure is formed in layer 35 to form a flexible part of the MEMS device, while layer 38 forms a structure that moves due to the flexible nature of layer 35. As can be seen, layer 38 constitutes the posts and walls and movable portions supporting the MEMS structure on substrate 30. The displaceable member may be formed as a stack of layers 38 and 35 (additional layers if desired), or may be formed solely of layer 38, or even entirely of layer 35. The composition of the movable and flexible elements depends on the ultimate desired stiffness or flexibility, the ultimate desired electrical conductivity, the MEMS device being constructed, and so forth.

The micro-mirrors formed in accordance with fig. 1 to 4 are preferably constructed on a light-transmissive substrate and have a non-deflected "off" state and a deflected "on" state, however, the micro-mirrors may be constructed on the same substrate as the micro-mirror drive circuit and electrodes, and likewise the "on" and "off" states of the micro-mirrors may be in positions other than the flat non-deflected state. In the embodiments shown in fig. 5-9, the micromirror and the circuitry and electrodes for moving the micromirror are constructed on the same substrate. Also, the micro-mirror has not only deflected "on" and "off" states, but also a different deflection angle between "on" and "off". As shown in fig. 5A to 5G, the semiconductor substrate having the circuits and electrodes formed thereon may be a starting (starting) substrate formanufacturing a micromirror plate according to the present invention.

As can be seen in FIG. 5A, a semiconductor substrate 10 having circuitry for controlling the micromirror plate hasThere is a patterned metal layer, typically aluminum (e.g., the last metal layer in a semiconductor process), that forms the discrete regions 12a-12e thereon. As can be seen in fig. 5B, a sacrificial layer 14 is deposited thereon. As in the previous embodiments, the sacrificial layer material can be selected from a number of materials depending on the adjacent structure and the etchant required. In the present example, the sacrificial layer material is a novolac photoresist. As can also be seen in fig. 5B, holes 15a to 15c are formed in the sacrificial material by standard patterning for novolac photoresist in order to form the apertures 15a to 15c connected to the metal regions 12a to 12 c. As can be seen in fig. 5C, after the holes 15a to 15C are formed, the holes are formed according to a standard plug (plug) forming methodAs plugs or other connectors 16a to 16 c. For example, tungsten (W) can be deposited by CVD with the following reaction: a) silicon reduction: (this reduction is typically produced by allowing WF6 gas to contact a region of exposed solid silicon on the wafer substrate at a temperature of about 300 ℃), b) hydrogen reduction: (this process is carried out under reduced gas pressure, typically at temperatures below 450 ℃), c) silane reduction: (this reaction (LPCVD at about 300 ℃) is widely used to produce hydrogenated W core layers). Other conductive materials, particularly other refractory metals, may be used for the plugs 16a to 16 c. After deposition of the plug material layer, chemical mechanical polishing is performed up to the sacrificial layer in order to form the plug as shown in fig. 5 c. For some plug materials, it may be desirable to first deposit a layer of liner to avoid spalling (e.g., for tungsten plugs TiN, TiW or TIWN liners can be deposited in the holes of the sacrificial material to surround the tungsten and then in the sacrificial layer).

As can be seen in fig. 5D, a conductive layer is deposited and patterned so as to form discrete metal regions 18a to 18c, each of which is electrically connected to the underlying metal regions 12a to 12c, respectively, by plugs 16a to 16c, respectively. The conductive layer may be any suitable material (aluminum, aluminum alloy, or other metal alloy, conductive ceramic, etc.) that is deposited by a suitable means, such as physical vapor deposition or electroplating. The material should preferably be electrically conductive and have a suitable combination of stiffness and elasticity etc. (as will be seen, the area 18c will be formed as a hinge for the micromirror). Of course, the discrete regions 18 a-18 c need not be formed simultaneously, if different materials or properties are desired to make up one discrete region to the next. (and there are other areas that are built into the device, such as areas 12a through 12e and plugs 18a through 18 c). Of course a small number of processing steps are involved if each discrete area in a layer is material that is deposited at the same time. In a preferred embodiment, this conductive layer may be an aluminum alloy or a conductive binary or ternary higher compound such as those disclosed in U.S. patent application Ser. No. 60/228007 to Reid filed on 8/23/2000 and U.S. patent application Ser. No. 60/300533 to Reid filed on 6/22/2001. Both patent applications are incorporated herein by reference. These compounds are deposited by reactive sputtering. An appropriate etch chemistry is used to pattern the conductive layer. (e.g., chlorine chemistry for aluminum) to form the discrete conductive regions 18 a-18 c.

As further shown in fig. 5E, a second layer of sacrificial layer 20 is deposited, which may be the same as or different from the sacrificial material of layer 14 (preferably the same material so that both layers can be removed at the same time). Layer 20 is then patterned to form apertures 20a up to regions 18 c. As with the formation of holes in sacrificial layer 14, this may be accomplished with an additional layer, or layer 20 can be directly patterned if the material is a photoresist or other material that can be directly patterned. As can be seen in fig. 5F, a plug or contact 22 is formed by depositing a preferably conductive material on the sacrificial layer 20, followed by chemical mechanical polishing, such that the plug 22 is connected to the discrete region ("hinge") 18 c. Then, as can be seen in FIG. 5G, the micromirror plate body 24 is constructed by depositing a (preferably conductive) layer, followed by patterning into the desired micromirror shape. Many micromirror shapes are possible, such as that shown in FIG. 6A, and will be discussed further herein. However, the shape of the micro-lenses according to this example of the invention may have any shape, square or diamond as shown in fig. 6B and 6C. Of course, these shapes that allow for a compact micro-lens package and thus a high fill factor are preferred. (e.g., the shape of the micro-lenses of fig. 6A in a close-fitting array as shown in fig. 7). The dotted line in fig. 6C (in fig. 12 later) is the axis or rotation of the micromirror.

