MOVABLE PIEZO ELEMENT AND METHOD FOR PRODUCING A MOVABLE PIEZO ELEMENT

Abstract

A movable piezo element and to a method for producing the element are provided. The movable piezo element may have a structured substrate, in which an intermediate layer is arranged between a first substrate layer and a second substrate layer. The element may also have a first electrode layer. The element may also have a second electrode layer arranged on the ferroelectric, piezoelectric, or flexoelectric layer. The second substrate layer may be structured such that at least one bar of the second substrate layer is formed. The bar may be clamped on one side and may be physically spaced from the first substrate layer. A surface of the bar facing away from the first substrate layer, and/or a lateral surface of the bar, may be at least partly covered by another layer.

Claims

1-14. (canceled)

15. A movable piezo element, comprising: a structured substrate in which an intermediate layer is arranged between a first substrate layer and a second substrate layer; a first electrode layer of an electrically conductive, non-ferroelectric material arranged on the second substrate layer; a ferroelectric, piezoelectric, and/or flexoelectric layer arranged on the first electrode layer; and a second electrode layer of an electrically conductive, non-ferroelectric material arranged on the ferroelectric, piezoelectric, and/or flexoelectric layer, wherein the second substrate layer is structured such that at least one bar of the second substrate layer mounted on one side is formed that is spatially spaced apart from the first substrate layer; and a surface of the bar remote from the first substrate layer and/or a side surface of the bar is/are at least partially covered by a layer stack of the first electrode layer, the ferroelectric, piezoelectric and/or flexoelectric layer and the second electrode layer.

16. The movable piezo element in accordance with claim 15, wherein the first electrode layer, the ferroelectric, piezoelectric, and/or flexoelectric layer and/or the second electrode layer have a thickness variation at the side surface of below 10% or of a maximum of 5 nm.

17. The movable piezo element in accordance with claim 15, wherein the bar is connected with material continuity along its longitudinal axis at at least one end to the further second substrate layer.

18. The movable piezo element in accordance with claim 15, wherein the bar is meandering or spiral.

19. The movable piezo element in accordance with claim 15, wherein the ferroelectric, piezoelectric, and/or flexoelectric layer has undoped or doped hafnium oxide, undoped or doped zirconium oxide, or an alloy therefrom, with the doped hafnium oxide or the doped zirconium oxide is doped with silicon, aluminum, germanium, gallium, iron, cobalt, chromium, magnesium, calcium, strontium, barium, titanium, zirconium, yttrium, nitrogen, carbon, lanthanum, gadolinium, and/or an element of the rare earths.

20. The movable piezo element in accordance with claim 15, wherein the movable piezo element is used as a MEMS switch, as a MEMS filter, as a MEMS phase shifter, as a cantilever for atomic force microscopy, as a microfluid switch, as a microfluid valve, as a micromirror, as a micropositioner, as an ultrasound transducer, as an ultrasound sensor, as a loudspeaker, as a microphone, as a seismograph, as a microspectrometer, as a micromechanical latching mechanism, as a micromechanical step motor, as a Fabry-Pérot interferometer, or as a flagelliform drive for a micromechanical application.

21. A method of producing a movable piezo element, comprising: a substrate in which an intermediate layer is arranged between a first substrate layer and a second substrate layer is structured such that the second substrate layer is removed in at least one region such that at least one elevated portion of the second substrate layer is formed in the region; a first electrode layer of an electrically conductive, non-ferroelectric material is applied to the second substrate layer; a ferroelectric, piezoelectric, and/or flexoelectric layer is applied to the first electrode layer; and a second electrode layer of an electrically conductive, non-ferroelectric material is applied to the ferroelectric, piezoelectric, and/or flexoelectric layer; and then at least one bar of the second substrate layer that is mounted on one side is generated in that the intermediate layer between the bar of the second substrate layer and of the first substrate layer is removed.

22. The method in accordance with claim 21, wherein the intermediate layer is formed from an electrically insulating oxide that has a thickness between 100 nm and 10 μm.

23. The method in accordance with claim 21, wherein before the removal of the intermediate layer, a filler layer is applied that covers the second electrode layer and that is subsequently structured such that as a hard mask it does not cover at least one side surface of the bar.

24. The method in accordance with claim 23, wherein the filler layer is removed by a wet chemical etching process and in so doing the at least one side surface of the bar is also exposed.

25. The method in accordance with claim 21, wherein as the last method step, the first electrode layer and the second electrode layer are electrically contacted by an electrical voltage source.

26. A component having the movable piezo element in accordance with claim 15 and having a transistor or a circuit, wherein the movable piezo element and the transistor or the circuit are electrically contacted by an electric contact having a distance of less than 50 μm.

