DE102009021936A1 - Optical filter and a method for producing an optical filter - Google Patents

Abstract

An optical filter (10) comprises a first partially transparent mirror (1a), a second partially transparent mirror (1b) and a deformable spacer (7; 15) disposed between the first partially transparent mirror (1a) and the second partially transparent mirror (1b) is. A distance (20) between the first partially transparent mirror (1a) and the second partially transparent mirror (1b) is variable as a function of a voltage (U) applied to the optical filter (10).

Description

The The present invention relates to an optical filter and a method for producing an optical filter and in particular to a linear variable optical filter, which is used for optical spectroscopy suitable is.

The Optical spectroscopy involves the analysis of electromagnetic radiation in a certain frequency band. Each frequency becomes one Assigned intensity. Essentially limited you look at the electromagnetic spectrum from the UV to the MIR range (UV = ultraviolet, IR = infrared, MIR = mean infrared). However, some investigations are also carried out in DUV (deep ultraviolet) or in FIR (far infrared).

through optical spectroscopy it is possible the spectral composition from radiation sources. It can also be investigated to what extent the spectral composition of a radiation modified by introducing samples (transmission or reflection) becomes. From the particular wavelength-dependent Transmission, reflection or absorption may be substance specific Ingredients, properties, bonding conditions, band structure (Band gaps) and other physical and chemical properties getting closed. Thus, by means of optical spectroscopy Materials are characterized. Likewise, dynamic, Time-dependent physical and chemical processes investigated become.

• 1) Dispersive Spektrometer. Der am weitesten verbreitete Ansatz beruht auf der Verwendung eines dispersiven Elements, wie z. B. eines Prismas oder, häufiger, eines Gitters. Das einfallende Licht wird durch das dispersive Element in seine spektralen Bestandteile zerlegt und auf einen Multikanaldetektor abgebildet, so dass jedem Detektorelement eine Intensität und ein Wellenlängenbereich zugeordnet wird. Alternativ wird das dispersive Element verkippt, so dass das Spektrum zeitlich sequentiell von einem Einzeldetektor erfasst werden kann. Um hohe Flexibilität für die Bandbreite und die Zentralwellenlänge des Spektrometers zu erreichen und auch hohe Flexibilität in Bezug auf die Auflösung zu erzielen, sind verschiedene Bauteile, wie Linsen oder Spiegel, einstellbare Schlitze, gegebenenfalls verschiedene dispersive Elemente und Detektoren erforderlich. Dies macht den Aufbau komplex und die Abmessungen groß, was sich auch in einem verhältnismäßig hohen Preis widerspiegelt.
• 2) FT-Spektrometer. Fourier-Transform-(FT-)Spektrometer basieren auf der Interferenz von zwei Teilstrahlen, die üblicherweise mittels eines Michelson-Interferometers generiert werden. Bei Änderung der optischen Weglängendifferenz werden Interferenzminima und -maxima durchlaufen. Die Intensität wird abhängig von der optischen Weglängendifferenz der Teilstrahlen mit einem Detektor aufgenommen. Mittels einer Fourier-Transformation wird die wellenlängenabhängige Intensität – das Spektrum – errechnet. FT-Spektrometer haben vor allem für geringe Intensitäten im Vergleich zu vielen anderen Spektrometertypen Vorteile (z. B. Jaquinot-Vorteil). Nachteilig wirken sich die großen Abmessungen solcher Spektrometer und die hohe Empfindlichkeit gegenüber Schocks und Vibrationen aus. Dieser Nachteil wird potentiell durch neuartige auf MEMS beruhenden Ansätzen beseitigt, wobei im Allgemeinen die Auflösung durch diese Ansätze sinkt. Da das Spektrum immer sequentiell aufgenommen wird, erlauben FT-Spektrometer nicht die Erfassung schneller spektraler Veränderungen.
• 3) AOTF (Acousto Optical Tunable Filter): Eine stehende akustische Welle wird eingesetzt, um in einem transparenten Kristall periodische Brechungsindexvariationen hervorzurufen. Diese periodische Variation wirkt für das einfallende Licht als Gitter. Durch Frequenzänderung der akustischen Welle wird die Periodizität des Gitters geändert. Üblicherweise wird der Detektor so angeordnet, dass er die Intensität der ersten Beugungsordnung erfassen kann. Wird die Anregungsfrequenz geändert, so trifft auf den Detektor die 1. Beugungsordnung einer anderen Wellenlänge. Durch Variation der akustischen Anregungsfrequenz wird also das Spektrum zeitlich sequentiell aufgenommen. Nachteilig neben der sequentiellen Erfassung sind die hohen Kosten für solche Modulatoren und der hohe Justageaufwand.
• 4) Fabry-Perot-Filter. Die Wirkungsweise von Fabry-Perot-Filtern beruht auf Vielfachinterferenz. Im einfachsten Fall werden zwei teiltransparente Spiegel planparallel angeordnet. Nur Wellenlängen, welche die Interferenzbedingung erfüllen werden von dem Filter transmittiert. Entscheidend ist dabei der Abstand der Spiegel. Zur Erfassung eines kompletten Spektrums wird der Abstand der Spiegel verändert und die Intensität als Funktion des Abstands aufgenommen. Aus der rechnerischen Beziehung zwischen Abstand und Wellenlänge kann die Intensität wellenlängenabhängig aufgetragen werden. Auch hier wirkt sich nachteilig die zeitlich sequentielle Erfassung des Spektrums aus.
• 5) LVOF (Linear Variable Optical Filter): Der Ansatz ist ähnlich dem des Fabry-Perot-Filters mit dem Unterschied, dass die beiden Spiegel nicht planparallel sondern leicht verkippt zueinander angeordnet werden. Entlang der Verkippung ergeben sich somit unterschiedliche Interferenzbedingungen. Dies führt dazu, dass entlang der Verkippung unterschiedliche Wellenlängen transmittiert werden.

