MEMS actuator and MEMS actuator array with a plurality of MEMS actuators

Description

[0001] The present invention relates to a MEMS actuator (MEMS = microelectromechanical system) and its structure, and further to a MEMS actuator array comprising a plurality of MEMS actuators. Embodiments further relate to a MEMS actuator with digital control and uniformly stepped deflection. Embodiments further relate to a comb-drive MEMS actuator with multiple electrically isolated, fixed electrode structures per MEMS actuator.

[0002] MEMS actuators according to embodiments can be used for a wide range of applications, such as scanner mirrors, optical switches for coupling optical fibers ("optical cross-connect"), microvalves, electrical microswitches, and other applications. Another relevant application is area light modulators, in which the MEMS actuators, which are often arranged as a 2-dimensional array, move and position connected micromirrors as desired.

[0003] In the following, the technical background is discussed, with the inventors' findings and technical conclusions regarding the technical background being summarized, e.g., with reference to the cited references.

[0004] Typically, the position of a movable actuator element is controlled by an applied electrical signal. Electrostatic attraction is often used as a physical effect, but electromagnetic forces, piezoelectric forces, or thermal expansion can also be used.

[0005] Based on the type of movement that can be performed, a distinction is made between rotary / tilting actuators and translational actuators, as well as actuator types that enable both types of movement. In the latter case, the movement components can either be rigidly coupled by the type of suspension or individually adjusted using multiple control signals (e.g., piston-tip-tilt). The present inventive concept can be used for all of these types of movement. To simplify the description, however, translational actuators will be primarily discussed, especially those coupled with micromirrors for phase adjustment of reflected light.

[0006] Such actuators are often arranged in large numbers, densely packed on a carrier substrate in a plane, and the desired deflection direction is translational perpendicular to this plane or tilting around an axis that lies in this plane. If micromirrors are coupled to the actuators, this is also referred to as a micromirror array. The shape and size of the micromirrors and the required deflection are usually determined by the application and optical constraints. A very large number of mirrors (up to several million) are often preferred. To keep the size of the entire array within limits, the individual mirrors should be as small as possible. However, their size also limits the available space for the underlying memory cells of the electronic control system and for the structural design of the actuator, and thus the possible driving force.The dense packing also promotes crosstalk, so that an actuator can react in an unfavorable way to the control signal of the neighboring actuators.

[0007] In the plate actuators discussed so far, the electric field is reasonably homogeneous, and the movable actuator element can be deflected largely in the direction of the field. Alternatively, there are actuators often referred to as comb drives. These, however, are characterized by the fact that inhomogeneous boundary fields play a decisive role, and the movable part can be deflected largely perpendicular to the direction of the strongest part of the electric field.

[0008] For electrostatically controlled analog micromirror arrays with high to very high pixel counts (approximately >1000 to several million), plate actuators are commonly used [1, 2]. These are relatively simple in design and fabrication and can represent sufficient solutions.

[0009] The present inventive concept is suitable for micromechanical actuators that are electrostatically controlled and feature a restoring elastic suspension that applies a corresponding counterforce for a static equilibrium deflection. The deflection can thus be adjusted as desired within a predetermined range and is not limited by mechanical stops.

[0010] Such actuators are usually controlled by an analog voltage between one (or two) fixed electrodes and the movable actuator element, see, for example, [1, 2]. However, digital control is also possible. For this purpose, several electrodes are provided, each of which can then be selectively applied with one of only two possible address voltages.

[0011] For digital control, the electrodes are usually designed to exert varying influences on the deflection, e.g., as in [3, 4]. This can be achieved by varying the size of the electrodes or by varying their lever arms relative to a tilt axis of the actuator, or by varying the effective distance of the electrodes from the movable actuator plate. In contrast to the widely used binary actuators (e.g., DMD / DLP from Texas Instruments [5]), such 'multi-stage digital actuators' have more than two digitally controllable deflection positions, which are not defined by mechanical stops but by a balance of electrostatic forces and spring forces.

[0012] However, plate actuators exhibit the well-known pull-in effect, which, with parallel plates, renders all positions unstable and thus unusable beyond a deflection of one-third of the initial separation of the plates at 0V applied voltage [1]. In fact, the deflection characteristic near the pull-in is already so steep that in practice only a significantly smaller range of the initial separation can be used, approximately only 20%, possibly 25%. If, for example, a deflection of at least 320nm is required for phase modulation of visible light (for a 2π modulation range), the initial separation is at least 1.3µm, preferably 1.6µm. This large separation results in very small electrostatic forces (since the force decreases proportionally to the inverse of the square of the separation) and, for pixel sizes around or below 10µm, also in almost unmanageable direct crosstalk between the electrodes of a pixel and the neighboring pixel.

[0013] Electrostatic comb drives have so far been used primarily in microsystems with single or a few larger actuators (with micromirrors or other moving components such as the mass in inertial sensors), e.g., [7, 8, 9]. The bulk micromechanical manufacturing methods commonly used in this field usually have feature sizes (e.g., finger widths) of several micrometers and are not well suited for pixels that are intended to be only a few micrometers in size.

[0014] Comb drives can be designed for actuator deflection parallel to the electrode planes or fingers, or alternatively, essentially perpendicular to them. The latter is often referred to as a vertical comb drive, since the electrode planes are usually located parallel to a substrate surface as manufacturing planes. In this case, the actuator's up and down movement is not restricted by the drive combs, and very large deflections can be achieved in resonant operation [7]. Of course, the actuators can also be oriented differently depending on the manufacturing technology; the reasoning then applies accordingly. The present inventive concept optimizes such vertical comb drives for the outlined boundary conditions. KWON S ET AL: "High Aspect Ratio Micromirrors With Large Static Rotation and Piston Actuation", IEEE PHOTONICS TECHNOLOGY LETTERS, IEEE SERVICE CENTER, PISCATAWAY, NJ, US, Vol. 16, No. 8, 1.August 2004 (2004-08-01), pages 1891-1893, ISSN: 1041-1135, DOI: 10.1109 / LPT.2004.831047, discloses silicon micromirror devices with electrically isolated vertical comb drives. All comb fingers are fabricated in a single silicon layer to create isolated, independently operable vertical comb drive sets, allowing independent up or down movement and bidirectional rotation.

[0015] The known microsystems with vertical comb drives can be divided into those in which the electrodes (combs) lie in a common plane after completion and without applied voltage, and others in which the static and movable electrodes each lie in a separate plane. The latter can be achieved by manufacturing the electrodes directly in different planes, e.g. [8], by moving one part of the electrodes from their original position to a new rest position in a late step in the manufacturing process, e.g. [9], or by shortening the two electrodes on different sides by a separate etching, e.g.

[10] . In all of these cases, an actuator is created that can be deflected in an analog manner when a static address voltage is applied. In contrast, an actuator with both electrodes in one plane can only be excited resonantly and would not be suitable for the present task.

[0016] In the case of electrodes in two planes, the prior art already includes electrodes that overlap or are immersed in each other in the resting position (without applied voltage). This means that the lower edge of the upper electrode comb is lower than the upper edge of the lower comb, with the fingers each lying in the gaps of the other comb. This is chosen because, with a given design and address voltage, a comb drive only develops its full power when the combs are immersed in each other.

[0017] Vertical comb drives have the advantage that they do not exhibit pull-in in the desired working direction, meaning the deflection can even be greater than the resting distance between the electrodes. Here, too, the desired vertical force is greater the smaller the horizontal distance between the electrode edges. However, a lower limit for the horizontal electrode distance is determined by the much larger horizontal forces that the individual fingers exert on each other. In perfectly manufactured systems, all horizontal forces add up to zero, but even the smallest inaccuracies can lead to enormous horizontal net forces that can even destroy such an actuator (horizontal pull-in). This effect is all the more critical the greater the immersion depth. The latter is, of course, even greater at full deflection of the actuator than at rest.

[0018] There are also known systems in which the electrodes are positioned edge-to-edge in the resting position, or systems with a small vertical separation. The latter is usually due to manufacturing limitations, such as an etch stop layer or bonding layer between the planes from which the electrodes were manufactured.

