Multi-level deformable mirror device.

Description: TECHNICAL FIELD [Detailed description of the invention]

【0001】

[Industrial application field]

The present invention relates to a deformable mirror device. More specifically, the present invention relates to a deformable mirror device in which a torsion hinge is configured in a plane different from the reflective surface.

【0002】

[Conventional technology and problems]

Deformable mirror devices (DMDs) have many applications in the field of light deflection technology. In operation, the device resembles a small mirror that rotates around a fixed axis. This rotation deflects the light under the control of this rotation. Therefore, in some applications, an array of DMDs is arranged so that when individual DMDs, called pixels, are selectively rotated, patterns can be created for different purposes.

【0003】

This DMD device is typically used in darkfield projection optics and can be used, for example, in high definition television equipment (HDTV). HDTV applications require a large array of pixels to achieve the required sharpness. These large pixel arrays require high density of individual DMD pixels and high modulation efficiency to maximize brightness and minimize light scattering to maximize contrast. To.

【0004】

Light scattering is partially controlled by the size of the non-rotating (dead) surface area area illuminated by light. The reason for this dead area area is that it is used to obtain a torsion hinge that supports a rotatable DMD pixel. One configuration and structure of such torsion hinges is disclosed in US Pat. No. 4,662,746 in the name of the applicant, received May 5, 1987. This patent is incorporated herein by reference in the following.

【0005】

Therefore, in this field, there is a demand for a DMD matrix that has a combination of high density and high modulation efficiency and has a small inherent light scattering.

【0006】

Further, in this field, there is a demand for a structure that maximizes the light reflection area region for DMD pixels having a given density.

【0007】

There are further demands for DMD architectures that support HDTV applications.

【0008】

[Means for solving problems of conventional technology]

The spatial light modulator device is a torsion hinge under the surface of the rotatable mirror and is configured in one position within the substrate, eg, between the address electronics and the surface. There, the surface "dead" area area is significantly smaller. The structure can then be used to construct a DMD device with some technical advantages. The first technical advantage is that the structure has low light scattering (high contrast). The reason is that there is no light scattering caused by the torsion hinge. Another advantage of this structure is that the surface area area occupied by the rotatable mirror is large, which results in greater modulation efficiency (greater brightness). Another technical advantage of the present invention is that several different arrangements of pixel elements can be obtained. All of these pixel elements are controllable and deflectable and are on the reflective surface of a substrate having a mechanically connected rotatable support that is contained within a surface other than the reflective surface.

【0009】

Multi-layer mirror surfaces and hinges are created in multiple stages, combining a combination of spin deposition and sputter deposition with several plasma etching and photoengraving stages.

【0010】

[Example]

The present invention and its advantages can be further fully understood by the following description with reference to the accompanying drawings.

【0011】

FIG. 1 shows a conventional 45 degree 2 stable torsion beam array 10 with a 12 micron pixel pitch. Each of the pixels 100 is held so that it can rotate by a hinge 101 held on the substrate by the support 102.

【0012】

Prior art constructs should be compared to the structures shown in Figure 2. Figure 2 shows the hidden hinge array 20 and also this structure has a pitch of 12 microns. When looking at this hidden hinge architecture from above, only the square torsion beam reflecting surface 200 and the beam support post 201 of each pixel are visible, as shown in Figure 2. Comparing element 100 (FIG. 1) with element 200 (FIG. 2), it is clear that the hidden hinge provides a larger rotatable reflective surface for a given pixel size.

【0013】

The dotted line in Figure 3 shows the hinge below, the address electrode, and the landing electrode. The beam support post 201 tightly connects the beam 200 to the underlying torsion hinge 401. A detailed view of the hinges and electrodes below is shown in FIG. The beam support post 201 allows the beam 200 to rotate under the control of a hinge 401 connected to the post 406. This allows the rotatable surface (beam) 200 to rotate under the control of the electrodes supported by the post 403. The beam 200 comes into contact with the landing electrode 405 and lands. The contact body 402 extends through the substrate and contacts the underlying address electrode. The configuration and operation of this device will be described below.

【0014】

Cross-sectional views along the hinges are shown in FIGS. 5a and 5b. FIG. 5a is a cross-sectional view along the hinge, showing the hinge support post 406, the hinge 401, and the beam 200. FIG. 5b is a cross-sectional view in the direction perpendicular to the hinge 401, showing the address electrode 404 and the electrode support post 403. A protective oxide layer 501, a metal layer 2 (502), and a CMOS circuit and substrate layer 503 are also shown.