The layers used to manufacture the micro-lenses are shown as individual layers in accordance with fig. 5A to 5G. However, each layer (whether a structural layer or a sacrificial layer) can be provided as a stack. For example, one layer of the stack has improved mechanical properties and the other layer has improved electrical conductivity. Likewise, while the structural layers are electrically conductive in the preferred embodiment, it is possible tomake the micromirror element 24 (or one of the layers in the stack 24) and the driving electrodes 12d and 18d (and the layers/materials connecting the electrodes 12d and 18d to the semiconductor substrate) electrically conductive. Still further, the materials disclosed above (metals, metal alloys, metal-ceramic alloys, etc.) need not comprise any metal, but can comprise, for example, silicon (e.g., polysilicon) or compounds of silicon (e.g., Si3N4, SiC, SiO2, etc.). If Si3N4 is used as the structural material and amorphous silicon is used as the sacrificial material, xenon difluoride can be used as a vapor etchant to remove the sacrificial amorphous silicon. If desired, the silicon or silicide (or other compound) used as the structural material can be annealed (annealed) before and/or after removal of the sacrificial layer to improve the stress characteristics of the structural layer. Fig. 8 is an exploded perspective view of the micro-mirror constructed in accordance with fig. 5A to 5G.

One of the final steps in the manufacture of the microlens is the removal of the sacrificial layers 14 and 20. Fig. 9A is a view of the micromirror plate after removing the sacrificial layer, showing the micromirror plate 24 connected to the substrate 10 through the posts 22, the hinges 18c, the posts 16c and the metal areas 22. Since no voltage is applied to any underlying electrode (discrete metal formed in the above process)Region) such as electrode 18bOr 12d, the micromirror shown in fig. 9A is not moved or deflected. This non-deflected position is not the "off" position of the micromirror, which is typically the farthest angle away from the "on" position for the projection system (in order to achieve the best contrast ratio for the projected image). The "on" position of the micromirror, i.e., the position where the micromirror deflects light into the acceptance cone of the collection optics, is shown in fig. 9B. Voltage V_AIs applied to electrode 12d to electrostatically pull the micromirror plate 24 down until the edge of plate 24 hits electrode 12 e. The mirror plate 24 and the electrode 12e are at the same potential, in this example the voltage Vo. As shown in fig. 9C, when the voltage V is applied_BIs applied to electrode 18b and the mirror plate 24 is deflected to an opposite position, its movement being stopped by electrode 18 a. The electrode 18a and the mirror plate 24 are at the same potential. (in this example the voltage Vo). Depending on the electrode 18 b. The voltage applied to electrodes 18b and 12d need not be the same for the size of electrode 12d and the distance between these electrodes and the mirror plate 24. The deflected position shown in fig. 9C is the "off" position, and the deflected light is furthest from the collection optics.

As can be seen by comparing fig. 9B and 9C, the "off" position forms a lower angle (with the substrate) than the "on" position. Hereinafter, when referring to the on position and the off position (or such an angle or non-deflected micromirror position relative to the substrate), the sign of one angle will be used (positive or negative relative to the substrate, or non-deflected position). The symbol is arbitrary, but indicates that the micromirror is rotated in the direction of the "on" position and in the opposite direction of the "off" position. The benefits of such asymmetry will be discussed in more detail below. In one example of the present invention, the on position is from 0to +30 degrees and the off position is from 0 to-30 degrees. Move to the open position, more than to the closed position: for example, the on position can be from +10 to +30 degrees (or +12 to +20 degrees or +10 to +15 degrees) and the off position can be greater than 0 and between 0 and-30 degrees (or within a smaller range, between 0 and-10 or-12 degrees, or from-1 to-10 or-11 degrees, or from-2 to-7 degrees). In another example, the micromirror is capable of rotating at least +12 degrees to an onposition and an off position at-4 to-10 degrees. Depending on the material used for the hinge, larger angles can be used, such as an on rotation from +10 to +35 degrees and an off rotation from-2 to-25 degrees (material fatigue and plastic deformation can, of course, be problematic at large angles). Regardless of the direction of rotation, it is preferred that the on and off positions are at an angle greater than 3 degrees but less than 30 degrees to 0to the substrate, it is preferred that the on position is greater than +10 degrees, and the mirror plate is rotated 1 degree (or more) more in the on direction than in the off direction.

Fig. 10AA to 10DD illustrate another method and microlens structure. Variations in materials, layers, sacrificial etches, deposition of structural layers, etc., are all in respect of the processes described above. For the method shown in fig. 10A to 10D, the substrate 40 can be a light-transmissive substrate (later connected to a second substrate having circuitry and electrodes) or a semiconductor substrate already having circuitry and electrodes thereon. In this example, which can be seen in fig. 11A to 11B, the circuit and the electrodes are formed or on separate substrates.