[0019] For micromirror arrays with a large number of pixels, it is often advantageous to split the applied voltage into two components: a fixed voltage, the same for all (or a large number of) electrodes, called the bias voltage, which results in an actuator output position that differs from the voltage-free rest position. The other component is an address voltage specific to each individual electrode, which varies depending on the desired position and can be selected from two fixed voltage values ​​with digital addressing. Providing the bias voltage can generate a greater actuator force at maximum deflection when the address voltage is limited, allowing the use of stiffer springs, which is advantageous for fast switching times. On the other hand, the bias voltage increases the risk of horizontal pull-in.

[0020] For all electrostatic actuators, the actuator force is proportional to the square of the voltage for a given configuration and displacement, see also the formula below. Furthermore, the force increases further as the movable actuator element approaches the electrode, resulting in a highly nonlinear displacement characteristic, which is usually undesirable, with analog control. One way to linearize the characteristic curve of plate actuators is described in [6]. Unfortunately, this approach may be difficult to apply for very small pixels.

[0021] Even with multi-stage digital plate actuators, a non-linearly increasing force occurs in the same way when the movable actuator element approaches the electrodes: F elektrostatisch = ε 0 2 A Elektrode U 2 g − d 2

[0022] Whereby here U denotes the applied voltage, gthe distance (gap) of the movable actuator element from the considered electrode in the rest position (without voltage applied to all electrodes) and d the approach or deflection of the actuator. This effect leads to the influence of each electrode on the amount of deflection being determined by the current value of the effective plate distance ( g - d ) depends on the voltage values ​​of the other electrodes. This results in unevenly graduated actuator positions for increasing digital control values. Fig. 7 This is shown as an example for addressing with 3 bits (= values ​​from 0 to 7) for an otherwise ideal plate actuator in the deflection range 0 to 0.25 of the gap. At large deflections, only poor resolution is achieved. The last step (from address value 6 to 7) is around 3.4 times larger than the first (from 0 to 1). If one wanted to make the last step as small as the first in this example, one would need around 2 more bits in the addressing and accordingly more, more finely divided electrodes. This is particularly disadvantageous for very small actuators.

[0023] Based on the prior art presented and the technical findings derived therefrom by the inventors, the object underlying the present invention is to provide a discretely controlled, multi-stage, electrostatic MEMS actuator with which a displacement that is as uniformly graduated as possible can be achieved. Even with very small, densely packed MEMS actuators, the maximum displacement and actuator force should be as large as possible, and crosstalk between the electrodes or electrode structures of a pixel and those of neighboring pixels should be as small as possible.

[0024] This object is achieved by the MEMS actuator according to independent claim 1. Further developments according to the invention are defined in the dependent claims.

[0025] A MEMS actuator according to the invention thus comprises a substrate, a first electrode structure which is arranged stationary with respect to the substrate, wherein the first electrode structure has a plurality of partial electrode structures, each of which has an edge structure and is electrically controllable separately from one another, and a second electrode structure with an edge structure, wherein the second electrode structure is deflectably coupled to the substrate by means of a spring structure and is electrostatically deflectable by means of the first electrode structure in order to move the edge structure of the second electrode structure into one of a plurality of discrete deflection positions, wherein the edge structures of the first and second electrode structures are formed opposite one another in a plan view and the opposing sections are spaced apart from one another by a lateral distance,and wherein the individual partial electrode structures of the first electrode structure are designed to exert a different, rectified electrostatic force on the second electrode structure based on an electrical drive voltage and to deflect the second electrode structure into the respective discrete deflection position.,

[0026] A MEMS actuator according to the invention can comprise a third electrode structure that is arranged stationary with respect to the substrate, wherein the second electrode structure is arranged between the first and third electrode structures and is deflectable. The second electrode structure is electrostatically deflectable by means of the third electrode structure in order to move the edge structure of the second electrode structure into a further discrete deflection position. The first and third electrode structures are configured to exert an opposing electrostatic force on the second electrode structure upon electrical excitation.

[0027] According to another aspect, a MEMS actuator array comprises a plurality of

[0028] MEMS actuators according to the type just mentioned, and a control device for individually controlling the respective partial electrode structures of the first electrode structures and / or for individually controlling the respective third electrode structure of the plurality of MEMS actuators, wherein the second electrode structures of the MEMS actuators can be deflected into at least one discrete deflection position in each case between the minimum deflection position and the maximum deflection position in the first electrode structure based on the control voltage, and / or wherein the second electrode structures of the MEMS actuators can be deflected into at least one discrete deflection position in each case between the maximum deflection position in the first electrode structure and the maximum deflection position in the third electrode structure based on the further control voltage.

[0029] According to one aspect, the MEMS actuator comprises a control device for selectively or individually controlling at least a subset of the partial electrode structures of the first electrode structure of the MEMS actuator with a discrete voltage value of the control voltage in order to obtain, based on the selected subset of partial electrode structures, a resulting electrostatic force on the second electrode structure with a corresponding change in the discrete deflection position of the edge structure of the second electrode structure.

[0030] According to one aspect, the electrical drive voltage has a plurality of different, discrete voltage values, wherein based on the different discrete voltage values ​​of the electrical drive voltage, the edge structure of the second electrode structure can be deflected into different, discrete, vertically spaced deflection positions.

[0031] According to one aspect, the electrical drive voltage has two different, discrete voltage values ​​to provide digital control of the MEMS actuator.

[0032] According to one aspect, the first electrode structure has "n" sub-electrode structures which, at a discrete voltage value of the electrical drive voltage, exert a rectified electrostatic force on the second electrode structure that differs by a predetermined factor. Each of the "n" sub-electrode structures can be assigned to a different bit position of a bit word. The greater the electrostatic force of this sub-electrode structure on the second electrode structure upon electrical excitation, the higher the value of the respective sub-electrode structure is assigned to a bit position of the bit word. The value of the respective bit of the bit word reflects the activation state of the assigned sub-electrode structure.

[0033] According to one aspect, the partial electrode structures of the first electrode structure differ from one another with regard to their size and / or the respective lateral distance from the second electrode structure.

[0034] For example, a higher-order bit of the n-bit word can have a greater effect (= exertion of a rectified electrostatic force on the second electrode structure) than all other lower-order bits of the n-bit word combined. The greater effect can, for example, be one (1) LSB (= least significant bit).

[0035] The solution to this problem includes, for example, the use of a comb drive with several electrically separated fixed electrodes (= separately controllable sub-electrode structures of the first electrode structure) per MEMS actuator. The sub-electrode structures differ in their size, e.g., the number of fingers or edge length, and / or the horizontal distance from the movable actuator element, i.e., the deflectable second electrode structure, and thus have varying degrees of influence on the deflection.

[0036] Comb drives can be used to generate a force that is largely independent of the deflection. This ensures that the gradation is uniform if, for example, the following conditions are met: According to one aspect, the first and second electrode structures are spaced apart by the lateral distance in a plane parallel to the substrate, wherein the second electrode structure has a first immersion depth in the minimum deflection position in the first electrode structure, wherein the first immersion depth or first deflection position lies, for example, between -0.25 times (= distance), 0 times (= flush), or 0.5 times the value and 1.5 times the value of the lateral distance.

[0037] According to one aspect, in a maximum deflection position of the second electrode structure in the first electrode structure, a vertical projection between the first and second electrode structures has at least 1, 1.5 or 2 times the value of the lateral distance.

[0038] Embodiments of the inventive concept thus make it possible to realize a digitally controlled, micromechanical, electrostatic MEMS actuator with a large deflection range and small lateral dimensions, which, compared to the plate actuators commonly used in this area, shows an improved gradation of the deflections and very low crosstalk between neighboring pixels.

[0039] Preferred embodiments are explained in more detail below with reference to the accompanying drawings. They show: Fig. 1 shows a schematic perspective view of a MEMS actuator according to an embodiment; Fig. 2a-b show schematic partial cross-sectional views through a portion of the edge structure of the first electrode structure and an opposite portion of the edge structure of the second electrode structure in an initial state (minimum address value - Fig. 2a ) and a final state (maximum address value - Fig. 2b ) according to one embodiment; Fig. 3 shows a qualitative curve of the actuator force at a fixed control voltage versus the immersion depth of the second electrode structure in the first electrode structure for a comb drive according to one embodiment; Fig. 4 shows a schematic partial cross-sectional view through a section of the edge structure of the first electrode structure and an opposite section of the edge structure of the second electrode structure of the MEMS actuator with an inverted design according to another embodiment; Fig. 5 shows a schematic partial cross-sectional view through part of a MEMS actuator according to another embodiment, wherein the MEMS actuator is designed as a double-acting actuator; Fig. 6 shows a schematic plan view of a regular array of MEMS actuators according to another embodiment; and Fig.7An exemplary, simulated curve of a non-linearly stepped deflection of a digitally controlled eight-stage plate actuator according to the state of the art.