27. The component in accordance with claim 26, wherein the movable piezo element and the transistor or the circuit are formed as an integrated circuit on a single substrate.

28. The component in accordance with claim 26, wherein the movable piezo element and the transistor or the circuit are formed in a single wiring plane of a CMOS process.

Description

[0027] There are shown:

[0028] FIG. 1 a schematic representation of a method of producing a highly integrated piezoelectric element in a side view;

[0029] FIG. 2 a schematic representation of a method of producing a single piezoelectric element in a view corresponding to FIG. 1;

[0030] FIG. 3 a cross-section of a piezoelectric element in a view corresponding to FIG. 1;

[0031] FIG. 4 a schematic representation of the excursion of a piezo element oscillating in plane in a view corresponding to FIG. 1;

[0032] FIG. 5 a schematic representation of the excursion of a piezo element oscillating both in plane and out of plane in a view corresponding to FIG. 1;

[0033] FIG. 6 a schematic representation of the excursion of a piezo element oscillating in plane in a plan view and a side view;

[0034] FIG. 7 a piezoelectric element for the deflection of an atomic force microscope tip in a view corresponding to FIG. 6;

[0035] FIG. 8 a plan view of a piezoelectric element that is used in a lateral movement as a switch or as a microfluidic sluice;

[0036] FIG. 9 a view corresponding to FIG. 8 of a use of a piezoelectric element in a lateral movement as a switch or as a microfluid valve;

[0037] FIG. 10 a plan view of an oscillating, meandering piezo element;

[0038] FIG. 11 a perspective view of a spiral system having a piezo element;

[0039] FIG. 12 a simulation of an oscillating, meandering membrane in a plan view;

[0040] FIG. 13 a schematic representation of a miniaturized drive in a plan view;

[0041] FIG. 14 a plan view of a micropositioner;

[0042] FIG. 15 a schematic representation of a miniaturized piezo tube;

[0043] FIG. 16 a schematic plan view of a micromechanical step motor.

[0044] FIG. 1 shows a method of producing a piezoelectric element in a schematic view. In Figure la, a substrate is shown in a cross-sectional view in which an intermediate layer or sacrificial layer 101 is arranged between a first layer 100 as a first substrate layer and a second layer 102 as a second substrate layer. The substrate is a so-called “silicon-on insulator” wafer in the embodiment shown, i.e. the first layer 100 and the second layer 102 consist of intrinsic or heavily doped silicon, while the intermediate layer 101 in this embodiment is produced from a typical sacrificial layer material known from the production of microelectromechanical systems, typically silicon oxide. Metals such as aluminum, copper, molybdenum, vanadium, chromium, iron, nickel, palladium, cadmium, platinum, cobalt, gold, tin, zinc, indium, or alloys therefrom, conductive oxides such as doped strontium titanate, lanthanum strontium manganite, and also further elastic materials, preferably electrically conductive materials, such as silicon nitride, carbon nanotube films, or polymers having a high glass transition temperature can furthermore in particular be considered for the second substrate layer 102.

[0045] The second layer 102 as a later, oscillating element can be applied with a layer thickness of 50 nm to 10 μm, preferably 100 nm to 2 μm. The intermediate layer 101 as a sacrificial layer or an insulating layer can have a layer thickness between 100 nm and 10 μm, preferably 200 nm to 3 μm.

[0046] The structure shown in FIG. 1b) in which the second layer 102 has at least one elevated portion, typically a columnar or wall-shaped elevated portion, is obtained by applying a hard mask or resist layer 103 and a subsequent structuring (for example by wet chemical etching, ion etching, or reactive ion etching) of the second layer 102 as the elastic layer.

[0047] The hard mask or the resist layer or the resist film 103 is removed by an etching, preferably a dry etching, as shown in FIG. 1c). A first electrode layer 104 is applied in a compliant manner as a back electrode on the second layer 102 of the substrate. The first electrode layer 104 is applied by atomic layer deposition from an electrically conductive material such as titanium nitride to obtain a compliant deposition. Alternatively, other metals can, however, be used as the electrode material such as aluminum, copper, molybdenum, vanadium, chromium, iron, nickel, palladium, cadmium, platinum, cobalt, gold, tin, zinc, indium, or alloys therefrom, and also further elastic materials, preferably electrically conductive materials, such as silicon nitride, doped or undoped alloys of silicone and germanium such as B:SiGe, carbon nanotube films, or polymers having a high glass transition temperature. In this case, the second substrate layer 102 and the first electrode layer 104 can be present in a common layer.