In fig. 10A, a sacrificial layer 42 is deposited and patterned to form holes 43, and then plugs 46 are formed as shown in fig. 10B (preferably as in the process of fig. 5A through 5B-a deposited metal, metal alloy or other conductive layer or planarized (e.g., by CMP) to form plugs). Then, as can be seen in fig. 10C, the hinge 50 is constructed by depositing a conductive material (with suitable amorphous microstructure, elasticity, hardness, density, etc.). In this example, the hinge (and/or micromirror) is a silicon nitride of early transition metal. Such as Ta-Si-N, later transition metal silicon nitrides such as Co-Si-N or metal-ceramic alloys such as titanium aluminum oxide alloys. After deposition of such material, a photoresist is deposited and patterned so as to allow etching/removal of all areas except the hinge area 50. Then, as can be seen in fig. 10D, the micro-mirror plate 44 is constructed by first protecting the hinges with photoresist, and then depositing and patterning a hinge structure layer so as to construct the micro-mirror plate 44 partially overlapping the hinges 50 and thus connected to the hinges 50. As in other embodiments, thousands or even millions of such arrays of lenses are constructed simultaneously in the array.

The substrate with the micro-mirror plate is then connected to a substrate with drive circuitry and electrodes, whether at the wafer level or the die level. In this case there should be at least two electrodes per micromirror. Each for one deflection direction, preferably with a third electrode for allowing the micromirror to stop its movement (in one direction) by hitting on a material with the same potential as the micromirror itself. A second substrate 60 having electrodes 72 and 74 for deflecting the micromirror and a pad or electrode 70 are shown in fig. 11A. In fig. 11A the micromirror is in a non-deflected position. When the voltage V is_AIs applied to electrode 72 and the micromirror 44 is deflected until it hits electrode 70 (fig. 11B). This is the open position of the micromirror, which allows light to enter the collection optics of the system. It is possible to design a gap between the substrates so that the end of the sheet 4 hits both the electrode 70 and the substrate 40. When V is_BApplied to electrode 74, the micromirror plate 44 deflects in the opposite direction until the end of the micromirror plate hits the substrate 40. This is the off position of the micromirror. Due to the position of the hinge 50 and the post 46, the angle of the micromirror in this off position is smaller than that in the on positionThe angle of (c). An array of such micro-mirrors is shown in fig. 12. And an exploded view of a micromirror plate manufactured according to the process of fig. 10A to 10D is shown in fig. 13.

Fig. 14A is a cross-sectional view of a plurality of micro-mirrors in an array in which the micro-mirrors in the "off" state are not deflected (set 100), while the micro-mirrors in the "on" state (set 102) are moved from a flat state to project light to where the light energy is seen (directly, onto a target in the overall device, through a room onto a screen, etc.). The arrangement of such a micromirror array is better shown in fig. 14B and 14C. As can be seen in fig. 14B, in the on state of the micromirror, the incident cone of light 50 is deflected to the off state of the micromirror (all the micromirrors are on in this figure), and the cone of light 52 is projected distally to the output aperture 60, and in most cases will proceed to the image system (e.g., projection lens or lens group). The light cone 54 represents the specular reflection from the transparent cover. Fig. 14C is a diagram of the micromirror plate in the off state. Where cone 52 represents the reflected light from the micromirror in this off state. The cone of incident and reflected light will shrink onto the entire array, although in these figures, for ease of illustration, the cone of light is shown as a cone of draw on a single micromirror.

The arrangement of fig. 14B and 14C has the benefit that when the microlenses are in their off-deflected state, little light can propagate through the gaps between the microlenses, causing unwanted "gap scattering". However, the diffracted light is caused by the repeating pattern of the micromirror as shown in FIG. 14C (light 61a and 61b extending out of the cone of deflected off-light 52). This unwanted light is caused by scattering or diffraction from the edges of the micromirror (edge diffraction), particularly because the incident cone of light (and such an exit cone of light) is made as large as possible to increase efficiency, e.g. diffracted light of light 61a extending outside the cone of deflected off-light can enter the output aperture 60 (e.g. collection optics) and undesirably reduce the contrast ratio.

To avoid this overlap of off-state light (including diffracted light) and on-state light that reduces the contrast ratio, the on-state light and the off-state light can be further separated from each other by deflecting the micro-mirrors for the on and off states. As can be seen in fig. 15A, if the micromirror is deflected in its off state as shown in this figure, some light will be properly deflected in the direction of the far-off state (e.g., an optical collection system) as shown by ray 116. The other light 112 will not strike the micromirror, but will be scattered at the upper surface of the lower substrate (e.g., on the lower circuitry and electrodes) and enter the optical collection system even if the adjacent micromirror is in the off state. Or as can be seen by ray 114, incident light energy strikes the micromirror, yet still causes the slit to scatter rather than be properly directed to an off angle as does ray 116. As in the arrangement of the openings shown in fig. 15B and fig. 14B. However, as shown in fig. 15C, the off state with diffraction 61a, caused by the synchronicity of the micro-lenses, is moved further away from the angle of "on" in order to result in an improved contrast ratio due to diffraction and edge diffraction. (although as mentioned above the reduction in contrast ratio is due to the gap).

The improved micromirror array will maximize the distance between the off and on cones of light (reducing the edge scatter into the acceptance cone region) and also minimize the gap between adjacent micromirrors (minimizing the gap scatter). One solution that has been tested provides a micromirror array having micromirrors that deflect in opposite directions for on and off states as in fig. 15A to 15C, and a light absorbing layer is provided under the micromirrors in order to reduce slit scattering. Unfortunately, this adds complexity to the process, or absorbs light impinging on the micromirror array assembly (into the light valve), which increases the temperature on the light valve and causes problems due to thermal expansion, increases fatigue and sagging of the micromirror structure, increases cracking of the passivation film, self-assembled monolayers and/or lubricants, and the like.