There are different approaches to the realization of a spectrometer:

• 1) Dispersive spectrometer. The most common approach is based on the use of a dispersive element, such. As a prism or, more often, a grid. The incident light is decomposed by the dispersive element into its spectral components and mapped onto a multi-channel detector, so that each detector element is assigned an intensity and a wavelength range. Alternatively, the dispersive element is tilted, so that the spectrum can be detected sequentially in time by a single detector. In order to achieve high flexibility for the bandwidth and the central wavelength of the spectrometer and also to achieve high flexibility in terms of resolution, various components such as lenses or mirrors, adjustable slots, possibly different dispersive elements and detectors are required. This makes the structure complex and the dimensions large, which is also reflected in a relatively high price.
• 2) FT spectrometer. Fourier transform (FT) spectrometers are based on the interference of two sub-beams, which are usually generated by means of a Michelson interferometer. When changing the optical path length difference interference minima and maxima are traversed. The intensity is recorded depending on the optical path length difference of the partial beams with a detector. By means of a Fourier transformation, the wavelength-dependent intensity - the spectrum - is calculated. FT spectrometers have advantages especially for low intensities compared to many other spectrometer types (eg Jaquinot advantage). Disadvantageous are the large dimensions of such spectrometers and the high sensitivity to shocks and vibrations. This disadvantage is potentially eliminated by novel MEMS-based approaches, which generally lower the resolution through these approaches. Since the spectrum is always recorded sequentially, FT spectrometers do not allow the detection of fast spectral changes.
• 3) AOTF (Acousto Optical Tunable Filter): A standing acoustic wave is used to produce periodic refractive index variations in a transparent crystal. This periodic variation acts as a grating for the incident light. By changing the frequency of the acoustic wave, the periodicity of the grating is changed. Usually, the detector is arranged so that it can detect the intensity of the first diffraction order. If the excitation frequency is changed, the first diffraction order of a different wavelength strikes the detector. By varying the acoustic excitation frequency, the spectrum is thus recorded temporally sequentially. Disadvantages in addition to the sequential detection are the high costs for such modulators and the high adjustment effort.
• 4) Fabry-Perot filter. The mode of action of Fabry-Perot filters is based on multiple interference. In the simplest case, two partially transparent mirrors are arranged plane-parallel. Only wavelengths that satisfy the interference condition are transmitted by the filter. The decisive factor is the distance between the mirrors. To detect a complete spectrum, the distance of the mirrors is changed and the intensity is recorded as a function of the distance. From the mathematical relationship between distance and wavelength, the intensity can be plotted as a function of wavelength. Here, too, adversely affects the temporal sequential acquisition of the spectrum.
• 5) Linear Variable Optical Filter (LVOF): The approach is similar to that of the Fabry-Perot filter with the difference that the two mirrors are arranged not plane-parallel but slightly tilted to each other. Along the tilting thus result in different interference conditions. This causes that along the Verkip pung different wavelengths are transmitted.

In US 6,909,548 B2 It is described that the variability with respect to the transmitting wavelength is achieved by positioning the light beam at a specific position along the tilt, depending on the desired wavelength to be transmitted. Alternatively, the filter is translated translationally.

As an integrated spectrometer LVOFs z. B. as in US 2002/0131047 A1 used. 2 shows such a conventional linear variable optical filter (LVOF), with a first partially transparent mirror 1a , a second semi-transparent mirror 1b that tilts to each other from a left end 4 to a right end 6 are arranged. As in 2 shown, the LVOF 10 with the mutually tilted arranged mirrors 1a and 1b with a line detector 2 combined. With vertical (vertical) light incidence 3 will be on the left end 4 where the distance of the two mirrors is the lowest, transmits the wavelength λ ₁ ± Δλ and strikes the underlying detector element 5 , At the right end 6 the wavelength λ ₂ ± Δλ satisfies the constructive interference condition, where: λ ₂ > λ ₁ . Δλ describes the bandpass width which depends on the finesse of the filter, which in turn is essentially determined by the reflectances of the mirrors. The central wavelength of the filter is (λ ₂ - λ ₁ ) / 2.