[0040] Before exemplary embodiments of the present concept are explained in more detail below with reference to the drawings, it is pointed out that identical, functionally equivalent or equivalent elements, objects, functional blocks and / or method steps in the different figures are provided with the same reference numerals, so that the description of these elements, objects, functional blocks and / or method steps shown in different exemplary embodiments is interchangeable or can be applied to one another.

[0041] Various embodiments will now be described in more detail with reference to the accompanying drawings, in which some embodiments are illustrated. In the figures, dimensions of illustrated elements, layers, and / or regions may not be drawn to scale for clarity.

[0042] It should be understood that when an element is described as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or intervening elements may be present. Conversely, when an element is described as being "directly connected" or "coupled" to another element, no intervening elements are present. Other terms used to describe the relationship between elements should be interpreted similarly (e.g., "between" versus "directly between," "adjacent" versus "directly adjacent," etc.).

[0043] To simplify the description of the various embodiments, the figures feature a Cartesian coordinate system x, y, z, where the xy plane corresponds to or is parallel to the main surface area of ​​the carrier or substrate, and the vertical direction is perpendicular to the xy plane. In the following description, the term "lateral" or horizontal means a direction in the xy plane (or parallel thereto), while the term "vertical" indicates a direction in the ±z direction (or parallel thereto).

[0044] The following examples refer to microelectromechanical systems (MEMS) designed to deflect a movable electrode structure that can be mechanically coupled to a functional element. Although the following embodiments refer to movable functional elements that comprise a mirror, in particular a micromirror, any other functional elements can also be arranged, both in the optical area, such as lenses, filters, or the like, but also in other areas, such as for establishing an electrical contact or changing a mechanical distance.

[0045] MEMS can be manufactured using semiconductor technology, with multilayer arrangements being particularly suitable for this purpose, including conductive, insulating, and semiconducting layers, which can be spaced apart by similar layers or air gaps. MEMS can, for example, be obtained using a multilayer stackup, which is reduced by selective removal of stack material, for example through an etching process, to expose MEMS structures. A silicon material, such as monocrystalline silicon, polycrystalline silicon, or a doped silicon material, can be used as the substrate. Conductivity can be created in different layers, for example to provide the functionality of an electrode. Other layers can, for example, be metallized, for example to create a reflective surface and / or an electrically conductive surface.

[0046] In the following, the Fig. 1 and 2a-b an exemplary embodiment of a MEMS actuator 10 according to an embodiment is explained in more detail.

[0047] Fig. 1 shows a schematic perspective view of the MEMS actuator 10 according to an embodiment, while the Fig. 2a-b each a schematic partial cross-sectional view along a section line AA in Fig. 1 and parallel to the xz-plane through a portion of a partial electrode structure 14-1 of the first electrode structure 14 and an opposite portion of the second electrode structure 16 in an initial state ( Fig. 2a ) and a final state ( Fig. 2b ) show.

[0048] The MEMS actuator 10 comprises a substrate 12, for example, a complete wafer or semiconductor wafer, or alternatively, a partially or completely separated section of the wafer. The substrate 12 can form a main surface area 12-A parallel to the substrate plane (parallel to the XY plane) and extend at least partially in the substrate plane. The substrate plane is, for example, parallel to a main side 12-A of a wafer (not shown in Fig. 1 ), which can be simply referred to as the top or bottom, without these terms being intended to have a restrictive effect. Since terms such as top, bottom, left, right, front, and back can be changed or interchanged based on a changed orientation of the MEMS actuator 10 in space.

[0049] The MEMS actuator 10 further comprises a first electrode structure 14, which is arranged stationary with respect to the substrate. The first electrode structure 14 has a plurality (e.g., n) of partial electrode structures 14-1, ..., 14-n, which each have an edge structure 14-0 and are electrically controllable separately from one another, ie, for example, can be supplied with an electrical control signal or a control voltage. The first electrode structure 14 with the partial electrode structures 14-1, ..., 14-n can be arranged on or directly on the main surface region 12-A of the substrate 12. According to one embodiment, the first electrode structure 14 can be arranged, for example, by means of spacer elements (not shown in Fig. 1 ) spaced from the main surface area 12-A of the substrate 12, in this regard, for example, Fig. 4 and the corresponding description.

[0050] The MEMS actuator 10 further comprises a second electrode structure 16 with an edge structure 16-0, wherein the second electrode structure 16 is deflectably coupled to the substrate 12 by means of a spring structure 18 and is electronically deflectable by means of the first electrode structure 14 in order to move or deflect the edge structure 16-0 of the second electrode structure 16 into a discrete deflection position z of a plurality of possible, discrete deflection positions z with respect to the edge structure 14-0 of the first electrode structure 14.

[0051] As in Fig. 1 As shown by way of example, the first electrode structure 14 has, for example, three (n = 3) partial electrode structures 14-1, 14-2, 14-3, wherein sections of the partial electrode structure 14-1 are arranged at the corner regions of the first electrode structure 14, wherein sections of the second partial electrode structure 14-2 are arranged laterally (in the x-direction) between sections of the first partial electrode structure 14-1, and wherein sections of the third partial electrode structure 14-3 are arranged laterally (in the y-direction) between further sections of the first partial electrode structure 14-1. As in Fig. 1 As shown by way of example, the first partial electrode structure 14-1 has four electrically interconnected sections, while the second partial electrode structure 14-2 has two electrically interconnected sections and the third partial electrode structure 14-2 has two electrically interconnected sections.

[0052] The different partial electrode structures 14-1, 14-2, 14-3 are designed to be separately controllable in order to exert a different, rectified electrostatic force through the different partial electrode structures 14-1, 14-2, 14-3 onto the second electrode structure 16 based on an electrical control voltage VS of the individual partial electrode structures 14-1, 14-2, 14-3 and to deflect the second electrode structure 16 into one of the discrete deflection positions z.

[0053] The selectively or individually controllable partial electrode structures 14-1, 14-2, 14-3 of the first electrode structure 14 differ in their geometric configuration, e.g., in their size, e.g., the number of fingers or edge length, and / or the horizontal distance from the deflectable second electrode structure 16. The selectively controllable partial electrode structures 14-1, 14-2, 14-3 of the first electrode structure 14 thus have a varying degree of influence on the deflection of the second electrode structure 16.

[0054] The Fig. 1 However, the configuration of the partial electrode structures 14-1, 14-2, 14-3 of the first electrode structure 14 shown is only to be regarded as an example, wherein, depending on the respective application, the number "n" and the geometric configuration and arrangement of the controllable partial electrode structures 14-1, ..., 14-n of the first electrode structure 14 as well as the geometric configuration and arrangement of the respective sections of the controllable partial electrode structures 14-1, ..., 14-n differ from the representation in Fig. 1 can be distinguished. The number n of partial electrode structures can, for example, be in a range of 2 - 8 (2, 3, 4, ..., 7, 8 or even more): In the MEMS actuator 10, the opposite sections of the edge structures 14-0, 16-0 of the first and second electrode structures 14, 16 are spaced from each other by a lateral or horizontal distance x 0 (in the xy plane) (see on Fig. 2a-b ). The lateral distance x 0 thus refers (in a plan view) to laterally opposite sections of the edge structures 14-0, 16-0 of the first and second electrode structures 14, 16. The individual partial electrode structures 14-1, ..., 14-n of the first electrode structure 14 are now designed to exert a different, rectified electrostatic force on the second electrode structure 16 based on an electrical drive voltage VS and to deflect the second electrode structure 16 into one of the discrete deflection positions z.

[0055] At the Fig. 1 In the illustrated embodiment of the MEMS actuator 10, the second electrode structure 16 is deflectable translationally with respect to the first electrode structure 14. As in Fig. 1 As shown by way of example, the spring structure 18 can have two posts or substrate extensions 18-1, wherein a spring element 18-2 is arranged between the two posts 18-1, ie, is clamped between the two posts 18-1 or is mounted on the same and thus forms a spring element 18-2 clamped on both sides. A connecting element 18-3 is now also arranged on the spring element 18-2 and is mechanically connected to the second electrode structure 16. The spring element 18-2 sets a spring force which, for example, counteracts the electrostatically induced deflection of the second electrode structure 16 vertically (in the -z direction) relative to the first electrode structure 14.