As can be seen in fig. 16A to 16C, the micro-mirrors are deflected in the on and off states and deflected at different angles are provided. As can be seen in fig. 16A, the micromirror 100 is deflected in the off state, i.e. its deflection angle is smaller than the deflection angle of the micromirror 102 in its on state (deflected in the opposite direction from the flat or non-deflected state). As can be seen in fig. 16B, the on state and some of the specular reflection 54 are unchanged (incident light 50 is projected into output aperture 60 as outgoing light 52). In fig. 16C the micromirror is in the off state in a deflected position sufficient to minimize edge scattered light 61a entering the output aperture 60, but deflected just so much that such edge scattered light is outside the cone of acceptance light in order to minimize the gap scattered light from under the micromirror due to the large off state deflection.

An additional feature of the present invention is the packaging of the device. As mentioned above, reflection off of an optically transparent substrate can cause specular reflection. As can be seen in fig. 17A, an incident cone of light 50 reflects off the micromirror plate in an open position, shown as a reflected cone of light 52. The specular light reflected from the surface of the transparent substrate 32 is shown as a cone of light 54. In manufacturing projection systems, it is desirable to increase the divergence angle of the cone in order to increase the etendue and the efficiency of the projection system. However, as can be seen in FIG. 17A, increasing the divergence angle of the cone 50 will result in an increase in the divergence angle of the cones 52 and 54, so that specularly reflected light from the cone 54 will enter the output aperture 60 even though the micromirror plate is in its off state (thus reducing the contrast ratio).

To allow for a larger divergence angle of the light cone and to avoid specular reflections into the output aperture, the light transmissive substrate 32 is placed at an angle relative to the substrate, as can be seen in fig. 17B. In many cases, the substrate 30 is a substrate on which the micro-mirrors (or other optical MEMS elements) are formed, while the substrate 32 is a light-transmissive window in the package for the optical MEMS device. The angle of the window is greater than-1 degrees (the minus sign is coincident with the direction of the angle or the direction of the micromirror). In one example, the window is in an angle from-2 to-15 degrees, or in a range from-3 to-10 degrees. In any event, the window is at an angle relative to the substrate, preferably in the same direction as the off position of the micromirror. (relative to the micromirror substrate and/or the package bottom). As can be seen in fig. 17B, when the micromirror is in the on state, there is a gap between the reflected light (light reflection cone 52) as light from the on micromirror and the specular reflected light (light cone 54). This "gap" is due to the specular reflection cone 54 being reflected at a greater distance due to the angled optically transparent substrate. This arrangement allows, as can be seen in fig. 17C, increasing the divergence angle of the incident light cone (and corresponding reflected light cone) from the open micro-mirror (cone 52) and the optically transparent substrate (cone 54). (for ease of illumination, the point of reflection of the cone is intermediate the micromirror and the transparent substrate,although the actual cone 52 reflects from the micromirror and the specularly reflected cone 4 reflects from the substrate 32). The angled light-transmissive windows as in diagrams 17B and 17C allow for greater throughput, greater system efficiency, greater optical value etendue (solid angle multiplied by area). The light valve as shown in fig. 17B and 17C can modulate one large etendue light beam and can pass more light from the light source, thus being more efficient.

One packaged device is shown in fig. 17D and 17E. As seen in fig. 17D, incident light 40 (this view is opposite the previous view) is incident on the array and reflected therefrom. As can be seen in fig. 17E, an angled transparent substrate 32 (with mask regions 34a and 34b) not only allows the divergence angle of the light cone to be increased as described above, but additionally allows the gap between the mask of window 32 and the microlens array to be minimized, which reduces light scattering and temperature build-up on the package. The angle of the light transmissive substrate with respect to the substrate is from 1-15 degrees, preferably from 2-15 degrees, or even from 3-10 degrees. As can be seen in fig. 17D to 17E, the connection lines 37 at one end of the substrate in the package (electrically connecting the substrate to the package for actuation of the micromirror or other micromechanical element) are arranged at a greater distance from the angled window than at the opposite end of the substrate. In this way, the angled window allows for the presence of the wire, and also minimizes the distance between the light transmissive window and the micromirror substrate at the end of the substrate where the wire is not present. Note that light is incident on the micromirror array from one end of the package corresponding to the position of the raised edge of the angled window and the wiring. Additional ingredients that can be present in the package are the package adhesive, molecular scavengers or other getters, a source of a static friction scavenger (e.g., chlorosilane, perfluorinated n-basic acid, cyclohexanedisilicic acid, etc.).