[0048] A ferroelectric, piezoelectric, or flexoelectric layer 105 of hafnium oxide, zirconium oxide, or alloys therefrom is deposited as a ferroelectric material on the first electrode layer 104, for which purpose atomic layer deposition has likewise been used. The ferroelectric layer 105 is in turn formed as a compliant layer. In further embodiments, an alternating atomic layer deposition of hafnium oxide and a respective dopant or an alternating atomic layer deposition of hafnium oxide and a respective dopant and alternatingly a further oxide, for example Al.sub.2O.sub.3 can take place. In this case, nitrogen, yttrium, carbon, strontium, scandium, silicon, aluminum, gadolinium, iron, germanium, gallium, lanthanum, and also rare earths can be considered as the dopant.

[0049] The second electrode layer 106 is in turn applied as a compliant layer on the ferroelectric, piezoelectric, or flexoelectric layer 105 by means of atomic layer deposition and the structure shown in FIG. 1d) is thus achieved. Instead of atomic layer deposition, physical vapor deposition can also be used. Alternatively to this, a further layer can also be applied that acts as a hard mask. In this respect, the materials already named for the first electrode layer 104 can be considered as the materials.

[0050] All the layers are in direct contact with the respective adjacent layers and cover these layers completely. The structure formed in this manner is, as shown in FIG. 1e), filled with a filler layer 107 completely covering the second electrode layer 106 so that it forms a planar surface. The filler layer 107 is here typically formed from SiO.sub.2 and is applied by means of chemical vapor deposition. Other oxides can also be considered as the materials here.

[0051] The filler layer 107 is subsequently structured such that a respective side of the oscillator is liberated from the filler layer 107 (FIG. 1f). The metal ferroelectric metal layer stack formed from the first electrode layer 104, the ferroelectric, piezoelectric, or flexoelectric layer 105, and the second electrode layer 106 is subsequently etched, preferably by means of a wet chemical etching, which results in the configuration shown in FIG. 1g). A central connection of the layer stack between the remaining bars of the second semiconductor layer is subsequently separated in the embodiment shown and the remaining intermediate layer 106 beneath the bar is removed so that they can be present mounted on one side or at two sides, but can oscillate (FIG. 1h).

[0052] In this respect, side surfaces facing one another can be completely covered by the layer stack, while the upper sides are only covered by the layer stack by half. An electrical contacting of the first electrode layer 104 and of the second electrode layer 106 can be provided by a voltage source 110 as the last step. The configuration obtained by means of the method shown in FIG. 1 is shown in FIG. 11.

[0053] The described method can be easily integrated in the CMOS process flow of a high k metal gate process flow in that so-to-say a ferroelectric, piezoelectric, or flexoelectric capacitor is applied to a membrane (namely the substrate) and the piezoelectric properties are thus implemented. In this respect, the ferroelectric, piezoelectric, or flexoelectric phase of the materials is used. The piezoelectric expansion or shrinking in the plane of the membrane while applying an electrical voltage to the first electrode layer and the second electrode layer by an electrical voltage source results in a bending of the membrane. Unlike in electrostatic systems, this direction of movement is implemented in both mechanical strain directions.

[0054] The ferroelectric, piezoelectric, or flexoelectric layer 105 as a thin film is, as already mentioned, CMOS compatible and is often implemented as a gate dielectric in common CMOS processes. The described piezoelectric elements can therefore be produced in a CMOS process line, which allows lower production costs and a higher throughput than with conventional methods. The small thickness of the capacitor thereby formed enables a high scalability for very greatly miniaturized systems. Since the piezoelectric element is lead-free, RHoS compliance is also present. A capacitor having an insulating layer is formed in the described method whose piezoelectric properties result in distortion. A vertical integration is also made possible by the compliant deposition of the ferroelectric, piezoelectric, or flexoelectric in three-dimensionally structured substrates. Significant tensions of the film and thus a bending of the bar is generated even at small electrical voltages below 5 V by use of a thin film ferroelectric, piezoelectric, or flexoelectric. The required control voltage is thus considerably below currently available low voltage solutions or other oscillators based on electrostatic approaches. In the embodiment shown, a thin film ferroelectric, piezoelectric, or flexoelectric having a thickness below 50 nm is used. Changes of the polarization state thus already result at small electrical voltages and the required control voltage is considerably smaller than in already known low voltage solutions. This is in particular sensible at low power solutions.