The required tilt angles are usually of the order of 0.05 °. The precision mechanical production is therefore very difficult. A first approach to producing LVOF filters in surface micromachining is in A. Emadi, H. Wu, S. Grabarnik, G. de Graaf and RF Wollfenbuttel, FABRICATION OF TAPERED OPTICAL STRUCTURES USING RESIST REFLOW, Proc. of 19th MicroMechanics Europe Workshop, Aachen (2008) described.

All previously known solutions have in common that the wavelength range λ ₂ - λ ₁ is not adjustable due to the fixed tilt angle. The resolution which can be achieved in principle is therefore likewise determined and results from the number N of the detector elements according to: Resolution = (λ ₂ -λ ₁ ) / N.

outgoing from this prior art, the present invention is the Task is to provide a spectrometer, characterized by a small size, low Production costs, and low shock and vibration insensitivity while offering high flexibility in terms of on the wavelength range and the resolution offers.

This The object is achieved by an optical filter according to claim 1 and a Process for producing an optical filter according to claim 20 solved.

Of the The present invention is based on the finding that an optical Filter with variably adjustable central wavelength and Resolution can be created by having a first and second semi-transparent mirror by a deformable spacer are separated from each other, wherein the deformable spacer by applying an electrical voltage to the optical filter changes the distance between the first and second semi-transparent mirrors.

The change the distance may be that the first and second partially transparent mirrors by area are arranged parallel to each other or one above the other and changed the distance between the mirror arranged parallel to each other becomes. This allows the central wavelength of the optical filter vary. In further Ausfüh approximately examples Further, the spacer may be deformed such that the first semitransparent mirrors a tilt in relation to the second partially transparent mirror and the tilting through Apply the voltage or by changing the applied Tension is changed.

at Further embodiments, the variable spacer an electroactive material, which upon application of an electrical voltage either contracts or expands, so that by a Voltage change, a change of the distance is effected. By the electroactive material both the Tilting as well as the mean distance of the first and second Mirror are adjustable, so that the central wavelength (Center frequency) and the wavelength range (resolution) be set independently.

Of the for example, first and second partially transparent mirrors each form a plane or a plate, so that at a relative tilting results in a linear variable optical filter. The tilt is due to the tilt angle (= angle of intersection the surface normals of the two mirrors). This Optionally, linear variable optical filters can also be replaced by a Surface micromechanics are produced.

at Further embodiments includes the optical Filter one or more transparent mirror plates over Aktorelemente can be tilted from electroactive material, the distance between the mirror plates being adjusted can.

Optionally, the two mirrors can be a Vor have tilting, which is realized by an additional spacer. The pre-tilting can be realized for example by a fusible material, wherein the fusible material may comprise, for example, a polymer. In further embodiments, the addi tional spacer for setting a pre-tilting of a passive or an active (optical) material consist.

at Further embodiments, the optical filter a detector row, wherein the detector row a plurality of one-dimensional arranged detector cells (eg., Photocells) and the optical filter is arranged on the detector line.

at Further embodiments is the optical filter realized as a two-dimensional arrangement, so that, for example the tilt is formed along one of the two directions and along the other of the two directions either the tilt does not change or different detectors arranged for different wavelength ranges are. In such two-dimensional arrangements, both a wide Spectrum as well as a selected area with high resolution be recorded. This can for example be realized in such a way that the center distance for all along the other of the two Directions of successive filters is the same, the tilt however, the individual filter is different. embodiments thus also comprise a two-dimensional optical filter, wherein the tilt in the one direction of the detector elements uniform is and the optical filter thus as a spectral, spatially resolving (One-dimensional) sensor can be used.

The Electroactive material can optionally be replaced by two transparent electrodes be activated, the two transparent electrodes on two opposite sides of the electroactive material are arranged.

at Further embodiments is between the first and second semi-transparent mirrors a wedge or a plane-parallel Stack of electroactive material disposed, wherein the electroactive Material simultaneously fulfills an optical function by the incident light is the electroactive material in the propagation passes between the first and second partially transparent mirrors. In such a case, for example, an electrode as a resistor or resistance layer may be formed over the one Voltage drops, so that, for example, a voltage profile along the Verkippungsrichtung trains when different electrical Potentials to different contact areas of the resistance layer be created. The potentials can then, for example be chosen so that a uniform Tilting results. Optionally, additional Electrodes used to enhance the optical properties - in particular a possibly disturbing dispersion - compensate.

Of the First and second mirrors may be used in embodiments for example, made by a thin metal film which are vapor-deposited on a substrate such as glass so that this results in a partially transparent mirror. The transparency can in such a case over the layer thickness the thin metal film layer are adjusted. The semi-transparent Mirrors in the form of thin metal films can also be made in combination with dielectric layers. The partially transparent mirrors of the optical filter can also from a dielectric layer stack comprising a Bragg reflector forms, be prepared.