[0056] Embodiments thus relate to micromechanical actuators (MEMS actuators) 10, which are electrostatically controlled, for example, with a bias voltage U BIAS and have a restoring elastic suspension 18, which applies a corresponding counterforce for a static equilibrium deflection in order to return the second electrode structure 16 to the initial position or the initial state ( Fig. 2a ) bring to.

[0057] The MEMS actuator 10 can further comprise a functional element 20, which is also mechanically firmly coupled to the second electrode structure 16 by means of the connecting element 18-3. The functional element 20 can be an element whose translational and / or rotational position can be adjusted, controlled, or at least influenced by the electrostatic deflection between the first and second electrode structures 14, 16. For example, the functional element 20 can be a micromirror and / or an electrically conductive structure.

[0058] According to one embodiment, the connecting element 18-3 can be mechanically coupled to the functional element 20 at a centroid of the surface of the functional element 20, wherein the connecting element 18-3 can further be mechanically coupled to the functional element 20 at a centroid of the surface of the second electrode structure 16. This symmetrical arrangement is to be considered only as an example, and other configurations can be selected, as will be illustrated in subsequent embodiments.

[0059] As in Fig. 1 As further illustrated by way of example, the edge structure 16-0 of the second electrode structure 16 can be configured to interlock with the edge structure 14-0 of the first electrode structure 14 with respect to a plan view (and parallel to the xy plane). Thus, the edge structure 14-0 of the first electrode structure 14 can have a finger or comb structure (with finger elements 14-A, 16-A), wherein the edge structure 16-0 of the second electrode structure 16 can have a further opposing finger or comb structure.

[0060] In this case, for example, we also speak of a "comb drive," which is formed by the first and second electrode structures 14, 16. In this context, however, it should be noted that the term "comb drive" is not intended to be restrictive, since only a few finger or comb elements 14-A, 16-A can also be used for the first and / or second electrode structures 14, 16.

[0061] Furthermore, according to the present functional principle, the first and / or second electrode structures 14, 16 can also function as edge elements without finger or comb elements according to the principles described here. For example, the first electrode structure 14 can be designed in a plan view (and parallel to the xy plane) as a circumferential structure relative to the second electrode structure 16. In general, this can then also be referred to as an electrostatic edge actuator, since the actuator force is proportional to the length of the opposing edge structures 14-0, 16-0 of the first and second electrode structures 14, 16.

[0062] Therefore, to the extent that the producible feature sizes allow, a finger or comb structure can be used more effectively for the edge structure 14-0, 16-0 of the first and second electrode structures 14, 16. In Fig. 1a, according to one embodiment, the MEMS actuator 10 has a first and second electrode structure 14, 16 with only a few finger elements 14-A, 16-A, so that the resulting MEMS actuators 10 can be densely arranged in the substrate plane and can have pixel sizes of, for example, only approximately 8 to 16 times the minimum feature size. The minimum feature size or minimum feature size is generally referred to as the value of the smallest structure that can be reliably produced photolithographically.

[0063] In this context, it should be noted that the initial state (with a minimum address value, i.e., a minimum drive voltage VS) does not need to correspond to a force-free position of the second electrode structure 16 arranged on the spring structure 18, since, for example, the electrical drive voltage VS can further comprise a constant component in the form of an electrical bias voltage V BIAS. According to one exemplary embodiment, the drive voltage VS can therefore comprise a bias voltage component (a bias voltage) V Bias and an operating voltage component VB. According to a further exemplary embodiment, the drive voltage VS can only comprise the operating voltage component VB.

[0064] According to one embodiment, the MEMS actuator 10 may further comprise a control device 22 for providing the control voltage VS between the first and second electrode structures 14, 16 of the MEMS actuator 10. The control device 22 may be configured to selectively control at least a subset or all of the partial electrode structures 14-1, ..., 14-n of the first electrode structure 14 of the MEMS actuator 10 with a discrete voltage value of the control voltage VS in order to obtain, based on the selected subset of the partial electrode structures 14-1, ..., 14-n, a resultant electrostatic force on the second electrode structure 16 with a corresponding change in the discrete deflection position z of the edge structure 16-0 of the second electrode structure 16 with respect to the edge structure 14-0 of the first electrode structure 14.

[0065] According to one embodiment, the electrical drive voltage VS provided by the drive device 22 can have a plurality of different, discrete voltage values ​​(signal levels) V Si in order to provide a discrete drive of the MEMS actuator 10, wherein, based on the different discrete voltage values ​​V Si of the electrical drive voltage VS, the edge structure 16-0 of the second electrode structure 16 can be deflected into different, discrete, vertically spaced deflection positions z with respect to the edge structure 14-0 of the first electrode structure 14. For example, a plurality of different, discrete voltage values ​​V Si with i = 2, 3 or 4 (or even more) can be used. Thus, different partial electrode structures 14-1, ..., 14-n of the first electrode structure 14 can each be driven with a different discrete voltage value V Si of the electrical drive voltage VS.

[0066] According to one embodiment, the electrical drive voltage VS provided by the drive device 22 may have two different, discrete voltage values ​​V S1 , V S2 in order to provide digital control of the MEMS actuator 10.

[0067] The control device 22 can, for example, be integrated into the semiconductor material of the substrate 12 or can also be arranged externally to the substrate and be electrically connected to the partial electrode structures 14-1, ..., 14-n of the first electrode structure 14 and the second electrode structure 16.

[0068] According to one embodiment, the first electrode structure can therefore have "n" sub-electrode structures that, at a discrete voltage value of the electrical drive voltage, exert a rectified electrostatic force on the second electrode structure 16, e.g., one that differs by a predetermined factor. Thus, each of the "n" sub-electrode structures can be assigned to a different bit position of a bit word, wherein the respective sub-electrode structure is assigned to a higher-order bit position of the bit word, the greater the electrostatic force of this sub-electrode structure 14-n on the second electrode structure 16 upon electrical excitation, and wherein the value of the respective bit of the bit word reflects the activation state of the assigned sub-electrode structure. Thus, an n-bit word can be used to drive the "n" sub-electrode structures 14-1, ..., 14-n of the first electrode structure.

[0069] According to one embodiment, the partial electrode structures 14-1, ..., 14-n can be designed such that their influence on the deflection differs by a factor of 2, e.g. by a number of fingers or edge lengths that differ by a factor of 2 with a constant horizontal gap, or by the same edge lengths with horizontal gaps that are stepped by a factor of √2.

[0070] According to a further embodiment, the partial electrode structures 14-1, ..., 14-n can be graded both in edge length and in horizontal gap, whereby one of the factors (per bit) can then be selected and the other results from the requirement for dual influence. Then, the partial electrode structures 14-1, ..., 14-n can simply be assigned directly to the various bits of the address value.

[0071] These implementations and designs are to be considered as examples only and may vary depending on the application area of ​​the MEMS actuator 10.

[0072] Fig. 2a now shows a schematic partial cross-sectional view along a section line AA in Fig. 1 and parallel to the xz-plane through a portion of a partial electrode structure 14-1 of the first electrode structure 14 and an opposite portion of the second electrode structure 16 in an initial state. In the initial state, the operating voltage VB (i.e., the variable portion of the drive voltage VS) can have a minimum operating voltage value V B-MIN. For a number n = 3 of partial electrode structures 14-1, ..., 14-n of the first electrode structure, a minimum address value can thus be present that can be assigned to an n-bit word (here a 3-bit word) with the bit value "000".

[0073] Fig. 2b now shows a schematic partial cross-sectional view along a section line AA in Fig. 1 and parallel to the xz-plane through a section of a partial electrode structure 14-1 of the first electrode structure 14 and an opposite section of the second electrode structure 16 in a final state. In the final state, the operating voltage VB (i.e., the variable component of the drive voltage VS) can have a maximum operating voltage value V B-MAX. For a number n = 3 of partial electrode structures 14-1, ..., 14-n of the first electrode structure, a maximum address value can thus be present that can be assigned to an n-bit word (here a 3-bit word) with the bit value "111".