If the micromirror of the present invention is used for a projection display, there should be a suitable light source that illuminates the array and projects the image onto the target through the collection optics. The arrangement of the light sources and the light beams incident on the array and on each micromirror can improve the contrast ratio while minimizing the footprint of the projection system in the present invention, as can be seen in fig. 18 and 19A to 19C. As can be seen in fig. 18, the light source 114 directs a beam 116 at a 90 degree angle to the front edge 93 of the active area of the array. (the active area of the array is shown as rectangle 94 in the figure). The active area 94 typically has from 64,000 to about 2,000,000 pixel points in a generally rectangular array, such as shown in fig. 18. The active area 94 reflects light (through the on-state micromirror) through the receiving optical system 115 to the target to form a corresponding rectangular image on the target (e.g., a wall or screen). Of course, the array can be other shapes than rectangular, and produce corresponding shapes on the target (unless through a mask). Light from the light source 114 reflects off specific micro-mirrors in the array (those in the on state) and passes through an optical system 115 (simplified as two lenses for clarity). The micro-mirror in its off state directs light to the area 99 in fig. 18. Fig. 18 is a simplification of a projection system that may have additional elements such as TIR prisms, additional focusing or magnifying lenses, color wheels (color wheels) for providing color images, light pipes, etc., all of which are known in the art. Of course, if the projection system is used for maskless lithography or non-color applications, rather than color image projection applications (e.g., front or rear screen projection televisions, computer monitors, etc.), then both a color wheel and different collection optics can be used. Also, the target may not be a screen or a photoresist, but may be a retina of an observer who is directly observing the display. As can be seen in fig. 18, all the open micro-mirrors in the array direct the light together into a single collection optical system, which may be a lens or a set of lenses for directing/focusing/projecting the light onto the target.

Whether the observed image is on a computer, television or film screen, the pixel points on the image on the screen (the pixel points on each observed or projected image corresponding to the micromirror elements in the array) have at least two sides that are not parallel to the four sides forming the rectangular screen image. As can be seen in one example of the micromirror elements in fig. 19A-19E, the incident light beam does not strike the sides of any micromirror element perpendicularly. Fig. 19A is a perspective view of light impinging on an individual micromirror element, while fig. 19B is a top view and fig. 19C is a side view. The incident light beam may be 10 to 50 degrees (e.g., 20 degrees) from normal to the plane of the micromirror array.

Regardless of the angle that the incident beam makes with the plane of the micromirror, the edge without the micromirror will be perpendicular to the beam incident thereon (see fig. 19D). In a preferred embodiment, the sides of the microlenses will be arranged at an angle (131) of less than 80 degrees or preferably 55 degrees or less, more preferably 45 degrees or less, most preferably 40 degrees or less, with respect to the projection of the optical axis of an incident beam onto the plane of the microlenses. Conversely, the angle 132 should be 100 degrees or greater, preferably 125 degrees or greater, more preferably 135 degrees or greater, and most preferably 140 degrees or greater. The on-off (i.e., rotational) axis of the micromirror is marked as dotted line 103 in fig. 19D. This switch axis can be at other locations along the micromirror plate, such as line 106, depending on the type of hinge used. As can be seen in fig. 19D, the switching axis (e.g., 103 or 106) when projected onto the plane of the micromirror is perpendicular to the incident light beam 102 projected onto the plane of the micromirror. Fig. 19E is a top view as in fig. 19D, however one micromirror array is shown in fig. 19E along with the incident light beam 102 impinging on the 2-dimensional (2-D) array of micromirrors. Note that each of the micro-lenses in fig. 19E has the shape of the micro-lens shown in fig. 19A-19D. As can be seen in fig. 19E, the shape of the entire micromirror array is rectangular. Each of the four sides of the array, 117 and 120, is formed by scribing between the most marginal pixel points in the last row and column of the active area (121 and 124) (e.g., side 119 is formed by lines that extend through pixel points 123 and 122 at the corners). Although the edges 119, 117 of the active area "in front" (closest to the light source) and "behind" (furthest from the light source) can be seen in fig. 19E to be "saw tooth shaped" due to the shape of the micro-lenses in the active area, it should be remembered that as many as 3,000,000 or more micro-lenses can be in an area from 1 square centimeter to 1 square inch. Thus, unless under extreme magnification, the active area will be substantially rectangular with the sides 118 and 120 (or 117 and 119) of the active area parallel to the sides 107 and 109 of the micromirror plate in fig. 19D (the micromirror plate in fig. 19D is one micromirror element within the active area of fig. 19E); and at the same time the active areas 117 and 119 (or 118 and 1290) are not perpendicular to the anterior or posterior edges 125a-D of the micromirror plate (see fig. 19D). Fig. 19E can also be regarded as an image including a large number of projected pixels. (each projected pixel has the shape shown in fig. 19D). In accordance with the above, the sides 118 and 120 (or 117 and 119) of the projected image are thus parallel to the sides 107 and 108 of the projected pixel, and the sides 117 and 119 (or 118 and 120) of the projected image are not perpendicular to the sides 125a-d of the projected pixel.

Fig. 20 is a view of a 2-dimensional micro-mirror array (of course with much fewer pixels than in a typical active area). For ease of illustration (as in fig. 20 and fig. 21-26 and 29-32), fewer than 60 micro-mirrors/pixels are shown. Although typical displays have from 64 pixels (320 × 200) to 1920K pixels (1600 × 1200 pixels U × GA), or higher (e.g., 1920 × 1080 ═ HDTV; 2048 × 1536 ═ Q × GA). Since each pixel size in the present invention is very small, the achievable resolution is substantially unlimited. As can be seen in fig. 20, the sides of each pixel are the corresponding sides of the parallel active area. Thus, the edge of each micromirror is either perpendicular to the edge of the active area or parallel to the edge of the active area. In contrast, as shown in fig. 21, the sides of the micro-mirror plate are neither parallel nor perpendicular to the sides of the active area. As will be seen below, in other embodiments, some of the edges are neither parallel nor perpendicular to the edges of the active area, and some of the edges can be parallel to the edges of the active area (so long as the edges are parallel to a line from the incident light beam that is disposed on the plane of the micromirror).