When electroactive materials come with materials, for example dielectric, electrostrictive, electrochemo-mechanical, piezoelectric or elastic properties in question. Alternatively, the electroactive Materials, for example, an ionic polymer-metal composite, a mechanical-chemical polymer / gel, a liquid crystal elastomer, or shape memory polymer / alloys. In particular, you can for example as electroactive materials polymers or copolymers and polymer blends are used. The electroactive material can further comprising an alloy or a mixture (composite / nanocomposite). Furthermore, in embodiments, the electroactive Material may be a ceramic or polymer-ceramic composite.

at For example, a wafer bonding method may be used to fabricate the optical filter used, for example, to apply the upper mirror (either on the spacer or the electroactive material). To achieve a sufficient optical quality, is in embodiments also the Dektektorzeile planarized and then the lower mirror (the Mirror facing the detector) on the detector / detector line arranged.

at Further embodiments may have the characteristics of the electroactive material by nanoparticles or nanoparticles be optimized - both in terms of electrical as also optical properties.

In further embodiments, the optical filter has an integrated position sensor, with optionally integrated light-emitting diodes or additional photodiodes or else Ele elements of the detector line can be used. Alternatively, laser diodes can also be used for the position sensor. It is also possible to use external light sources for calibration.

Around the tilt angle and the distance of the partially transparent mirror For example, capacitive ones can be used to determine Measurements are used. The position sensor can, for example piezoresistive or piezoelectric work and also can also have interference effects in the position sensor system be exploited.

The Signals of the position sensor can in further embodiments also used for a control system of the optical filter be to the desired wavelength range and / or to adjust and maintain the central wavelength. This for example, can be important to drift and over external influences, such as a temperature change, to be independent.

at Further embodiments, the electrostatic Attraction or repulsion exploited to a tilt / deflection to achieve the partially transparent mirror, for example Electrodes or electrode plates either opposite to two lying sides of a deformable (flexible) material formed may be or are arranged on the mirror plane and the mirrors limit the flexible material on both sides, so that when applying a voltage to the two electrode areas a electrostatic force is exerted that causes that the two mirrors either tilt towards each other or but attract or repel in parallel.

embodiments of the present invention can be used for a variety of Applications are used. For example, the optical filter in telecommunications (eg optical data transmission) or in the areas of DWDM (Dense Wavelength Division Multiplex) or WDM (Wavelength Division Multiplex). Furthermore, spectrometers, those on the linear variable optical filters according to embodiments based in mobile devices and / or chemical analysis and / or food control and analysis. After all can also chemometric with such kind of spectrometer Analyzes are carried out.

It is also possible in further embodiments, the optical filter in combination with a one- or multi-dimensional To operate shift device, so that a spectrally resolved two-dimensional image can be taken. The spectrally resolved For example, a two-dimensional image can be used for a plastic fitting or used for sorting waste. Other applications include water analysis, food analysis, blood analysis, gas analysis and environmental metrology.

embodiments can also be used to organic light-emitting Diodes (OLEDs) or liquid crystal displays (LCD displays) for example, in terms of color determination, the intensity and to influence Mura.

According to the invention Thus the task set above is solved by a LVOF according to the prior art is extended so that the distance and The tilt angle of the mirror realized flexibly adjustable are. Flexible means that both parameters (distance and tilt angle) during the operation of the filter independently and without Intervention in the optical system can be adjusted. By adjusting the center distance, the central wavelength be determined. By adjusting the tilt angle can the wavelength range can be selected. At N Detectors can be the principal resolution in the range of Central wavelength can be increased by the wavelength range (Tilt angle) is reduced. This will make it possible a spectrum within a large wavelength range and, if necessary, in selected (or all) areas to record the spectrum with higher resolution without having to replace optical or dispersive components. At the same time, such a filter is small in size produce, inexpensive to manufacture and has due the stiffness of the actuators used a very low shock and Vibration sensitivity.

embodiments The present invention will be described below with reference to FIG the accompanying drawing explained in more detail. It demonstrate:

1 a schematic representation of an optical filter according to embodiments of the present invention;

2 a representation of a conventional linear variable optical filter;

3 a cross-sectional view through a variable optical filter with four contact surfaces;

4 a cross-sectional view through a variable optical filter, which is arranged on a detector line;

5 a variable optical filter with pre-tilting;

6 a cross-sectional view of a variab len optical filter with electroactive material as an optical transmission medium;

7a , b is a representation of the dispersion and an electrode arrangement for dispersion compensation;

8th a plan view of a two-dimensionally formed variable optical filter array; and

9 an illustration of process steps for the production of an optical filter according to embodiments.

In terms of the following description should be noted that at the same embodiments or the same functionally identical functional elements have the same reference numerals and thus the description of these functional elements in different Embodiments are interchangeable.

1 shows a schematic representation of an optical filter with a first partially transparent mirror 1a , a second partially transparent mirror 1b and a deformable spacer 7 . 15 that is between the first and second semi-transparent mirrors 1a . 1b is arranged. A distance 20a between the first semitransparent mirror 1a and the second partially transparent mirror 1b is variable as a function of an applied to the optical filter (electric) voltage U. The spacer 7 . 15 can, for example, the distance 20 thereby changing that the first semi-transparent mirror 1a with respect to the second partially transparent mirror 1b is tilted, so that the tilting 20r to a variable distance 20a along the tilting direction (horizontal direction in the 1 ) leads. The tilt 20r corresponds to a tilt angle, which is adjustable for example in the order of 0.05 ° or in a range between 0 ° and 1 ° or between 0 ° and 0.1 °.

