FR2838421A1 - COMPONENT COMPRISING A MATRIX OF CELLS WITH MICROELECTROMECHANICAL STRUCTURES - Google Patents

Abstract

Microelectronic component comprising a matrix of cells of microelectromechanical structures, each cell comprising a movable part connected to a fixed frame (12) by at least one connecting portion (13), said matrix also comprising means for ensuring the movement selective of the mobile part of each structure, characterized in that said displacement means comprise:. means for generating a magnetic field (B) acting on the entire matrix ;. means for selectively ensuring, for each cell, the passage of an electric current in the connecting portion (13) connecting the mobile part to the fixed frame (12), so that said connecting portion (13) undergoes Laplace forces deforms causing the moving part of the selected cell to move.

Description

1 2838421

  COMPONENT COMPRISING A MATRIX OF CELLS OF

MICROELECTROMECHANICAL STRUCTURES.

  Technical Field The invention relates to the field of the electronic industry and more particularly to the sector of microeleckomechanical components

commonly known as MEMS.

  The invention relates more precisely to a new type of actuator making it possible to control the movement of MEMS structures within a matrix of such structures. This type of actuator can be used in different types of MEMS components, including in particular but not exclusively, components operating in the field of optical communications, whether

  switches, optical attenuators or even projection systems.

  In the rest of the description, the invention is therefore described in more detail.

  specifies in its application to optical components, and more precisely to optical switches, but without this constituting a limitation of the scope of

the invention.

  Prior techniques In general, it is advantageous to gather a certain number of MEMS structures within a matrix to perform different operations.

  combined, particularly in the context of optical applications.

  Such a matrix therefore comprises a set of elementary cells, arranged according to the arrangements of rows and columns. Each cell includes a micro-electromechanical or MEMS structure which includes a part that can move relative to a fixed frame. The mobile part is connected to the fixed frame by a connecting portion which has a certain deformation capacity to allow the

  moving part to adopt different positions relative to the frame.

  The movement of the movable part is effected by an actuator which, when controlled, makes it possible to exert a force on a part of the structure, and

therefore to move the mobile part.

  Various types of actuators have already been proposed, which act on different physical principles. Thus, apart from actuations by thermal phenomena, we mainly know actuators using

  electrostatic or magnetic forces.

  Actuators of the electrostatic type generally operate by allowing the application of a potential difference between an area linked to the fixed frame and an area associated with the movable part. The different opposite parts being subjected to a potential difference, they tend to attract each other, causing

  thus the displacement of the mobile part.

  Actuators of the electrostatic type have drawbacks in certain configurations, and in particular those in which the displacements have a large amplitude. Indeed, in this case, the intensity of the electrostatic force being a function of the square of the inverse of the distance between the electrodes, there is a great variation in the intensity of this force as the displacement of the mobile part. In addition, the use of a potential difference requires that the two electrodes do not come into contact with one another, and even do not come too close, to avoid the risk of short-circuit or of breakdown. Furthermore, the value of the electrostatic forces obtained is also a function of the applied voltages, which induce constraints on the design of the structures. In other words, to obtain satisfactory forces, it is sometimes necessary to use values of

  tension which are incompatible with certain geometries of structures.

  & There are also known actuators using the magnetic force exerted by the application of a magnetic field on a magnetic area produced in Ia

mobile part of the MEMS structure.

  This type of actuator has the advantage of allowing larger displacement strokes, and also of allowing contact between the zone generating the magnetic field and the moving part. Indeed, the magnetic field is generally generated by the passage of a current in a winding located opposite the mobile part. When the actuator must be deactivated, the command

  the corresponding electrical current in a winding is interrupted.

  However, magnetic type actuators have certain drawbacks, in particular because they require the use of magnetic materials. During the manufacturing processes of MEMS structures, the use of this type of material is relatively delicate, in particular because of the great diversity of the manufacturing processes used. In general, when several actuators are used to control a matrix MEMS structure, the magnetic field generated for a cell can disturb the neighboring cell (s) due to the configuration of the field lines. This phenomenon is all the more sensitive when one wishes to obtain high strength intensities, and therefore when large magnetic fields are used. We therefore observe crosstalk phenomena which can disrupt the proper functioning of the component, or impose sizing constraints if we wish to overcome them. It has also been envisaged to combine electrostatic and magnetic phenomena, to control a structure matrix. MEMS, as described in the document "MAGNETICALLY ACTUATED MICROMIRORS FOR FIBER OPTIC Switching", published in Solid-State Sensor an Actuator Workshop, Hilton Head Island, South Carolina, June 8-ll, 1998, pages 273 to 276. More specifically, the principle described in this document consists in ensuring a global displacement of all the mobile structures under the effect of a magnetic field. The magnetic field is here advantageously used to effect relatively large amplitude displacements of all of the mobile structures. When the different mobile structures are in the active position, the position of some of them can be maintained under the effect of an electrostatic force which is applied to the only cells which must be maintained in a

determined position.