[0055] It is possible to use ultralaminates to increase a breakdown strength. They are oxide layer of, for example, Al.sub.2O.sub.3, SiO.sub.2, or ZrO.sub.2 having a layer thickness of a maximum of 3 nm. They are introduced alternatingly to the doped or undoped hafnium oxide or zirconium oxide or alloys therefrom with single layer thicknesses of 3 nm to 20 nm. In addition to a breakdown voltage, a switching voltage is thus also increased and raised by a factor of at least 5. An alternating series control of the ferroelectric, piezoelectric, or flexoelectric capacitors can additionally be carried out for high voltage applications.

[0056] Hafnium oxide doped with silicon, aluminum, germanium, magnesium, calcium, strontium, barium, titanium, zirconium, nitrogen, carbon, silicon, gallium, iron, cobalt, nickel, cadmium, scandium, yttrium, lanthanum, vanadium, and elements of the rare earths or undoped hafnium oxide as well as further compliant ferroelectrics that can be deposited can be considered as materials. In comparison with other ferroelectrics, these materials have a much smaller permittivity; considerably reduced loss currents are therefore caused by the capacitive load. It is thus possible due to the CMOS compliance of the hafnium oxide (HfO.sub.2) or of the zirconium oxide (ZrO.sub.2) and of said dopants to produce further electronics on the same substrate, that is an on-chip production, as a so-called system-on-chip (SoC). The described element can, however, also be produced as a single miniaturized SMD component (surface mounted device) so that even very small construction shapes such as the 01005 format can be operated. The oxide layer can be formed as an aluminum oxide layer (Al.sub.2O.sub.3), a silicon oxide layer (SiO.sub.2), and/or a zirconium oxide layer (ZrO.sub.2).

[0057] The described piezo element is suitable for different applications, for example, uses in sound, ultrasound, microfluids, micropumps, or microoptics. A use in radio frequency technology can equally also take place. In these application fields, considerable miniaturizations can be achieved in comparison with known techniques. A high degree of design freedom that thus allows a good scaling of the resonances can be achieved for the sound and ultrasound applications by the integration capability in the CMOS and MEMS process flow. A vibration compensation that is necessary in a rough environment to ensure the functional capability is furthermore possible by a configuration of out-of plane and in-plane oscillators on a single chip.

[0058] In FIG. 2, an analogous method with a single cantilever is shown in a view corresponding to FIG. 1. Recurring elements are provided with identical reference numerals in this Figure and also in the following Figures. Since one of the side surfaces is completely covered and the surface remote from the first substrate layer 100 is covered by at least half by the layer stack at the bar itself, a monomorphic in-plane oscillator can be implemented. The layer stack applied to the side surface and the layer stack applied to the surface remote from the first substrate layer 100 are here formed continuously as a single layer stack, i.e. are applied with material continuity. The side surface disposed opposite the covered side surface is not covered by the layer stack, i.e. is completely exposed.

[0059] In FIG. 3, a heterostructure of the first layer 102 is shown in a view corresponding to FIG. 1. The same materials of the layer 102 can here be considered as the materials for the three layers 111, 112, 113 in this example. It can furthermore consist of more than the shown three layers. Since one of the side surfaces is completely covered and the surface remote from the first substrate layer 2 is covered by at least half by the layer stack at the bar itself, a monomorphic in-plane oscillator can be implemented.

[0060] The mechanism of the movement of the layer movable in the plane is shown schematically in FIG. 4 in a side view corresponding to FIG. 1 with a possible contacting of the oscillator. The first electrode layer 104 and the second electrode layer 106 are connected to the electrical voltage source 110. On application of a voltage, the oscillator or cantilever moves by the distance 108 in the plane. There is a proportional relationship here between the excursion and the applied voltage in the static case. This thus also enables negative excursions or a good controllability of the excursion.

[0061] The mechanism is shown schematically in FIG. 5 of both in-plane and outplane movability, that is the three-dimensional movability, of the layer 102 or of the cantilever formed therefrom in a side view corresponding to the preceding Figures. In this respect, the layer stack of the first electrode layer 104 and the second electrode layer 106 and the ferroelectric, piezoelectric, and/or flexoelectric layer 105 is structured correspondingly so that a separate contacting of the now separated electrode layers 106 and 104 is made possible by means of the voltage source 110. On application of an electrical voltage, the oscillator moves in the plane or in the cantilever by the distance 108 in-plane and by the distance 109 out-plane. There is a proportional relationship here between the excursion and the applied voltage.

[0062] FIG. 6 shows, analogously to FIG. 4, the excursion of the bar mounted on one side in a corresponding side view and in a plan view.

[0063] FIG. 7 shows a cantilever for atomic force microscopy (AFM) in a plan view and in a side view with which an in-plane movement can likewise be achieved.