[0074] Based on the illustration in Fig. 2a-b of the sections or finger elements 14-A, 16-A of the edge structure 14-0, 16-0 of the first and second electrode structure 14, 16, both the relative positions and deflection paths of the edge structures 14-0, 16-0 of the first and second electrode structure 14, 16 of the MEMS actuator 10 to one another as well as typical dimensions of the edge structures 14-0, 16-0 of the first and second electrode structure 14, 16 will now be explained.

[0075] So in Fig. 2a-b two individual finger elements 16-A of the edge structure 16-0 of the second electrode structure 16 and a finger element 14-A of the edge structure 14-0 of one of the partial electrode structures 14-1, ..., 14-n of the first electrode structure 14, moving relative thereto. Fig. 2a-b The arrangement of the finger elements 14-A, 16-A of the edge structures 14-0, 16-0 shown can be continued periodically (at least in sections) in order to form the circumferential edge structures 14-0 and 16-0 of the first and second electrode structures 14, 16.

[0076] As in Fig. 2a-b As shown by way of example, the finger elements 14-0 of the first electrode structure 14 have a vertical thickness d 14 (in the z-direction) and a lateral width b 14 (in the xy-plane). The finger elements 16-A of the second electrode structure 16 have a vertical thickness d 16 and a lateral (horizontal) width b 16. As already explained above, Fig. 2a represents an initial state or ground state in a minimum deflection position Δz 1 (minimum immersion depth or minimum immersion depth) of the MEMS actuator 10, while Fig. 2b represents a final state in a maximum deflection position Δz 2 (= maximum immersion depth) with a remaining vertical projection Δz 3 between the first and second electrode structures 14, 16. The minimum and maximum deflection positions indicate a maximum deflection path Δz 2, wherein the discrete deflection position z is located at a vertical position (parallel to the z-direction) along the maximum deflection path Δz 2.

[0077] According to one embodiment, the MEMS actuator 10 may include a micromirror element as the functional element 20, which is coupled to the second electrode structure 16. For applications involving visible light, for example, the following dimensions of the MEMS actuator 10 may be present for a minimal structure: x 0 ≈ 200 nm, d 14 = d 16 ≈ 1000 nm, b 14 = b 16 ≈ 400 nm. These values ​​are to be considered only as examples and may vary depending on the application of the MEMS actuator 10.

[0078] Fig. 3 now shows a qualitative curve of the (relative) actuator force at a fixed control voltage VS versus the immersion depth or the discrete deflection positions z of the second electrode structure 16 with respect to the first electrode structure 14, as an example for a comb drive according to an embodiment. In the embodiment, the finger elements (= combs) 14-A, 16-A have a vertical thickness (height) d 14 , d 16 that corresponds to 5 times the value of the lateral distance (= horizontal gap) x 0. Thus, in the embodiment of Fig. 3 At point z = 0, the plane of the lower edge of the finger elements 16-A of the second electrode structure 16 coincides with the plane of the upper edge of the finger elements 46-A of the first electrode structure 14, while point z = 5 corresponds to a complete vertical overlap of the finger elements 14-A, 16-A of the first and second electrode structures 14, 16. Negative values ​​of the immersion depth correspond to a vertical distance between the first and second electrode structures 14, 16. As can be seen from Fig. 3 As can be seen, the relative actuator force is almost independent of the deflection position over a wide range (e.g. between 0.5 times the value and 2.5 or 3 times the value of the lateral distance x 0 ).

[0079] According to one embodiment, the first and second electrode structures 14, 16 or the edge structures 14-0, 16-0 are spaced apart in a plane (xy plane) parallel to the substrate 12 by the lateral distance x 0, wherein the second electrode structure 16 has a first immersion depth in the minimum deflection position with respect to the first electrode structure, wherein the first immersion depth or first deflection position Δz 1 has, for example, at least -0.25 times, 0 times or 0.5 times the value of the lateral distance x 0 or wherein the first immersion depth lies between -0.25 times, 0 times or 0.5 times the value and 1.5 times the value of the lateral distance x 0.

[0080] In Fig. 3 As an example, 0.5 times the value of the lateral distance x 0 is given for the first immersion depth Δz 1. The first immersion depth Δz 1 at minimal deflection (in the initial position) can therefore be at least half as large as the horizontal distance x 0 of the opposite sections of the effective edge structures 14-0, 16-0 of the first and second electrode structures 14, 16 for the best possible linearity of the relative actuator force.

[0081] If a lower accuracy with lower requirements for the linearity of the relative actuator force is tolerable, the first immersion depth Δz 1 at minimum deflection iW can only be greater than zero, with Δz 1 > 0. If an even lower accuracy with lower requirements for the linearity of the relative actuator force is tolerable, the immersion depth or first deflection position Δz 1 at minimum deflection iW can have a negative value, and a vertical initial distance Δz 1 of up to a quarter of the horizontal distance x 0 , with Δz 1 ≥ -1 / 4 x 0 , can be permitted. This implementation can be attractive in some cases, as it allows a relatively simple manufacturing process even for cases without bias voltage V BIAS.

[0082] According to one embodiment, in a maximum immersion or deflection position Δz 2 of the second electrode structure 16 with respect to the first electrode structure 14, a vertical projection or offset Δz 3 between the first and second electrode structures 14, 16 is at least 1 time, 1.5 times or 2 times the value of the lateral distance x 0 of the opposite sections of the effective edge structures 14-0, 16-0 of the first and second electrode structures 14, 16 in order to obtain the best possible linearity of the relative actuator force up to the maximum deflection (in the end position) Δz 2.

[0083] In Fig. 3 For example, twice the value of the lateral distance x 0 is given for the projection (= second immersion depth) Δz 3. The projection Δz 3 at maximum deflection (in the end position) can therefore be at least twice as large as the horizontal distance x 0 between the opposite sections of the effective edge structures 14-0, 16-0 of the first and second electrode structures 14, 16 for the best possible linearity of the relative actuator force.

[0084] If a lower accuracy with lower requirements on the linearity of the relative actuator force is tolerable, the second immersion depth z 2 at maximum deflection can be at least greater than 1.5 times the value of the horizontal distance x 0.

[0085] If an even lower accuracy with lower requirements on the linearity of the relative actuator force is tolerable, a projection of at least the horizontal distance x 0 may be sufficient for the second immersion depth or deflection position z 2 at maximum deflection.

[0086] According to embodiments, a digitally controlled, multi-stage, electrostatic MEMS actuator 10 can be provided, with which a deflection that is as uniformly graduated as possible can be achieved, wherein even with very small, densely packed MEMS actuators 10, the maximum deflection and the actuator force are as large as possible and crosstalk between the electrodes or electrode structures 14, 16 of a pixel and those of the neighboring pixels is as small as possible.

[0087] According to embodiments, the MEMS actuator 10 can be used with a comb drive, wherein the first electrode structure 14, which is arranged stationary with respect to the substrate, has a plurality (n) of partial electrode structures 14-1, ..., 14-n, which form an edge structure 14-0 of the first electrode structure 14 and can be electrically controlled separately from one another. The partial electrode structures 14-1, ..., 14-n differ in their size, such as the number of fingers or edge length, and / or the horizontal distance x 0 to the movable second electrode structure and thus have a different influence on the deflection, for example.

[0088] Comb drives can be used to generate a force in the range Δz 3 that is largely independent of the deflection, as shown in the previous example using Fig. 3 This ensures that the gradation is even when the Fig. 3 explained conditions for the immersion depth Δz 1 and the projection Δz 3 are observed.

[0089] This means that the finger elements (= combs) 14-A, 16-A of the first and second electrode structures have a vertical thickness (height) d 14 , d 16 which corresponds to the sum of the immersion depth Δz 1 , the linear area Δz 2 and the projection Δz 3 , with d 14 = Δz 1 + Δz 2 + Δz 3 or d 16 = Δz 1 + Δz 2 + Δz 3 . According to one embodiment, the vertical thickness (height) d 14 , d 16 of the finger elements (= combs) 14-A, 16-A of the first and second electrode structures 14, 16 can be the same, with d 14 = d 16 . The vertical thickness (height) d 14 , d 16 can also be different, for example due to manufacturing reasons, with d 14 ≠ d 16 .

[0090] In this regard, it should be noted that the immersion depth or first deflection position Δz 1 can also have a negative value at minimal deflection, e.g. with Δz 1 ≥ - 1 / 4 x 0 .