The micro-mirror array as shown in fig. 22 achieves a high contrast ratio. However, the arrangement of the micro-mirrors as shown in fig. 23-29 simplifies the addressing scheme. More particularly, FIGS. 23-29 have the benefit of not placing pixel points on grid points that are arranged at an angle to the X-axis and Y-axis of the array. Because typical video image sources provide color data for pixels on an X-Y axis grid, the arrangement of pixels in fig. 23-29 avoids non-trivial video pre-processing in order to render an acceptable image on a display. The same arrangement of fig. 23-29 avoids the more complex layout of the rear plane of the display (with respect to fig. 13 and 14, twice as many row or column lines as pixel control units are required). The horizontal lines 80 in fig. 22 connect the top rows of micromirror elements, while the vertical lines 81A-D extend from the top rows of micromirrors (these horizontal and vertical lines correspond to the addressed rows and columns in the array). As can be seen in fig. 22, only every other micromirror is connected in this way. Thus, twice as many rows and columns are required for all of the micro-mirrors to be addressed, which results in increased complexity in addressing the array. Fig. 22 also shows support posts 83 at the corners of the micromirror plate, which are connected to hinges (not shown) below each micromirror element (the "overlying hinges" are discussed above) and to a light transmissive substrate (not shown) above the micromirror element.

In the preferred embodiment of the present invention as shown in fig. 23, an array 92 is provided. The beam 90 is directed onto the array such that the side without the micromirror is perpendicular to the incident beam. In fig. 23, the edge in front of the micromirror (relative to the incident light beam 90) makes an angle of about 135 degrees with the incident light beam 90. Preferably, this angle is greater than 100 degrees, and more preferably greater than 130 degrees. The contrast ratio is further improved if the angle between the leading edges of the incident beam is 135 degrees or more, and can even be as much as 140 degrees or more. As can be seen in fig. 23, the positioning of the micromirror elements does not cause the addressing problems discussed above in connection with fig. 22. The posts 95 are connected to hinges (not shown) below each of the micromirror elements in fig. 23. The hinges extend in a direction perpendicular to the incident beam (and parallel to the front and rear edges 91B and 91D of the active area). The hinge allows rotation of the axis of the micromirror perpendicular to the incident light beam.

Fig. 24 is a view of a similar micromirror plate as shown in fig. 23. However, in FIG. 24, the micromirror elements are "opposite" and have their "concave" portions as their front edges. Even though the micromirror in fig. 24 is opposite to the micromirror in fig. 23, there is still no side of the micromirror perpendicular to the incident light beam. Fig. 24 shows the hinge 101 disposed in the same plane as the micro-mirror plates that are connected together with the hinge. Two types of hinges are disclosed in the above-mentioned patent No. 840. Fig. 25 also shows the hinge 110 in the same plane as the micromirror array, and shows double "convex" portions 112 ("protrusions") and "concave" portions 113 ("cut-outs") on the sides of the front face of each micromirror. Each micromirror is in the shape of a concave polygon due to a concave or cut-away portion of each micromirror. Although the micro-mirror can be a convex polygon (if the side of the micro-mirror without a convex polygon is the side parallel to the front of the active area), it is preferred that the micro-mirror has a concave polygonal shape. A convex polygon is known as a polygon through which a line that does not include a side of the polygon can pass. A polygon is a concave polygon if and only if it is not a convex polygon. The concave polygonal shape may be in the form of a series of (non-rectangular) parallelograms, or have at least one concave and at least one matching convex portion (for mating with the concave portion of an adjacent micromirror plate), although any concave polygonal shape is possible. Although less preferred, the shape of the micromirror plate can also be that of a single (non-rectangular) parallelogram, as mentioned above. Although not illustrated, the mating protrusion(s) and cut-out(s) need not be comprised of straight lines (without any edge of the micromirror plate for that matter), but may be curved. In this embodiment, the protrusions and cut-outs are semi-circular, although the angled protrusions and cut-outs are illustrated as being preferred.

Fig. 26A to 26F show another embodiment of the present invention. Although the shape of the micro-mirrors is different in each figure, each micro-mirror is the same in that no edge is perpendicular to the incident light beam. Of course, when the edge of a micromirror changes direction, there is a point, however, very small, at which point the edge can be considered vertical, if only instantaneous. However, when it is stated that no edge is vertical, it means that no substantial part is vertical, or at least that no such substantial part is present at the front or rear edge of the micromirror plate. Even if the direction of the leading edge is gradually changed (or a portion of the leading edge is perpendicular to the incident light beam, such as shown in fig. 29), it is preferred that no 1/2 beyond the leading edge be perpendicular to the incident light beam, more preferably no more than 1/4, and most preferably 1/10 or less. The fewer the portions of the leading and trailing edges that are perpendicular to the incident beam, the greater the improvement in contrast ratio.