[0064] This is sensible, for example, for optically assisted AFM methods. The signal of the damping is used to regulate the AFM peak here. The alternating voltage is applied between the second substrate layer 102 as the semiconductor layer and the upper, second electrode layer 106. FIG. 7b) shows the corresponding peak in a side view.

[0065] A microfluid spigot or a microfluid switch is shown in a plan view in FIG. 8. A plurality of movable bars can be combined here, as shown; for example, to implement a valve for the throughflow of a microfluid channel 132. A coupling to the outer electrical voltage source 110 results in a change of the flow path.

[0066] A microfluid switcher or a microfluid switch is shown in a plan view in FIG. 9. The bar movable in-plane can be arranged in a row and can, for example, be contacted by a common top electrode 130 to move a bar 131 linearly. The electrode 130 can equally be present in a further structured manner. The bar 131 can be introduced into a microfluid channel 132, for example. It here results in a control of the throughflow of the microfluid channel 132. The lateral position of the bar can thus be varied, and the throughflow can thus be controlled, by means of an external voltage of the voltage source 110, which can equally serve as a switch for the throughflow.

[0067] The oscillating bar can also be meandering, as shown in a plan view in the unloaded state and in the loaded state in FIG. 10. This enables a considerably increased excursion, as shown in the simulation shown schematically in FIG. 10.

[0068] A spiral shape or helical shape of the oscillating part is shown schematically in a perspective view in FIG. 11. This shape is particularly suitable for gyroscopes or (Cardan) mirror holders.

[0069] In further embodiments, the second electrode layer can also be applied compliantly as a mirror stack, for example by a heterostructure of titanium oxide and aluminum oxide (e.g. 67 nm Al.sub.2O.sub.3 and 49 nm TiO.sub.2 produce a mirror for a wavelength range from 420 nm to 500 nm). In particular laser light can be deflected by this and an integration into a Fabry-Perot system is possible.

[0070] FIG. 12 shows a meandering structure in a plan view. A particularly elastic material is used as the material for the membrane element. As shown in the non-deflected state in FIG. 12a), a plurality of meandering structures as inplane oscillators can also be connected to an inner spring. FIG. 12b) shows the deflected state.

[0071] Analogously to a miniaturized loudspeaker, the piezoelectric membrane can also be used to detect sound waves, that is it can be used as a microphone. The sound waves induce a movement of the membrane and a measurable electrical voltage and a measurable electric current are thus generated. Such a loudspeaker can also be used as a seismograph.

[0072] A miniaturized drive based on the already proposed cantilever, preferably in the meandered form 150, is shown in a schematic plane view in FIG. 13. It is possible to drive small objects, so-called nanobots, in a fluid by a battery or other electrical energy source. A body also has to be exposed for this purpose. An adaptation of the RC times for applying the voltage by small conductivity should takes place so that the CMOS circuit 151 drawn at the center of FIG. 13 regulates the voltage at the individual drivetrains or flagellums. Small antenna elements may also be contained that enable an external control.

[0073] A microspectrometer has a mirror element that can also be applied on the side and on the upper side by means of atomic layer deposition. These systems can then be integrated in a so-called “silicon photonics device” to rotate the beam between different optical paths, for example. A use as a spectrometer is likewise possible, with the meandering form being able to be used as an optical grid here.

[0074] With a plurality of cantilevers, a concave mirror shape can also be implemented with which a focus can be generated or also deactivated by means of an electric control of the individual cantilevers.

[0075] FIG. 14 shows a micropositioner in a plan view in which a bar or an object associated therewith can be positioned by means of a plurality of cantilevers. It is moreover also possible to implement a miniaturized loudspeaker by means of the discussed membrane structure.

[0076] A further application option is the slit piezotube shown in FIG. 15. A dimorph, i.e. electrodes on both sides of the tube have to be electrically separated from one another, is used here. A mirror is attached to the tube. The alignment of the mirror can be controlled by means of the distortion of the piezotubes. This can be used, for example, for LIDAR (light detection and ranging).

[0077] A micromechanical latching mechanism shown schematically in FIG. 16 can also be produced with which, for example, the rotational state of a micromechanical cog can be controlled.

[0078] It is furthermore possible to implement a linear micromechanical step motor having an oppositely disposed row of cantilevers with a coordinated movement. A microcavity that can be micromechanically coordinated or a Fabry-Pé-rot interferometer can finally also be produced. A membrane is used here. The incident light is filtered in dependence on the wavelength of the light. The distance between the cantilever and the reference window is here typically in the order of magnitude of the wavelength of the light used. The position of the membrane is modulated by means of an external voltage.

[0079] Only features of the different embodiments disclosed in the embodiments can be combined with one another and claimed individually.