[0091] The desired initial immersion depth Δz 1 of the mechanically connected finger elements or combs 14-A, 16-A of the first and second electrode structures 14, 16 into one another can, on the one hand, be formed directly during the manufacture of the MEMS actuator 10. However, it is often easier to manufacture the finger elements 14-A, 16-A in completely separate production planes, possibly even with a (thin) separating layer in between. The required immersion depth Δz 1 of the initial position can then be achieved by applying the bias voltage V BIAS . If this is not desired, the minimum immersion depth or deflection position Δz 1 can also be achieved by other measures, e.g., by a defined stress gradient in the spring plane, or by mechanical deflection during assembly in a housing (not shown).

[0092] In this context, it is again pointed out that the term "comb drive" should be understood very generally here (especially with regard to very small pixels 10), since electrode structures 14, 16 with only a few finger elements 14-A, 16-A or even electrode structures 14, 16 without any finger elements function according to the same principles.

[0093] Thus, according to the present functional principle, the first and / or second electrode structures 14, 16 can be used as edge elements without finger or comb elements, wherein the first and second electrode structures 14, 16 function equally according to the principles described here. For example, the first electrode structure 14 can be designed in a plan view (parallel to the xy plane) as a circumferential structure to the second electrode structure 16. The partial electrode structure 14-3 for the bit "0" in Fig. 1 In this sense it is a comb drive.

[0094] In general, this can also be referred to as an electrostatic edge actuator, since the actuator force is proportional to the length of the opposing edge structures 14-1, 16-1 of the first and second electrode structures 14, 16. The provision of finger elements 14-A, 16-A can provide further advantages, as far as the producible (minimal) structure sizes allow.

[0095] The Fig. 1 and 2a-b show finger elements 14-A, 16-A, which are connected at their ends in the same plane to form combs. Alternatively, the finger elements 14-A, 16-A could be connected to each other at their top or bottom. In this case, the connecting element would also attract the opposite comb, which would again lead to a deflection-dependent force and thus negatively influence the uniformity of the deflection gradation. Thus, laterally connected combs provide good properties. If combs connected vertically are to be used, the height d 14 , d 16 of the finger elements 14-A, 16-A and the projection Δz 1 required above should be increased by at least one distance between two fingers of the same comb in order to keep the disadvantages to a minimum.

[0096] According to exemplary embodiments, the electrostatic field in comb drives is spatially confined around the electrode fingers 14-A, 16-A due to the small electrode spacing x 0. This ensures actuator forces that can be significantly greater in the same pixel area than with a plate actuator, even if only a few fingers are possible due to the manufacturing-related minimum size of the structures. This alone is a decisive advantage, as it enables stronger springs and a faster response of the actuator. Surface micromechanics allows for significantly finer structures compared to bulk micromechanics and is therefore preferred. In addition, the comb drive allows for very low crosstalk between the electrodes of the same actuator and those of the neighboring actuator.

[0097] Since the majority of the electric field is located between finger elements 14-A and 16-A, the region of high field strengths is concentrated in a small spatial area. The low field strength of the electric field in the outer space results in low crosstalk with neighboring MEMS actuators 10.

[0098] Fig. 4 a schematic partial cross-sectional view through a section of the edge structure 14-0 of the first electrode structure 14 and an opposite section of the edge structure 16-0 of the second electrode structure 16 of the MEMS actuator 10 with an inverted structure according to a further embodiment.

[0099] The MEMS actuator 10 again comprises the first electrode structure 14 with an edge structure 14-0, wherein the first electrode structure 14 with the edge structure 14-0 is arranged stationary with respect to the substrate 12. The first electrode structure 14 with the edge structure 14-0 is connected to the substrate 12 by means of spacer elements (not shown in Fig. 4 )spaced from the main surface area 12-A of the substrate 12. The spacer elements (posts) may be part of the first electrode structure 14 and / or the substrate 12.

[0100] The MEMS actuator 10 has a top-lying first electrode structure (stator) 14 with the finger elements (stator fingers) 14-A of the edge structure 14-O, wherein the second electrode structure (actuator) 16 with the finger elements (actuator fingers) 16-O of the edge structure 16-O (in the rest position) is arranged vertically between the first electrode structure 14 and the substrate 12.

[0101] The lateral distance x 0 refers (in a plan view) again to laterally opposite sections of the edge structures 14-0, 16-0 of the first and second electrode structures 14, 16, where Fig. 4 represents an initial state or ground state in a minimum deflection position Δz 1 (minimum immersion depth) of the MEMS actuator 10.

[0102] The MEMS actuator 10 may further comprise a conductive base plate (not shown in Fig. 4 ) to shield the influence of the underlying electronics (not shown). The conductive base plate can be formed as part of the substrate 12. Unlike in the above embodiments, it can be advantageous to arrange the fixed electrode structure 14 (the stator) vertically above the movable electrode structure 16, i.e., on the side facing away from the substrate 12.

[0103] Fig. 4 thus shows a further example of a MEMS actuator 10 with only a few finger elements 14-A, 16-A, which can be densely arranged in the substrate plane or parallel to the substrate plane and can have dimensions for the MEMS actuator 10 (= pixel sizes) of only about eight to sixteen times the minimum structure size.

[0104] Fig. 5 shows a schematic partial cross-sectional view through a part of a MEMS actuator 10 according to a further embodiment, wherein the MEMS actuator 10 is designed as a double-acting actuator.

[0105] According to one embodiment, the MEMS actuator 10 can have a third electrode structure 24 that is arranged stationary with respect to the substrate 12, wherein the second electrode structure 16 is arranged (e.g., symmetrically) between the first and third electrode structures 14, 24 and is deflectable. The second electrode structure 16 can be electrostatically deflected by means of the third electrode structure 24 in order to move the edge structure 16-0 of the second electrode structure 16 into a further discrete deflection position z 3 . The first and third electrode structures 14, 24 are designed to exert an opposing electrostatic force on the second electrode structure 16 upon electrical excitation.

[0106] According to one embodiment, the third electrode structure 24 may be a partial electrode structure 24 or a plurality of partial electrode structures 24-m (not shown in Fig. 5 ), each having an edge structure 24-0 and electrically controllable separately from one another. The individual partial electrode structures 24-m of the third electrode structure 24 can be configured to exert a different, rectified electrostatic force on the second electrode structure 16 upon electrical excitation.

[0107] According to one embodiment, the third electrode structure can also comprise a plurality "m" of partial electrode structures 24-m (not shown in Fig. 5 ) which, at a discrete voltage value of the electrical drive voltage VS, exert a rectified electrostatic force on the second electrode structure 16, e.g., a different, rectified electrostatic force by a predetermined factor. Thus, each of the "m" sub-electrode structures can be assigned to a different bit position of a bit word, wherein the respective sub-electrode structure is assigned to a higher-order bit position of the bit word, the greater the electrostatic force of this sub-electrode structure 14-n on the second electrode structure 16 upon electrical excitation, and wherein the value of the respective bit of the bit word reflects the activation state of the assigned sub-electrode structure. Thus, an m-bit word can be used to drive the "m" sub-electrode structures 24-1, ..., 24-n of the third electrode structure 24.

[0108] According to one embodiment, the second and third electrode structures 16, 24 can be spaced apart in a plan view (xy plane) parallel to the substrate 12 by the lateral distance x' 0 , wherein the second electrode structure 16 in the minimum deflection position in the third electrode structure 24 has a projection Δz 4 (e.g., Δz 4 = Δz 3 ), wherein the vertical projection Δz 4 between the second and third electrode structures 16, 24 has at least 1, 1.5 or 2 times the value of the lateral distance x 0 , and wherein the immersion depth or deflection position Δz 1 of the second electrode structure 16 in the first electrode structure 14 is between -0.25 times (= distance), 0 times (= flush) or 0.5 times the value and 1.5 times the value of the lateral distance x is 0.

[0109] In the maximum deflection position (not shown in Fig. 5 ), the second electrode structure 16 has the projection Δz 3 in the first electrode structure 24, wherein the second electrode structure 16 has the immersion depth Δz 1 in the third electrode structure 24.

[0110] In Fig. 5 by way of example, two individual finger elements 14-A of the edge structure 14-0 of the first electrode structure 14 and two individual finger elements 24-A of the edge structure 24-0 of the third electrode structure 24 as well as a finger element 16-A of the edge structure 16-0 of the second electrode structure 16 moving relative thereto are shown. Fig. 5 The arrangement shown of finger elements 14-A, 16-A, 24-A of the edge structures 14-0, 16-0, 24-0 can be continued periodically (at least in sections) in order to form the circumferential edge structures 14-0, 16-0, 24-0 of the first, second and third electrode structures 14, 16, 24.