  This system has several drawbacks, and in particular that of requiring an arrangement of the various magnetic and electrostatic controls, which reduces the response time and complicates the structure of the component and of the control system. In addition, it is necessary to maintain these

  tensions, as long as one wishes the mirrors to remain in their active position.

  The invention seeks to take advantage of the advantages of magnetic forces, in particular in terms of intensity, while eliminating the problems of crosstalk: Description of the Invention The invention therefore relates to a microelectronic component including a matrix of cells of microelectronic structures . Each structure has

  a movable part which is connected to a fixed frame by at least one connecting portion.

  This matrix also includes means for ensuring the selective displacement

  of the mobile part of each structure.

  According to the invention, these means of movement include

  means for generating a magnetic field acting on the entire matrix.

  They also include means for selectively ensuring, for each cell, the passage of an electric current in the connection portion connecting the mobile part to the fixed frame. In this way, the connection portion, when it is traversed by the current and when it is exposed to the magnetic field, undergoes the Laplace forces which cause it to deform by causing the displacement of the mobile part of the

selected cell.

  Aukement dit, the invention consists in selecting the cells whose state one wishes to modify, and in making them selectively traverse by an electric current. This eleckic current, when it kavers the deformable link portion, is subjected to the magnetic field. This results in a magnetic force which is localized at the level of the deformable portion of the cell considered. This force therefore causes the

  displacement of the mobile part of the cell in question.

  The Laplace forces used being of magnetic origin, they have intensities significantly higher than the actuators equipped with electrostatic lO type. The characteristic forces appear only on the cells which are traversed by the currents themselves controlled in an appropriate manner. Although the magnetic field covers the entire matrix, there is no need for

  disruption from cell to cell, or in other words crosstalk.

  According to another characteristic of the invention, the component may also include means for imposing the direction of the electric current flowing in the connection portion, so as to control the direction of the forces undergone by said connection portions, and consequently the direction of movement of the moving part of each skucture. It is thus possible to move each movable part in a

  sense or in an aukre, and shift them from a rest position.

  In the same way, by controlling the intensity of the current, it is possible to determine the amplitude of displacement of the mobile part of each MEMS structure. According to an advantageous alternative embodiment, the connecting portions can have two positions of stable equilibrium, between which they can move under the effect of the forces characteristic of Laplace, so that the mobile part of each cell can adopt two stable positions . Aukement dit, the deformable connecting portion has a bistable character, that is to say that it has two positions of equilibrium in which it remains when it is not subjected to any external force. The transition from a position to the anchor is done when the deformable portion undergoes the forces characteristic of Laplace, with sufficient intensity. Outside these transition phases, no force is required to remain in stable equilibrium positions, which means that it is not necessary to circulate the characteristic electric current in the portions. link. In other words, the device consumes energy only to pass from one state to another, and therefore consumes no energy during the

rest of the time.

  The bistable nature of the connecting portion can be defined by a geometric shape of the connecting portion which can advantageously follow a shape

beam working in buckling.

  In practice, the component may include metallized zones, covering at least a fraction of the connection portions, and also forming conductive tracks for the flow of current to each connection portion. These metallized areas can be made in the same way as connections are

  electrics integrated on the MEMS structure.

  In practice, the magnetic field can either be continuous and applied permanently to the matrix, or even be pulsed, that is to say, be applied only during the phases in which it is desired to move the moving parts, in particular in the case where, as mentioned above, the connecting portions operate from

bistable way.

  Advantageously, in practice, the magnetic field can be generated by a single winding, itself traversed by a current, or even by

permanent magnets.

  Different configurations can be adapted with regard to the direction

  of the magnetic field and the direction of deformation of the connecting portions.

  Thus, the magnetic field can be parallel to the plane of the matrix.

  In this case, the connecting portions being perpendicular to the magnetic field and generally by the plane of the matrix, these connecting portions are subjected to forces perpendicular to the plane of the matrix. In this case, the moving parts of the MEMS structures generally move perpendicular to the

plane of the matrix.

  In another configuration, the magnetic field can be perpendicular to the plane of the matrix, so that the current which flows through the connecting portions parallel to the matrix, generates forces also parallel to the matrix. The mobile portions of each MEMS cell therefore move parallel to the plane

of the matrix.

Brief description of the fairs

  The manner of carrying out the invention as well as the advantages which flow therefrom

  will emerge clearly from the description of the embodiment which follows, given as

  nonlimiting example, in support of the appended figures in which: FIG. 1 is a top view of the active part of a component including a matrix of optical switches on which are arranged so

  different fiber optic device.