【0015】

Figure 6 shows the landing angle of beam 200-θ._LRotation to 200a and landing angle + θ_LShows a rotation of 200b to. Also shown are an address electrode 404 that controls the movement 200a, 200b and a landing electrode 405 that is located at the other end of the see-saw vibration of the beam 200. A method of controlling the rotational movement of the beam 200 is disclosed in detail in the following pending patent under the name of the improved 2-stable DMD addressing circuit and method.

【0016】

Temporarily leaving the main line here, as shown in FIG. 4, by arranging the hinge 401 between the address electrodes 404, a smaller space is available for the address electrodes. However, the effect of this is negligible with respect to address voltage requirements. The area area lost relative to the address electrode has a near zero moment arm, so the torque exerted by the address electrode is only slightly smaller.

【0017】

The hidden hinge architecture is a two-spacer process consisting of a hinge spacer and a beam spacer. The hinge spacer is a thin (~ 0.5 μm) layer. When it is removed, the resulting void allows the hinge to rotate freely. The beam spacer is thick (~ 1.5 μm) and determines the final angular rotation of the beam 200. Thin hinges and thick electrodes are made with a single plasma aluminum etch using the embedded hinge process disclosed in US Pat. No. 4,662,746 below.

【0018】

The process sequence for the hidden hinge architecture is shown in Figures 7a-7d. This process is a five-layer process (hinge spacer, hinge, electrode, beam spacer, and beam). This process should be compared to the conventional four-layer process (electrodes, spacers, hinges, beams).

【0019】

This process begins with the completed address circuit 503. The address circuit 503 has a contact opening created in the protective oxide 501 of the address circuit. Typically, this address circuit is a two metal layer / poly CMOS process. The contact opening provides access to the second level metal (metal 2) 502 adhesive pad and the metal 2 address circuit output junction.

【0020】

On the address circuit, the hinge spacer 701 is spin-deposited and patterned to create the hole 702. Hole 702 creates a hinge support post, an electrode support post, and a contact. The thickness of this spacer is typically 0.5 μm, and this spacer was cured at a temperature of 200 ° C to prevent flow and foaming during subsequent processing steps. , Positive photoresist deep UV.

【0021】

As shown in Figure 7b, the next two layers 703 and 704 are created by the so-called embedded hinge process. The aluminum alloy that creates the hinge is sputtered and deposited onto the hinge spacer. This alloy is typically 750 angstroms thick and is composed of 0.2% Ti, 1% Si and the rest Al. The masking oxide is plasma deposited and patterned into the shape of the hinge 401. This hinge oxide is then embedded by a second aluminum alloy layer 704. This second aluminum alloy layer 704 (typically 3000 angstroms thick) is for making electrodes.

【0022】

The mask oxide is then plasma deposited and patterned into the shape of the electrode 404, the electrode support post 406, and the beam contact metal 405.

【0023】

Finally, a single plasma-aluminum etching is used to create a pattern of hinges, electrodes, support posts, and beam contact metal. The electrode metal above the hinge region is removed by etching to expose the embedded hinge oxide. This oxide serves as an etching stop. When the plasma aluminum etching is complete, a region of thin hinge metal 703 and a region of thick electrode metal 704 are simultaneously formed in the pattern. The mask oxide is then removed by plasma etching.

【0024】

The beam spacer 705 is then spin-deposited onto the hinges and electrodes, as shown in Figure 7C, and patterned to create holes. This hole will create a beam support post 201. The spacer 705 determines the torsion beam angular deflection, and a typical value for its thickness is 1.5 microns, which is a positive photoresist. It is a deep UV cured at a temperature of 180 ° C to prevent flow and bubble formation in later processing stages. During this heating, the hinge spacer 701 does not deteriorate. That is because the hinge spacers were cured at a higher temperature (200 ° C).

【0025】

Finally, an aluminum alloy for creating the beam 200 (typically of which its thickness is 4000 angstroms) is sputtered onto the beam spacer 705. The masking oxide 707 is then plasma deposited and patterned into the shape of the beam. The beam is then plasma etched to create a beam and a beam support post.

【0026】

This completes the processing at the wafer stage. The masking oxide 707 on the beam 200 is left in place. The wafer is then coated with PMMA and cut to form a chip array, which is then spin washed with chlorobenzene. Finally, the chip is placed in the plasma etching chamber. In this plasma etching chamber, the masking oxide 707 is removed and the spacer layer 701 and spacer layer 705 are completely removed, with voids under the hinges and beams as shown in FIG. 7d. Is created.