[0111] Regarding the geometric arrangement of the first and second electrode structures 14, 16, reference is made to the explanations Fig. 1 and 2a-b which are equally applicable here.

[0112] The above statements regarding the MEMS actuator 10 of Fig. 5 thus make it clear that the present concept can also be advantageously used in double-acting actuators, in which the movable second electrode structure 16 is opposite fixed electrode structures 14, 24 in both directions (i.e., 'above' and 'below' with respect to the z-direction). As a result, each electrode level, i.e., the first and third electrode structures 14, 24, can possibly be designed for only a portion of the address bits of the bit word of the control signal VS, whereby the system, i.e., the MEMS actuator 10, can exhibit significantly improved properties with limited structural resolution. For example, the most significant bit can be realized by many electrode fingers 24-A of the third electrode structure 24 above the movable electrode structure (of the actuator comb) 16, while the other bits of the bit word can be realized by correspondingly smaller finger groups, i.e., the partial electrode structures 14-1, .., 14-n, of the first electrode structure 14 below the movable electrode structure 16. This can certainly offset additional manufacturing effort. In this case, the system, ie, the MEMS actuator 10, can be designed such that the immersion depth Δz 1 and the projection Δz 3 on both sides do not fall below the minimum values ​​specified above (for the initial and final positions) over the entire deflection range.

[0113] Fig. 6 shows a schematic plan view of a regular array 100 of MEMS actuators according to another embodiment. The MEMS array 100 of Fig. 6 has a plurality of symmetrically arranged MEMS actuators 10. It should be clear that the illustrated array arrangement can be implemented with all MEMS actuators 10 described above.

[0114] Fig. 6 shows, by way of example, the MEMS actuator array 100 in a px q array arrangement (p, q = positive integers), with p = 2 rows and q = 3 columns. However, the rows and columns can essentially be continued in any order to obtain the MEMS actuator array 100 with, for example, at least 10,000 MEMS actuators 10. For better visibility of the electrode structures 14, 16, the spring elements 18-2 are omitted, and the mirrors 20 are fully transparent and represented only by their (black) edge.

[0115] According to one embodiment, the MEMS actuator array 100 comprises a plurality of MEMS actuators 10 and further a control device 22 for providing a dedicated control voltage VS between the first and second electrode structures 14, 16 of the respective MEMS actuators 10. The control device 22 is designed, for example, to selectively or individually provide each individual MEMS actuator 10 or different groups of the MEMS actuators 10 with a dedicated control voltage VS.

[0116] The control device 22 can further be designed to provide a further separate control voltage V' S between the second and third electrode structures 14, 24 of the respective MEMS actuators 10.

[0117] According to one embodiment, the MEMS actuators 10 may comprise micromirror elements 20, each coupled to one of the second electrode structures 16, wherein the micromirror elements 20 are deflectable in accordance with the deflection of the associated second electrode structure 16.

[0118] According to one embodiment, the second electrode structures 16 of the MEMS actuators 10 can be deflected into at least one intermediate position z based on the drive voltage VS. According to one embodiment, the drive voltage VS and / or the further drive voltage V's have a respective address voltage VB , V' B for the MEMS actuators 10 and further a bias voltage V BIAS , V' BIAS for the MEMS actuators 10, e.g., all actuators.

[0119] According to one embodiment, the MEMS actuator array 100 further comprises a CMOS backplane as the substrate 12, wherein the CMOS backplane comprises the drive device 22 and further memory cells 23.

[0120] According to one embodiment, the MEMS actuator array 100 comprises at least 10,000 MEMS actuators 10 in a px q arrangement, with p rows and q columns. According to one embodiment, the MEMS actuators 10 of the MEMS actuator array 100 may have a pitch P of less than or equal to 20 µm.

[0121] Even if various embodiments and the wording of the description above refer to translational MEMS actuators 10, e.g. with parallel deflecting micromirrors (lowering mirrors), the present inventive concept is also suitable for other MEMS elements, especially also MEMS actuators 10 without mirrors.

[0122] Rotatory actuators can also be deflected in this way, whereby the deflection position z, which is well defined for lowering mirrors, can then be approximately replaced by the maximum deflection of the edge structure 16-0 of the second electrode structure 16, ie the deflection at the fingertips 16-A (if present).

[0123] The present invention can be very well combined with springs according to patent

[11] or even better to application

[12] .

[0124] In the following, some essential technical effects of the MEMS actuator 100 are summarized again.

[0125] The described embodiments enable the realization of a "digitally" controlled micromechanical electrostatic actuator 10 with a large deflection range Δz 2 with small lateral dimensions, which, compared to the plate actuators commonly used in this area, shows an improved gradation of the deflections and a very low crosstalk between neighboring pixels.

[0126] The present inventive concept is also suitable for MEMS actuators 10 in which more than two stages of deflection are achieved by addressing several electrodes or partial electrode structures 14-1, ..., 14-n of the first electrode structure 14 (or partial electrode structures 14-1, ..., 14-n and 24-1, ..., 24-n of the first and third electrode structures 14 and 24) independently of one another with a binary or discrete voltage value. For example, the movable electrode structure 16 (at a fixed electrical potential, bias voltage) can be opposed by four electrically separated, fixed partial electrode structures 14-1, ..., 14-4, each of which is connected to an SRAM memory cell of the address electronics 22. Each SRAM cell of the address electronics 22 can only assume one of two (or more discrete) states and apply one of two (or more discrete) voltages VS to the connected electrode.If the four partial electrode structures 14-1, ..., 14-4 generate electrostatic forces of varying strengths due to different edge lengths (e.g., number of fingers, etc.), up to 16 steps in the analog deflection range of the MEMS actuator 10 can be controlled with a purely digital (or discrete) control in the example, i.e., without a pull-in and without mechanical stops. In this case, the relatively good linearity of the characteristic curve according to the exemplary embodiments can achieve a significant improvement in the gradation of the deflection states compared to a plate actuator with multiple electrodes.

[0127] In principle, the same could also be done with one fixed and several movable electrodes.

[0128] The present inventive concept is suitable for micromechanical actuators, in particular for phase-shifting SLMs (SLM = Spatial Light Modulator, an "array" for modulating light) with very small pixels (measured by the manufacturable mechanical structure sizes or the desired deflection). Such SLMs are particularly interesting for digital holography, both for future holographic displays and for (somewhat more immediate) applications such as universal laser tweezers, wavefront modeling, and fast optical switches for fiber optic networks, where such SLMs enable the simultaneous splitting and control of the direction, divergence, and intensity of laser beams. However, their use in other devices for pattern generation or controlling light distribution also appears to be feasible. Furthermore, a wide variety of other applications in microactuators (even without micromirrors) and sensor technology are conceivable.

[0129] Although some aspects of the present disclosure have been described as features in the context of an apparatus, it is clear that such a description may also be considered a description of corresponding method features. Although some aspects have been described as features in the context of a method, it is clear that such a description may also be considered a description of corresponding features of an apparatus or the functionality of an apparatus.

[0130] In the foregoing detailed description, various features have been grouped together in examples in order to streamline the disclosure. This manner of disclosure should not be interpreted as intending that the claimed examples include more features than are expressly recited in each claim. Rather, as the following claims reflect, subject matter may lie in fewer than all of the features of a single disclosed example.