  Figure 2 is a partial top view showing some cells of the matrix for a preferred orientation of the magnetic field and the electrical circuit

set up on matoce.

  Figures 3 and 4 are summary perspective views illustrating the

  displacement of the movable part of this cell in both directions.

  Figures 5, 6 and 7 are views respectively similar to Figures 2, 3 and 4, illustrating the operation for a second orientation of the magnetic field. Figure 8 is a top view of part of the matrix showing a

  third configuration of the electrical tracks.

  Figure 9 is a cross-sectional view of the component including the

  matrix of switches, as well as the terminations of optical fibers.

  Manner of Carrying Out the Invention Figure 1 illustrates in general a component (1) forming an array of optical switches. More specifically, this component (1) comprises a matrix (2) consisting of a number of elementary cells (3) arranged in rows and columns, in alignment with channels (4) intended to receive optical fibers (5) , whose beam is then collimated by an organ of

collimation (6).

  Generally, each cell (3) has a movable part (10) accommodating in the illustrated form a reflecting structure (11) which is oriented at 45, to allow the reflection of the beams from a given line in the direction of the column corresponding. This mobile part (10) moves relative to a fixed frame (12) formed by the portions framing the different mobile parts (10). The mobile part (10) is connected by a junction zone (14)

  to a beam (13) forming the connection portion with the frame (12).

  According to the invention, and as illustrated in FIG. 2, the matrix comprises a certain number of electrical tracks or more generally of conductive zones, which make it possible to circulate an electric current inside the connecting portions (13 ). Thus, as illustrated in FIG. 2, the connection portion (13) comprises a conductive track (15), itself connected to additional tracks (16,17) allowing the flow of current to the track (15) from one of the corners of the matrix. Each of the cells has a conductive track (15,18,19) on its connecting portion from the movable part to the frame. We can plan to interpose an insulating layer between the sub stratum and the conductive tracks, at

  less in areas where several tracks run parallel.

  In addition, the component also includes a source of magnetic field which acts on the entire matrix. This field source can be a winding placed under the matrix, or as in the case corresponding to FIG. 2, a winding assembly located on the opposite side of the matrix, and supplied in the manner of a Helwholtz coil to form a field.

constant between them.

  In this case, when the current flows as illustrated in Figure 2, the connecting portion (13) is the seat of a Laplace force (= il AB), which is oriented

  as illustrated in the figures perpendicular to the plane of the substrate.

  In this way, and as illustrated in FIG. 3, this connecting portion (13) deforms and bends above the plane of the substrate, thereby causing the

rise of the mobile part (10).

  Advantageously, the position illustrated in FIG. 3 can correspond to a stable equilibrium position in which the link portion (3) works in buckling. When the current flowing in the track (15) is oriented in the opposite direction to FIG. 2, the force undergone by the connection portion (13) is also perpendicular

  in the plane of the substrate, but in the opposite direction, that is to say oriented downwards.

  In this case, as illustrated in FIG. 4, this force causes the deformation of the connecting portion (13) to cause the bending downwards, and consequently the descent of the movable part relative to the plane of the substrate. As mentioned above, this movement can be a stable position of the beam (13), so that the movable part (10) remains in this position when the electric current or the magnetic field disappears. By applying a current in the direction illustrated in FIG. 2, and in the presence of the field, the mobile part can pass

  from the position illustrated in FIG. 4 to that illustrated in FIG. 3.

  However, it should be noted that the bistability of the actuator constitutes a property

  advantageous, but which is not compulsory.

0 2838421

  The invention therefore covers variants in which the actuator is not

  bistable, but works under the effect of Laplace forces.

  The configuration illustrated in FIG. 2 has the advantage of not generating a Laplace force at the level of the portions (17) for bringing the current to the beams.

  of connection (13), since the direction of the current is collinear with that of the field.

  Other configurations can be adopted to obtain the displacements

  of the movable part, and in particular those illustrated in FIGS. 5 to 7.

  In this case, the magnetic field is perpendicular to the main plane of the substrate. It can for example be generated by a flat winding located parallel

  by the substrate, under the matrix (2), as illustrated in FIG. 9.

  lS In this case, the force exerted on the connecting portion (13) is, as illustrated in Figure 5, parallel to the plane of the substrate, and oriented towards the movable part. It can therefore move as illustrated in Figure 6 by approaching the side (21) of the housing (22) receiving the movable part (10). When the electric current changes direction there, the force F undergone by the connection portion (13) also changes direction, so that the mobile part (10) moves.

  as illustrated in Figure 7, towards the side (23) of the frame (12).

  As mentioned above, the two positions illustrated in Figures 6 and 7 can correspond in an advantageous form to stable equilibrium states of

the connecting portion (13).