The first semi-transparent mirror 1a is formed that an incident light beam 3 the first partially transparent mirror 1a can happen and the second semi-transparent mirror 1b splits the incoming light beam 3 in a transmission proportion 3t and a reflection component 3a on, with the reflected portion 3r at the second partially transparent mirror 1b reflected and then to the first partially transparent mirror 1a is reflected back. As a result, a part of the reflected light beam 3r the second partially transparent mirror 1b happen and with the transmission share 3t interfere so that, depending on the phase shift between the transmission component 3t and the reflection component 3r an interference pattern arises. When the phase shift between the transmitted and the reflected light beam 3t and 3r is an integer multiple of a wavelength, so it comes to a gain (constructive interference) and if the phase difference (2n + 1) π / 2 for n = 1, 2, 3, ..., then there is an extinction, so that these spectral parts are eliminated (destructive interference). In general, it comes at the partially transparent mirrors 1a , b to a multiple reflection, so that the reflection component 3r several times between the first and the second partially transparent mirror 1a and 1b is reflected and with each reflection a part passes the respective partially transparent mirror. For a certain distance 20a the constructive interference condition is given only for certain wavelengths, with the largest wavelength just twice the distance 20a is. The linear variable optical filter, as shown in 1 is shown, now has the effect that along the tilting direction (horizontal direction), the constructive interference condition for different wavelengths is satisfied, so that, depending on the incident position of the incident light 3 another wavelength from the incident light 3 is filtered.

3 shows a cross-sectional view of a first embodiment of deformable spacers 7 that is between the first semi-transparent mirror 1a and the second partially transparent mirror 1b a first deformable spacer 7a and a second deformable spacer 7b are arranged. Between the first deformable spacer 7a and the second partially transparent mirror 1b is a first electrode 8a formed between the first spacer 7a and the first partially transparent mirror 1a is a second electrode 8b formed between the second deformable spacer 7b and the second partially transparent mirror 1b is a third electrode 8c formed and between the second deformable spacer 7b and the first partially transparent mirror 1a is a fourth electrode 8d educated.

The 3 Thus, Figure 1 shows a first embodiment of an adjustable LVOF in cross-section, with adjustable LVOF also being referred to as Adjustable Linear Variable Optical Filter (ALVOF). Both deformable spacers 7a , b, for example, together with their respective electrodes 8a - 8d have the same nominal thickness after the manufacturing process. As electroactive material, for example, the following materials can be used: dielectric elastomers, electrostrictive polymers, electrochemo-mechanical polymers, ionic polymer-metal composites, mechanical-chemical polymers / gels, ferroelectric polymers, piezoelectric polymers and ceramics, liquid crystal elastomers, shape memory Polymers and alloys. Other examples are silicones, viscoelastic foams, BTO, PZT, PVDF, copolymers of PVDF (PVDF-TrFE, PVDF-TFE), odd-nylons, VDCN, polyurea, as well as composites such as e.g. Polymer blends (PVF ₂ nylon), polymer ceramics (PVDF-PZT), nanocomposites (ie, nanoparticles of an electroactive inorganic material such as BTO in a polymer matrix), carbon nanotubes, monomorphic, bimorph, or polymorphic structure. The electrodes 8a - 8d are made of an electrically conductive material, such. Al, Au, AlSiCu, ITO, CdO, Ag, Cu, conductive polymers, .... The electrodes 8a - 8d are electrically isolated from each other and can be applied independently of each other with an electric potential. In the production and for the operation, it may be favorable to electrically connect two of the electrodes to each other (eg in the combination 8a With 8c or 8b With 8d permitted).

Lies between the electrodes 8a and 8b an electrical voltage, so stretched or contracted the spacer 7a , Depending on the selected material, either the inverse piezoelectric effect (eg PVDF, PVDF copolymers or polymer-polymer composites), the contraction based on the dielectric effect (eg BCB, Elastomers), electrostriction or other effects that characterize the respective electroactive materials.

As a concrete example, between the first and the second electrode 8a and 8b a first voltage is applied, which in turn results in deformation of the first deformable spacer 7a leads. In an analogous manner, between the third electrode 8c and the fourth electrode 8d a second voltage is applied, which in turn results in deformation of the second deformable spacer 7b leads. If the first and the second voltage are different from each other or if the material of the first and second deformable spacers 7a and 7b is different from each other, the application of the first and second voltage leads to different deformations of the first spacer 7a and the second spacer 7b , As a consequence, the first semitransparent mirror becomes 1a with respect to the second partially transparent mirror 1b tilted and there is a linear variable optical filter.

Alternatively, equal voltages between the first and second electrodes 8a . 8b or between the third and fourth electrodes 8c . 8d be created, which lead to the first semitransparent mirror 1a compared to the second partially transparent mirror 1b pa rallel shifts, so that the distance 20 along the horizontal axis in the 3 remains constant.

There left and right but also created different voltages can be a different distance of the glass plates be reached on the left and right ends. This is both a Setting the tilt angle as well as a change the center distance possible. This results in the adjustability the wavelength range and the central wavelength.