Many embodiments of the micromirror plate can be viewed as a combination of one or more parallelograms (e.g., the same polygon). As can be seen in fig. 27A, a single parallelogram is effective for reducing diffraction of light (the light beam has a direction from the bottom to the top of the page and starts out of the plane of the page) because it has no sides perpendicular to the incident light beam. Fig. 27A shows a single parallelogram, with horizontal arrows indicating the width "d" of the parallelogram, and the switching axes ofthe micro-mirrors in fig. 27A (and fig. 27B to 27F) are also in this horizontal direction. For example, the switch axis can be along the dashed line in fig. 27A. Fig. 27B and 27C show two and three parallelogram micromirror plate patterns in which each subsequent parallelogram has the same shape, size and shape as the previous one. This arrangement forms the front and rear edges of the "saw tooth" shape of the micromirror element. Fig. 27D to 27F show 2 to 4 parallelograms. However, in FIGS. 27D through 27F, each subsequent parallelogram is the micromirror plane image of the previous parallelogram, rather than the arrangement of the images of the same parallelogram constituting the "saw-tooth" side of the micromirror element. It should be noted that the parallelograms need not each have the same width, and the lines connecting the vertices of the serrations or the lines connecting the sides of the serrations need not be perpendicular to the incident light beam. The width of each parallelogram, if they are configured to be the same width, will be "d" ═ M/N where M is the total width of the micromirror plate and N is the number of parallelograms. As the number of parallelograms increases, the width "d" is decreasing (assuming that the micromirror width is constant). However, the width "d" should preferably be substantially greater than the wavelength of the incident light. In order to maintain a high contrast ratio, the number of parallelograms N (or the number of times the front edge of the micromirror changes direction) should be less than or equal to 0.5M/λ, or preferably less than or equal to 0.2M/λ, and even less than or equal to 0.1M/λ, where λ is the wavelength of the incident light. Although the number of parallelograms is from 1 to 4 in fig. 27, any number is possible, although 15 or less, and preferably 10 or less, will result in a better contrast ratio. The number of parallelograms in fig. 27 is most preferred (4 or less).

As can be seen in fig. 28, the hinges (flexures) 191, 193 are disposed in the same plane as the micromirror element 190. An incident beam 195 from an out-of-plane light source of fig. 28 strikes the edge in front of the micromirror 190, none of which is vertical. Preferably, any portion of the hinge is not perpendicular to the incident light beam so as to reduce diffraction of light in the direction of switching of the micromirror.

Likewise, it should be noted that the sides of the "straight" micro-mirror plates (e.g., sides 194, 196 of the micro-mirror plate in fig. 28), which are shown as being parallel to the sides of the active area, may have other patterns as well. The above fig. 21 is an example in which there is no edge of the micromirror perpendicular to the incident beam 85. Fig. 30 and 31 are another example in which the sides without the micro-mirrors are perpendicular or parallel to the incident beam, and there is no added addressing complexity as in fig. 22. The incident beam can be directed substantially perpendicular to any of the edges of the four active areas in fig. 30 (see arrows 1-4) and not perpendicularly incident to the edges of any of the micro-mirrors. This unique characteristic also exists in the array shown in FIG. 31. As can be seen in fig. 29, it is also possible that part of the edge of the front face of each micromirror is perpendicular to the incident light beam and part is not perpendicular to the incident light beam.

Fig. 32A to 32J show possible hinges for the micro-mirror plate of the present invention. Similar to fig. 24, fig. 32A illustrates a micromirror (when viewed as a top view in this figure) with flexures 96 extending parallel to the incident beam and the flexural flexures couple the micromirror 97 to posts 98 that secure the micromirror elements to the substrate. The incident beam can be directed onto the array in the direction of arrows 5 or 6 in fig. 32A (as can be seen from above). Of course the incident beam originates out of plane (see fig.11A to 11E). Such an incident beam is the same for fig. 32B to 32L. Fig. 32C to 32E are another embodiment of this type of hinge. Fig. 32F to 32L are views of another embodiment of the hinge and the micromirror plate, in which the hinge does not extend parallel to the incident beam or the edge of the active area in front, except in fig. 32J, and still causes the micromirror plate to rotate about the rotation axis perpendicular to the incident beam.

When the sides of the micromirror parallel to the rotation axis of the micromirror (and perpendicular to the incident light beam) are not minimized, light diffracted by the sides of such a micromirror will pass through the collection optical system even though the micromirror is in the off state, thus reducing the contrast ratio. As can be seen in fig. 33A, a "+" shaped diffraction pattern (caused by illuminating a generally square array of microlenses, such as the microlens of fig. 20 that is at a 90 degree angle to the edge in front of the array) intersects the cone of acceptance light (circles in the figure). The diffraction pattern, which can be seen in this figure as a series of dark spots (with corresponding lighter backgrounds), forms a vertical and a horizontal line and passes right under the circle of the cone of acceptance rays, which is superimposed on the diffraction pattern as a circular solid black line. Although not shown, in the on state of the micromirror, two diffraction lines will pass within the circle of the acceptance cone. Thus, as can be seen in fig. 33A, the perpendicular diffraction lines will enter the acceptance cone of the collection optics, even when the micromirror is in the off state, thus reducing the contrast ratio. Fig. 33B is a diffraction pattern resulting from illuminating an array of square micro-mirrors at a 45 degree angle. As can be seen in fig. 38B, the diffracted light passing through the acceptance cone (small black filled circles in fig. 33B) is reduced compared to fig. 33A. However, as mentioned above, although diffraction is reduced by such illumination, other problems arise.

In contrast, as can be seen in fig. 33C, the diffraction pattern of the present invention (from the micromirror plate in the off state of fig. 28) does not have diffraction lines extending through the acceptance cone of the collection optical system, or the spatial area to which light is directed when the micromirror plate is in the on state. Thus, substantially no diffracted light passes through the area through which light passes when the micromirror is in the on state. Microlens arrays that produce such a diffraction pattern are new with illumination light that is orthogonal to the active area of the array (and/or orthogonal to the columns or rows). Likewise, the micromirror structure, and thus the hinge and the light source means of the micromirror, the sides of the active area and/or the addressed rows and columns are new.