【0027】

Determining the amount of cut underneath to free the hinges is made by using a modified dropout structure. The dropout structure was first disclosed in US Pat. No. 4,566,935. This U.S. patent is incorporated herein by reference. This modified dropout structure is composed of the structure of FIG. 5a of the present application, except that the hinge is not connected to any hinge support post. When the beam spacer 705 is removed, and when a sufficient amount of hinge spacer 701 is removed and the hinge is free to rotate, the beam 200 descends to one side and its end is on one electrode. Will occupy a position. Alternatively, the interrelationship of the amount of undercut between the hinge and the electrode can be achieved by using a conventional dropout structure and by removing the metal layer to observe the amount of spacers remaining.

【0028】

The hidden hinge architecture has several advantages. Some of these benefits are obvious, and some are confusing. First, as mentioned above, by placing the hinge under the beam, the area of surface area occupied by the rotatable mirror is increased, which increases the modulation efficiency and thus the display is brighter. is there. Second, the hidden hinges are shielded from the projected light and therefore cannot scatter the light into the pupil of the projection lens. Therefore, the contrast ratio increases. Third, for pixels of the same size, the hidden hinge architecture allows the torsion hinge to be longer without reducing brightness and contrast, thus increasing compliance and addressing. It is to reduce the demand for voltage.

【0029】

Fourth, since the hinge spacer flattens the underlying address circuit, a special flattening step (eg, the resist etch back flattening step of the protective oxide) is used for the final address circuit metallization. Is not necessary. Flattening the unevenness of the underlying structure is anisotropic to prevent residual metal filaments in the electrode metal and to flatten the small piles of aluminum created in metal 2. Required before target hinge / electrode etching.

【0030】

Fifth, the landing electrode is air separated from the address circuit, thus reducing the feedthrough of the high voltage reset pulse to the underlying address circuit. This coupling can cause a voltage spike at the gate of the address transistor if it is not controlled, which causes the transistor to temporarily turn on at the wrong time. This air separation also reduces the potential for dielectric breakdown through the protective oxide.

【0031】

Sixth, because the address electrodes are air separated, they have a small capacitance. Therefore, assembling the frame address DMD (US Pat. No. 4,615,595) is possible with a hidden hinge architecture. In this case, the address electrode, rather than the beam, is charge-addressed. This allows the manufacture of frame address torsion beam architectures without the need to electrically separate the beams.

【0032】

A hidden hinge for the 45 degree pixel architecture has been shown, but other pixel architectures are possible by changing the hinge architecture, as shown in Figures 8a-8d. When a hidden hinge is used for a bent beam pixel (Fig. 8C), the beam support post 201 moves vertically by a distance of λ / 4 (typically 1500 angstroms to 2000 angstroms). The 5000 angstrom hinge spacer gap allows for greater movement than this vertical movement. The bent beam pixel can be used as a phase-only modulator. The effect of hiding the hinge is to reduce the Inactive Background, which reduces the amplitude modulation component, and brings this bent beam closer to the modulator only in ideal phase.

【0033】

Although the present invention has described the present invention with reference to the specific examples, the present invention is not limited to the above description. Based on the above description, it will be immediately apparent to those skilled in the art that various modifications to the embodiment may be made and that other embodiments are possible. Therefore, all such modified examples are within the scope of the present invention.

【0034】

The following sections are further disclosed with respect to the above description. (1) To hold the base layer, the separating layer that defines the controllably rotatable area area, and the rotatable area area that has a controllably rotatable area area. The plane of the rotatable layer held by the base layer and connected to the rotatable area region to allow the rotatable area region to move within a range of movement defined with respect to the base layer. Spatial light modulator having a hinge arranged in a plane separated from.

【0035】

(2) In the first term, the device having a discrete area area for the base layer to obtain a signal for controlling the rotation of the rotatable region.

【0036】

(3) In the second term, the device in which the discrete area region has an address electrode.

【0037】

(4) In the third term, the apparatus in which the discrete area region further has an address circuit.

【0038】

(5) In the second term, the apparatus further comprising a substrate layer having a control circuit and an interconnect structure for communicating a signal from the control circuit to the discrete area region.

【0039】

(6) In paragraph 5, the device whose controllability is configured using CMOS technology.