The Production of such a device is ideally carried out by means of Micromechanical techniques, since here layer thicknesses set exactly be a submicrometer accurate adjustment and structuring of layers is possible and layers of very high Homogeneity can be applied.

4 shows an embodiment in which the optical filter, as shown for example in the 3 is shown on a detector line 9 is arranged, the detector line 9 a plurality of detector elements 19 having. The detector elements 19 For example, they may comprise photodiodes and be formed, for example, in a substrate as semiconductor components.

The individual detector elements 19 measure the intensity of the incident radiation 3 located at the respective location (along the connecting line between the first and second spacers 7a , b) of the detector element 19 the linear optical filter 10 can happen. As previously described, only that wavelength range can pass through the optical filter for which the constructive interference condition is satisfied, such that the individual detector elements 19 can detect a different wavelength range. On the other hand, it is also possible if the tilt of the first semi-transparent mirror 1a relative to the second partially transparent mirror 1b is very small, so that both are almost parallel, that the linearly variable optical filter is sensitive only for one wavelength. The individual detector elements 19 In this case, the total bandpass width Δλ can be detected. The bandpass width Δλ depends on the finesse of the filter, the finesse being given by the reflectivity of the partially transparent mirrors. The higher the reflectivity of the partially transparent mirrors, the greater the finesse becomes 1a and 1b is because with increased reflectivity multiple reflections play an increasing role, so that even the smallest deviations from the constructive interference condition lead to an extinguishing effect (destructive interference).

Consequently The result is a spectrometer with adjustable wavelength range and adjustable central wavelength.

Optionally, only the first deformity is needed bare spacers 7a with the first and second electrodes 8a , b be formed and instead of the second deformable spacer 7b can also be formed a solid spacer. Thus, by applying a voltage, only the distance on one side is changed, while the distance remains fixed on the other side. In such an embodiment, however, the central wavelength and the wavelength range are not independently adjustable.

The maximum stroke of the structures used as actuators as deformable spacers 7a and 7b is limited by material properties and / or by the maximum usable electrical voltage. If a very large wavelength range is to be detected, then it is advantageous if the ALFOV is manufactured in such a way that the two mirrors already have a tilting without applying an electrical potential (pre-tilting).

5 shows such a device in cross section, wherein compared to the embodiment as shown in the 3 is shown, the deformable first spacer 7a through another (passive) spacer 13 was supplemented so that even with absent tension pre-tilting between the first partially transparent mirror 1a (upper mirror) and the second partially transparent mirror 1b (lower mirror) is formed.

By unilateral introduction of the other (passive) spacer 13 the left layer stack is reduced by the thickness of 13 higher than the right one. Will the upper mirror 1a Applied by a pick-and-place method, so can the spacers due to the different heights 7 and 13 an adhesion-promoting layer 14a . 14b be necessary. If this layer z. B. realized by a polymer with suitably low melting temperature, so despite the different heights of the spacers, a wafer bonding method can be used. In the wafer bonding method, a plurality of optical filters may be simultaneously formed in parallel on a wafer, the partially transparent mirrors being formed by common layers which are later separated by dicing.

In this method, the difference in height spacers over the additional spacer 13 thicker layer of the polymer applied. During the bonding, the temperature is increased so much that the polymer can be plastically deformed. The process control is designed so that the upper mirror 1a with all spacers on the polymer suitable contact has. After Vereinze ment of the components, the temperature is then increased so much that the polymer can be very easily plastically deformed, so almost reached the liquid state. Due to adhesion, the mirror plate becomes 1a then pulled to the spacer as the polymer is displaced. Optionally, this process can be assisted by additionally applying a suitable high pressure to the mirror plate for plastic deformation 1a is exercised. In combination with a detector line 9 In turn, an integrated spectrometer is realized.

Specifically, for example, between the second electrode 8b and the first partially transparent mirror 1a a first adhesion-promoting layer 14a and between the fourth electrode 8d and the first partially transparent mirror 1a may be a second adhesive layer 14b be educated. By applying an electrical voltage between the first electrode 8a and the second electrode 8b and / or another electrical voltage between the third electrode 8c and the fourth electrode 8d The preset tilt can be increased or decreased. The additional spacer 13 Optionally, it can also be produced as an active element, or of an electroactive material, so that tilting can be increased even further when voltage is applied.

6 shows a further embodiment of an ALVOF, in which unlike the in 4 shown embodiment, the actuating material itself forms the optical wedge. The electrodes 8th are planar, so that a first electrode surface 8e from a second electrode surface 8f through an actuator layer 15 is disconnected. The actuator layer 15 For example, once again has an electroactive material, so that the distance between the first electrode layer 8e and the second electrode layer 8f is variable depending on the applied voltage. The first electrode layer 8e or the second electrode layer 8f have, for example, a first and second layer thickness d1, d2, which have a suitably chosen thickness d (see below).