The invention has been described with reference to specific embodiments. However, those skilled in the art will appreciate that many variations exist in the description of the embodiments described herein. For example, the shape of the microlenses of the present invention can be used for microlenses in optical switches (e.g., as disclosed in U.S. patent application 09/617,149 to Huibers et al, filed 7, 17, 2000 and U.S. patent application 60/231,041 to Huibers, filed 9, 8, 2000, both of which are incorporated herein by reference) to reduce diffraction in the switch. In addition, the micro-lenses of the present invention can be manufactured in a structure and manner. Such as those disclosed in U.S. patent application 09/767,632 to T rue et al, filed on 22/1/2001, U.S. patent application 09/631,536 to hubers et al, filed on 3/8/2000, U.S. patent application 60/293,092 to Patel et al, filed on 22/5/2001, and U.S. patent application 06/637,479 to Haiber et al, filed on 11/8/2000. Also, althougha standard red/green/blue or red/green/white degree wheel can be used in a projection display incorporating the micromirror plate of the present invention. Other chromaticity wheels can be used, such as those disclosed in U.S. provisional patent application 60/267,648 to Huibers, filed on 9.2.2001, and U.S. patent application 60/266,780 to Richards et al, filed on 6.2.2001, both of which are incorporated herein by reference.

Likewise, the present invention is applicable to processes using removable (and replaceable) substrates for both individual and combined purposes, such as disclosed in U.S. provisional patent application 60/276,222 to Patel et al, filed 3/15/2001. In addition, the present invention microlenses can be driven in an array by pulse width modulation, for example as in U.S. patent application 09/564,069 to Richards filed 3.20005, the subject matter of which is incorporated herein by reference. Further, if the halide or inert gas filed U.S. patent application 09/427,841 to Patel et al and U.S. patent application 09/649,569 to Patel et al filed 8/28/2000 can be used, both patent applications are incorporated herein by reference. Or the sacrificial materials and methods for their removal can be those set forth in U.S. patent application 60/298,529 to reid et al, filed on 5.6.2001. Still other materials of construction can be used, such as the MEMS materials set forth in U.S. patent application 60/228,007 filed on 8/23/2000 and U.S. patent application 60/300,533 filed on 6/22/2001. Each of the above patents and applications is incorporated herein by reference.

Throughout this application, structures or layers are disclosed as being on (or deposited on), across or on, or adjacent to other layers or structures, and the like. It should be appreciated that this means directly or indirectly above, across, above, adjacent, etc. As it should be appreciated that a variety of intermediate layers or structures can be incorporated in the prior art including, but not limited to, sealing layers, adhesive layers, conductive layers, layers for reducing friction, and the like. In the same way, a structure such as a substrate or a layer can be a stack due to additional structures or layers. Likewise, the use of the phrases "at least one" or "one or more" (or the like) when used herein is intended to emphasize the potential plural nature of a particular structure or layer, however, this phrase should in no way imply that a particular structure or layer does not lack plural nature in this manner. In the same way, when the phrase "directly or indirectly" is used, it should in no way be limited to where such phrase is not used, but rather is meant to be interpreted in any way whatsoever. Likewise, "MESE," "micromechanical," and "microelectromechanical" are used herein interchangeably, and structures may or may not have electrical components. Finally, the phrase "means" in the phrase "means for …" is specifically set forth in the claims, and does not mean any element in the claims, but is to be construed in accordance with a particular convention regarding the phrase "means for …".

Claims (10)

1. An individual chip made from a wafer and having formed thereon an array of micro-mirrors arranged in a two-dimensional form, in which chip:

the micromirror is disposed over a circuit and electrodes used to electrostatically drive the micromirror;

the micro-mirror is quadrilateral and no side is parallel to a side of the patch;

the micro-mirror array is rectangular and has an area of from 1 square centimeter to 1 square inch;

horizontal rows of the micro-mirrors extending landing from one corner to the other in rows parallel to one side of the rectangular image, and wherein a vertical line in each row corresponding to an addressed column extends from each micro-mirror to connect to the micro-mirrors of each other row; and is

Vertical columns of the micro-mirrors extending landing from one corner to the other in columns parallel to one side of the rectangular image, and wherein a horizontal line in each column corresponding to an addressed row extends from each micro-mirror and is connected to the micro-mirror of each other column;

wherein a number of row lines and column lines are provided for addressing the micromirror, and wherein the number of row lines multiplied by the number of column lines is larger than the number of pixels.

2. A die as claimed in claim 1, wherein the number of row lines times the number of column lines is twice the number of pixels.

3. The tablet of claim 1 wherein the number of micro-lenses is from 1,920,000 to 3,145,278.

4. The tablet of claim 1 wherein the micro-lens is in UXGA format.

5. The tablet of claim 1 wherein the micro-lens is HDTV formatted.

6. A platelet as in claim 1 wherein the axis of rotation of the micromirror is parallel to the edge of the platelet.

7. The tablet of claim 1 wherein the micro-lens is square.

8. The die of claim 1, wherein the micro-lenses are disposed on a grid that is aligned at an angle to the X-axis and Y-axis of the die.

9. The die ofclaim 1 wherein the micromirror plate comprises a micromirror plate connected to the substrate by a hinge, and wherein the substrate, the micromirror plate and the hinge are disposed on different planes.

10. The die of claim 9 wherein a first gap is formed between the hinge and the mirror plate and a second gap is formed between the mirror plate and the substrate.

Images