[0131] Although specific embodiments have been shown and described herein, it will be apparent to one skilled in the art that a variety of alternative and / or equivalent implementations may be substituted for the specific embodiments shown and illustrated therein without departing from the scope of the present application. This application text is intended to cover all adaptations and variations of the specific embodiments described and discussed herein. Therefore, the present application subject matter is limited only by the language of the claims. Literatur

[0132] 1. A. Gehner et al., "Micromirror arrays for wavefront correction", SPIE Vol. 4178, pp. 348-357 (2000). 2. Hubert Lakner et al.: "Design and Fabrication of Micromirror Arrays for UV-Lithography", Proc of SPIE Vol. 4561 (2001) 3. Christopher Aubuchon: 'Multi-Tilt Micromirror systems with concealed hinge structures', Patent US6900922 4. James Hall et al.: ' Adapting Texas Instruments (TI) DLP® technology to demonstrate a phase spatial light modulator', Proc. of SPIE Vol. 10932 (2019) 5. Texas Instruments Incorporated, "DMD 101 - Introduction to DMD technology", 2009 6. Peter Dürr et al.: 'Micro-actuator with extended analog deflection at low drive voltage', Proceedings of SPIE Vol. 6114 (2006) 7. Harald Schenk et al.: "A Novel Electrostatically Driven Torsional Actuator", Proc. 3rd Int. Conf. on Micro Opto Electro Mechanical Systems (1999) 8. L. Clark et al.: 'Vertical comb drive actuated deformable mirror device and method', Patent US 6,384,952 9. T.Sandner: 'Method of fabricating a micromechanical out of two-dimensional ...', Patent US 7929192 10. Veljko Milanovic et al.: 'Gimbal-less micro-electro-mechanical system ...', Patent US7295726 11. Peter Dürr et al.: 'MEMS actuator, system with a plurality of MEMS actuators and method ...', Patent DE102015200626, patent application US2017297897AA 12. Peter Dürr et al.: 'MEMS with a movable structural element and MEMS array', German patent application 102018207783.5 (not yet public).

Claims

1. MEMS actuator (10), comprising: a substrate (12), a first electrode structure (14) that is stationary with respect to the substrate (12), wherein the first electrode structure (14) comprises a plurality of partial electrode structures (14-1, ..., 14-n), each of which comprises an edge structure (14-0) and can be electrically controlled separately, and a second electrode structure (16) with an edge structure (16-0), wherein the second electrode structure (16) is deflectably coupled to the substrate (12) by means of a spring structure (18) and is electronically deflectable by means of the first electrode structure (14) to move the edge structure (16-0) of the second electrode structure (16) into one of a plurality of discrete deflection positions (z), wherein the edge structures (14-0, 16-0) of the first and second electrode structures (14, 16) are configured to be opposite to each other with respect to a top view and the opposite portions are spaced apart by a lateral distance (x0), and wherein the individual partial electrode structures (14-1, ..., 14-n) of the first electrode structure (14) are configured to apply a different, equally directed electrostatic force on the second electrode structure (16) based on an electric control voltage (VS) and to deflect the second electrode structure (16) into the respective discrete deflection position (z).

2. MEMS actuator (10) according to claim 1, further comprising: a control (22) for selectively controlling at least one subset of the partial electrode structures (14-1, ..., 14-n) of the first electrode structure (14) with a discrete voltage value of the control voltage (VS) to obtain, based on the selected subset of partial electrode structures, a resulting electrostatic force on the second electrode structure (16) with a corresponding change of the discrete deflection position of the edge structure (16-0) of the second electrode structure (16).

3. MEMS actuator (10) according to claim 1 or 2, wherein the electric control voltage (VS) comprises a plurality of different discrete voltage values (VSi), wherein, based on the different discrete voltage values (VSi) of the electric control voltage (VS), the edge structure (16-0) of the second electrode structure (16) is deflectable in different discrete, vertically spaced-apart deflection positions (z), wherein, for example, the electric control voltage (VS) comprises two different discrete voltage values to provide digital control of the MEMS actuator (10).

4. MEMS actuator (10) according to any one of the preceding claims, wherein the partial electrode structures (14-1, ..., 14-n) of the first electrode structure (14) differ with respect to their size and / or the respective lateral distance to the second electrode structure (16).

5. MEMS actuator (10) according to any one of the preceding claims, wherein the first electrode structure (14) comprises "n" partial electrode structures (14-1, ..., 14-n) that apply, at a discrete voltage value of the electric control voltage (VS), an equally directed electrostatic force differing by a predetermined factor to the second electrode structure (16), wherein, for example, each of the "n" partial electrode structures (14-1, ..., 14-n) can be allocated to a different bit position of an n-bit word, wherein the respective partial electrode structure is allocated to a bit position of the bit word of higher significance the greater the electrostatic force of this partial electrode structure on the second electrode structure (16) during electric excitation, and wherein the value of the respective bit of the bit word reflects the activation state of the allocated partial electrode structure.

6. MEMS actuator (10) according to any one of the preceding claims, wherein the first and second electrode structures (14, 16) are spaced apart by a lateral distance in a plane parallel to the substrate (12), wherein the second electrode structure (16) comprises a first immersion depth (Δz1) in the minimum deflection position in the first electrode structure (14), wherein the first deflection position (Δz1) is between -0.25 times, 0 times or 0.5 times the value and 1.5 times the value of the lateral distance (x0).

7. MEMS actuator (10) according to any one of the preceding claims, wherein, in a maximum deflection position (Δz2) of the second electrode structure (16) in the first electrode structure (14), a vertical overlap (Δz3) between the first and second electrode structures comprises at least 1 times, 1.5 times or 2 times the value of the lateral distance (x0).

8. MEMS actuator (10) according to any one of the preceding claims, wherein at least part of the edge structures (14-0) of the first electrode structure (14) comprise a finger or comb structure and wherein the edge structure (16-0) of the second electrode structure (16) comprises a further finger or comb structure.

9. MEMS actuator (10) according to any one of the preceding claims, wherein the edge structures (14-0) of the first electrode structure (14) and the edge structure (16-0) of the second electrode structure (16) are configured to interdigitate at least in some areas.

10. MEMS actuator (10) according to any one of the preceding claims, further comprising: a third electrode structure (24) that is stationary with respect to the substrate (12), wherein the second electrode structure (16) is arranged between the first and third electrode structures (14, 24) and is deflectable, wherein the second electrode structure (16) is electrostatically deflectable by means of the third electrode structure (24) to move the edge structure (16-0) of the second electrode structure (16) into a further discrete deflection position, wherein the first and third electrode structures (14, 24) are configured to apply, during electric excitation, an opposite electrostatic force on the second electrode structure (16), wherein, for example, the third electrode structure (24) comprises a plurality of partial electrode structures (24-1, ..., 24-m), each of which comprises an edge structure (24-0) and can be electrically controlled separately, and wherein the individual partial electrode structures (24-1, ..., 24-m) of the third electrode structure (24) are configured to apply, during electric excitation, a different, equally directed electrostatic force on the second electrode structure (16), or wherein, for example, the second and third electrode structures (16, 24) are spaced apart by a lateral distance in a plane parallel to the substrate (12), wherein the second electrode structure (16) comprises a second immersion depth in the minimum deflection position in the third electrode structure (24), wherein the second immersion depth is between -0.25 times, 0 times or 0.5 times the value and 1.5 times the value of the lateral distance.

11. MEMS actuator (10) according to any one of the preceding claims, wherein the second electrode structure (16) can be deflected in a translatory and / or rotatory manner with respect to the first electrode structure (14) or with respect to the first and third electrode structures (14, 24).

12. MEMS actuator array (100), comprising: a plurality of MEMS actuators (10) according to claim 10 or according to claim 11 when depending on claim 10, and a control (22) for individually controlling the respective partial electrode structures (14-1, ..., 14-n) of the first electrode structures (14) and / or for individually controlling the respective third electrode structure (24-1, ..., 24-m) of the plurality of MEMS actuators (10), wherein the second electrode structures (16) of the MEMS actuators (10) can be deflectable into at least one deflection position, each between the minimum deflection position and the maximum deflection position in the first electrode structure (14), based on the control voltage (VS), and / or wherein the second electrode structures (16) of the MEMS actuators (10) are deflectable into at least one deflection position, each between the maximum deflection position in the first electrode structure (14) and the maximum deflection position in the third electrode structure (24), based on the further control voltage (VSi).

13. MEMS actuator array (100) according to claim 12, wherein the MEMS actuators (10) comprise micromirror elements (20) that are each coupled to one of the second electrode structures (16), wherein the micromirror elements (20) are deflectable corresponding to the deflection of the second electrode structure (16).

14. MEMS actuator array (100) according to any one of claims 12 or 13, wherein the control voltage (VS) and / or the further control voltage (VSi) comprise a respective address voltage (VB) for the MEMS actuators (10) and further a bias voltage (VBIAS) for the MEMS actuators (10).

15. MEMS actuator array (100) according to any one of claims 12 to 14, comprising: a CMOS backplane as the substrate (12), wherein the CMOS backplane comprises the control (22) and further memory cells.