  The configuration illustrated in FIG. 5 has the advantage of keeping the mobile parts (10) of the cells in the same horizontal plane, which makes it possible to increase their robustness, generally the robustness of the assembly. In addition, and as illustrated in FIG. 5, the current flowing in the different tracks not being parallel to the magnetic field, it generates Laplace forces Fc which are exerted on the frame. These forces can advantageously be used in.,

11 2838421

  mechanisms for amplifying the deformation of the connecting portion (13). The effect of these forces can be canceled or reduced when the overall structure is based on a

  frame, as shown in Figure 9.

  In this regard, it will be noted that the connecting portion illustrated in the figures follows a particularly simple and basic form of beam, and this in order to allow the understanding of the invention. However, the invention also covers variants in which the connecting portions take on more complex shapes, as soon as they deform under the effect of Laplace forces.

  when they are crossed by electric currents.

  Movements of the mobile part analogous to those illustrated in FIGS. 3 and 4, can be obtained in the architecture illustrated in FIG. 8, in which the magnetic field is parallel to the plane of the substrate, and the different tracks for bringing the current (25,26) are essentially parallel to the track (15) traversing the connecting portion (13). In this case, forces Fc are exerted on the frame in accordance with the orientations illustrated in FIG. 8. Forces can for example advantageously be used to apply a prestress on the whole of the matrix, so as to increase its rigidity in terms of shock, and

  especially for accelerations perpendicular to the plane of the substrate.

  The component according to the invention can be produced according to different technologies, using in particular silicon substrates according to different surface machining methods, or even methods known under the names SCREAM or SBM. The substrate used can also be based on SOI, and as shown in FIG. 9. In this case, the platform (8) of the mobile part (10) is produced in the upper layer of the SOI substrate, the connecting portion ( 13) is itself produced in the lower layer (9) of the SOI substrate, going below the insulating layer (27). The matrix (2) can itself be arranged on a machined substrate (30) and produced independently. This substrate (30) receives on its underside the coil (32) generating a field (B) on

the whole of the matrix (2).

  The method of production from an SOI subskat can for example be

  that described in patent application FR 00.15625 of the Applicant.

  I1 emerges from the above that the component according to the invention has multiple advantages, and in particular: - a high intensity of the force exerted on the defomable parts of the MEMS skucture; - an almost total absence of crosstalk; a greatly reduced energy consumption, and more precisely only limited to the phases of kansition, when the mechanical structure is of the bistable type; - the possibility of generating forces on the maker's frame to

generate pre-stresses.

  Industrial applications: Although described in the case of a makice of optical switches, the invention is not limited to this single application. It can in particular be used to form optical attenuators, or even projection systems including controls for mirrors. The invention is also not limited to optical applications only, but may also be of interest

  in many MEMS skuctures organized in the form of makices.

3 2838421

Claims (8)

  1 / Microelectronic component (1) comprising a matrix of cells (2) of micro electromechanical structures, each cell comprising a movable part (10) connected to a fixed frame (12) by at least one connecting portion (13), Said matrix also comprising means for ensuring the selective displacement of the movable part of each structure, characterized in that said displacement means comprise: means for generating a magnetic field (B) acting on the entire matrix (2) ; means for selectively ensuring, for each cell, the passage of an electric current in the connection portion (13) connecting the mobile part (10) to the fixed frame (12), so that said connection portion ( 13) undergoing Laplace's forces deforms causing the   displacement of the mobile part (10) of the selected cell.   2 / microelectronic component according to claim 1, characterized in that it also comprises means for imposing the direction of the electric current flowing in the connecting portion (13), so as to control the direction   forces undergone by said connecting portions (13).   3 / microelectronic component according to claim 2, characterized in that the connecting portions (13) have two stable equilibrium positions, between which they can move under the effect of Laplace forces, so that the part mobile (10) of each cell can adopt two positions stable.   4 / microelectronic component according to claim 2, characterized in that the connecting portions (13) have a general beam shape, working in buckling. : 4 2838421   / Microelectronic component according to claim 1, characterized in that it comprises metallized zones (15-19), covering at least one Daction of the connection portions (13), and also forming conductive tracks (16,17)   for supplying current to each connecting portion (13).   6 / microelectronic component according to claim 1, characterized in   that the magnetic field is continuous or pulsed.   7 / microelectronic component according to claim 1, characterized in   that the magnetic field is generated by a single winding (32).   8 / microelectronic component according to claim 1, characterized in   that the magnetic field is parallel to the plane of the matrix.   9 / microelectronic component according to claim 1, characterized in that the connecting portions are substantially perpendicular to the magnetic field. / Microelectronic component according to claim 1, characterized in   that the magnetic field is perpendicular to the plane of the matrix. DEPOSITOR: MEMSCAP   AGENT: Cabinet LAURENT &CHARRAS; l