The preparation can be carried out, for example, as follows. On the detector line 9 is the first transparent electrode layer 8e (eg ITO: indium tin oxide or CdO) applied in layers. To achieve a high optical quality, the detector line should 9 be very planar. Typical production methods of detector lines seldom ensure this. Therefore, these are to be modified or it must be carried out subsequent to the actual production further process steps to achieve a sufficiently flat surface. In particular, planarization layers (for example oxides, such as doped silicate glass, polymers, for example polyimide or the like) are deposited and optionally treated by temperature and / or by chemical mechanical polishing (CMP).

On the first electrode layer 8e the electroactive material becomes an actuator layer 15 applied. This happens z. B. by spin coating, by spray coating or by bonding. On the actuator layer 15 eventually becomes the upper (second) electrode layer 8f applied, like the first electrode layer 8e is transparent.

In the first electrode layer 8e For example, a first voltage U _{1 can be} applied, and if the second electrode layer 8f has sufficient resistance, different potentials can be applied to opposite end points (eg a voltage U ₂ at the left horizontal end and a voltage U ₃ at a right horizontal end)

Lies between the first and second electrode layers 8e and 8f no voltage, so the actuator layer 15 a uniform thickness. In order to achieve a homogeneous change in the layer thickness, the potentials U ₂ and U ₃ are chosen equal. The electrical voltage U ₁ - U ₂ is then decisive for the homogeneous layer thickness. In order to achieve a wedge-shaped deformation, the potentials U ₁ , U ₂ and U _{3 are selected,} for example, as follows: U ₁ = 0 V (eg ground), U ₂ = 0 V, U ₃ ≠ 0 V. At the left end There is no tension and the layer is not deformed. On the right side, the voltage U ₁ - U ₃ ≠ 0 V is applied, which, depending on the effect and sign of the voltage, leads to a reduction in film thickness or enlargement. This case (layer thickness reduction on the right side) is the one in 6 shown case.

The tilt is exactly wedge-shaped when passing through the second electrode layer 8f formed voltage U ₃ - U ₂ linearly decreases, ie, a linearly changing potential gradient between the electrode layers is formed. This can be achieved by suitable choice of material, processing, design (eg by a layer thickness profile of the electrode layer 8f ) and wiring can be achieved in a very good approximation.

In the 6 not shown are the mirroring layers, preferably between the detector line 9 and lower electrode layer 8e or on the upper electrode layer 8f be applied. The mirroring layers may, for example, in turn comprise vapor-deposited thin metal layers or a transparent substrate such as glass.

Due to the dispersion of the material used as a wedge, the wavelengths are not linearly imaged on the detector line, so that, for example, different detector elements 19 detect a different large wave increment from Δλ. An approximate correction becomes possible if additional, separated, transparent electrodes are used, so that the layer thickness can be locally adjusted.

In the case of materials which have no dispersion, the refractive index is wavelength-independent and, in the case of such materials, the wavelength range from λ ₁ to λ _{2 is} uniformly imaged in the ALVOF due to the wedge-shaped tilting on the detector row arranged below it. In other words, the wavelength increment from detector element to detector element is identical (with equidistant arrangement of the detector elements). Each detector element 19 detects, for example, the wavelength increment Δλ / r, where r is the number of detector elements 19 is. However, this is an idealized situation. Strictly speaking, all materials have a dispersion, so that the refractive index n depends on the wavelength λ.

7a shows a typical course of the functional relationship between the refractive index n and the wavelength λ. In typical materials, the index of refraction n increases with increasing wavelength as in FIG 7a shown off. In gaseous media, such as air, however, the dispersion is generally negligible. Therefore, dispersion effects are negligible for all ALVOF where the gap between the two partially transparent mirrors 1a . 1b filled with a gas (eg air).

The dispersion, over the partially transparent mirror 1a . 1b and possibly other layers is introduced, is also negligible in many cases. However, if these layers (partially transparent mirrors and further layers) are relatively thick and have a material with significant dispersion, then in the case of an ALVOF, the wavelengths are no longer linear relative to the detector line arranged below them 19 be imaged. Thus, certain wavelength ranges are determined to be inaccurate with a resolution predetermined by the detector spacing, and other wavelength ranges too inaccurate. This is the case in particular when an actuator layer 15 is used with a high dispersion.

• 1) eine nicht-lineare Anordnung der Detektorelemente 19. Bei diesem Ansatz werden die Abstände der Detektorelemente 19 entsprechend der vorliegenden Dispersion der verwendeten Materialien angepasst. Dies ist sowohl für ein ALVOF mit einem Gas als auch für das mit einer aktiven bzw. deformierbaren Schicht zwischen den teiltransparenten Spiegeln möglich, wobei die deformierbare Schicht beispielsweise als Keil 15 zwischen den Spiegeln ausgebildet ist.
• 2) Für den Fall der deformierbaren Schicht (Aktorschicht 15) wird anstelle des linearen Spannungsabfalls an der oberen Elektrodenschicht 8f (siehe in 6) ein nicht-linearer Abfall der elektrischen Spannung eingesetzt, um die Dispersion zu kompensieren. Konkret würde also lokal die Spannung derart gewählt werden, dass sich ein linearer Abfall nicht der geometrischen, sondern der optischen Weglänge (= geometrische Weglänge × Brechungsindex) für die entsprechende lokale Wellenlänge ergibt. Da der Brechungsindex n Wellenlängenabhängig ist (siehe 7a), muss die geometrische Weglänge dies kompensieren, so dass die optische Weglänge in der Tat möglichst linear sich ändert. Es bil det sich somit eine nicht-linear deformierte Schicht zwischen den beiden Enden heraus.