【0040】

(7) An array of pixels in which each pixel can be individually rotated into at least two states, the array having a base structure supporting all the pixels of the array, and the base. The array is held in a plane separated from the plane so that each of the pixels defines a surface area and is separated from both the base structure and the pixels so that the pixels can be deformed with respect to the base. The array of pixels having intermediate portions arranged between them to connect the base structures to the pixels.

【0041】

(8) In paragraph 7, the device further comprising an HDTV device, wherein the array is a visible display device for a high-density television receiver.

【0042】

(9) In the seventh aspect, the apparatus having a discrete area area for supplying a signal for controlling the rotation of the pixel by the base structure.

【0043】

(10) In the ninth aspect, the apparatus in which the discrete area region has an address electrode.

【0044】

(11) In the tenth term, the apparatus in which the discrete area region further has an address circuit.

【0045】

(12) In the tenth aspect, the apparatus further comprising a substrate layer having a control circuit and an interconnect structure for communicating a signal from the control circuit to the discrete area region.

【0046】

(13) In paragraph 12, the device in which the control circuit is configured using CMOS technology.

【0047】

(14) A method of constructing an array of pixels in which each pixel can be individually rotated into at least two states, the stage of constructing a base structure for supporting all the pixels of the array, and It has a step of holding the array in a plane separated from the plane of the base, and the holding step is separated from both the base structure and the pixels, and said with respect to the base. A method of constructing the array of pixels, comprising the steps of forming an intermediate structure arranged between them to connect the base structures to the pixels so that the pixels can be deformed.

【0048】

(15) The method of claim 14, wherein the array is a display device for a high density television device.

【0049】

(16) The method of claim 15, wherein the steps constituting the base structure set a discrete area area for supplying a signal for controlling the rotation of the pixels.

【0050】

(17) The method in which the discrete area region has an address electrode in paragraph 16.

【0051】

(18) In paragraph 16, a step of setting a substrate layer having a control circuit and a step of interconnecting the discrete area region to the control circuit in order to communicate a signal from the control circuit to the discrete area region. The method further comprising.

【0052】

(19) A step of setting a substrate on which an address circuit is created and a first spacer formed in a pattern defining a first support post, and an address electrode and the first support on the first spacer. A step of setting a deformable hinge connected to a post, a step of depositing a second spacer made in a pattern defining a second support post on the deformable hinge, and a plurality supported by the second post. A method of manufacturing a deformable mirror having a step of setting the number of mirrors and a step of removing the first spacer and the second spacer.

【0053】

(20) 2 Stable deformable mirror device (DMD) pixel architecture is disclosed. In this architecture, the hinges are placed in a different layer than the torsion beam layer. As a result, the size of the pixel can be made smaller, and at the same time, a bright and high-density display device can be obtained in which a large partially active area region is maintained and used for application to high-definition television.

【0054】

Related application All pending patents below are cross-reference pending patents, and all these pending patents have been assigned to Texas Instruments. These pending patents were accepted at the same time. These pending patents are incorporated into this patent application for reference. Agent statement number TI-14568 Multi-level deformable mirror device. TI-14643 Improved 2-stable DMD addressing circuit And how. TI-14649 For integrated DMDs with control circuit boards Improved architecture and processing. TI-14715 Field updated deformable mirror device. The following are also referenced and incorporated herein by reference. TI-13173A Spatial light modulators and methods. Accepted on May 15, 1989. Serial number No. 335,049. TI-14481 Spatial light modulators and methods. Accepted on September 14, 1989. Serial number No. 408,355. US Pat. No. 4,662,746 Spatial Light Modulators and Methods. Accepted on May 5, 1987. US Pat. No. 4,566,935 Spatial Light Modulators and Methods. Accepted on January 28, 1986. US Pat. No. 4,615,595 Frame Addressed Spatial Light Modulator. Accepted on October 7, 1986.

[Simple explanation of drawings]

[Figure 1]

Array of DMD pixels with prior art architecture.

[Figure 2]

Array of DMD pixels with multi-level hinge architecture.

[Fig. 3]

Format diagram for the new architecture with the hinges, electrodes and supports below shown by dotted lines.

[Fig. 4]

Layout of the hinges, electrodes and support mechanism of the underlying layer.

[Fig. 5]

Cross-sectional view of the multilayer device taken along lines 5a-5a and 5b-5b of FIG.

[Fig. 6]

Operation diagram in two states of the beam under the control of the address electrode.

[Fig. 7]

Each stage diagram in the manufacturing process of a multi-layer device.

[Fig. 8]

Another physical layout of the multi-layer device.

[Explanation of symbols]

401 hinge