To compensate for the dispersion effects, two approaches can be pursued:

• 1) a non-linear arrangement of the detector elements 19 , In this approach, the distances of the detector elements 19 adjusted according to the present dispersion of the materials used. This is possible both for an ALVOF with a gas and for an active or deformable layer between the partially transparent mirrors, the deformable layer being for example a wedge 15 is formed between the mirrors.
• 2) In the case of the deformable layer (actuator layer 15 ) is used instead of the linear voltage drop across the upper electrode layer 8f (see in 6 ) a non-linear voltage drop is used to compensate for the dispersion. Specifically, therefore, the voltage would be selected locally such that a linear decrease results not in the geometric but in the optical path length (= geometric path length × refractive index) for the corresponding local wavelength. Since the refractive index n is wavelength-dependent (see 7a ), the geometric path length must compensate for this, so that the optical path length indeed changes as linearly as possible. Thus, a non-linearly deformed layer is formed between the two ends.

This goal can be achieved, for example, by suitable structuring of the electrodes (the upper or lower electrode layer 8e or 8f ) can be achieved. On the one hand, the thickness d of the electrodes passing through from left to right could 8e or 8f are chosen to match the dispersion so that they are not constant, for example. For example, the first layer thickness d1 of the first electrode layer 8f in the 6 from the potential U ₂ to the potential U ₃ are continuously changing, so that the voltage drop on the individual sections between the contacts, on which U ₂ and U ₃ abut, varies in different ways because of the change in resistance. The layer thickness d1 can increase or decrease linearly or non-linearly. The thicker the electrode, the lower the resistance and the lower the voltage applied at that point. Such a thickness structuring can be realized, for example, by grayscale lithography of a photoresist with subsequent etching to transfer the structure into the electrode material. In principle, a method of nanoprint lithography is also conceivable.

7b shows an alternative electrode shape, wherein perpendicular to the propagation of the light (perpendicular to the plane), for example, from left to right (and thus in their width b) the layer electrode 8f or 8e is structured. 7b shows a plan view of the optical filter, in the propagation direction of the incident light. Here are the detector elements 19 arranged equidistantly and are also through a lower layer electrode 8e at least partially covered. The upper layer electrode 8f However, in this embodiment is designed such that it in the direction of the detector line 19 continuously increasing in width (eg growing from a first width b1 to a second width b2). The width increase may be, for example, a non-linear radio tion and be chosen such that the nonlinear dispersion effect is compensated. Thus, the shows 7b a width variation of the electrode suitable for dispersion compensation.

As in the embodiment described above with variable Layer thickness d leads the variable layer width to one variable sheet resistance and thus to a variable voltage drop. A similar effect occurs here. Where the electrode is is narrow (eg on the left with a width b1) the electrical resistance correspondingly high, so that the voltage drop is higher there and the geometric layer thickness with it becomes lower, which compensates for a high refractive index n. Corresponding is the electrical resistance on the right side due to the larger width b2 of the electrode lower and so there is less tension there, which leads to that the geometric layer thickness is larger which compensates for a correspondingly lower refractive index n becomes.

By a suitable choice of the potentials can thus be an adaptation of the layer thickness and at the same time a tilt can be achieved so that both central wavelength as well as the wavelength range set flexibly selected can be.

Will instead of the detector line 9 If a transparent substrate (eg glass) is used, a pure adjustable filter can be produced, which can be introduced, for example, into an optical beam path.

8th shows an embodiment of a two-dimensional ALOVF in supervision. The cross-sectional views of 3 to 6 are for example along the cutting line 17 shown. In the embodiment of 8th By way of example, M = 4 mutually independent ALVOFs are arranged parallel to one another on a common substrate. The top layer in supervision 16 generally corresponds to the mirroring layer of the upper mirror 1a , 8th thus shows a first optical filter 10a , a second optical filter 10b , a third optical filter and a fourth optical filter 10d which are arranged vertically to each other. In horizontal direction along the cutting line 17 are the individual detector elements 19 arranged.

With this component it is possible, for example, to examine M different light sources in parallel. This can be, for example, the first and the second partially transparent mirror 1a and 1b extend over the entire two-dimensionally formed surface and have a uniform tilt in the horizontal direction (along the section line 17 ). The vertical direction then provides a spatial resolution. Thus, the illustrated filter is sensitive to the same central wavelength with the same resolution - but can determine a position along the vertical direction. For example It is thus possible, a flowing past or passing medium perpendicular to the current direction (eg, parallel to the cutting line 17 ) spectrally analyze. In the execution as in 8th Thus, the tilt from right to left for all M ALOVFs uniformly designed (or the tilting mirror is made of a part and there are not separate electrodes for each line). From right to left, the spectral information is thus captured. From top to bottom the spectral information as a function of the place. Thus, the element can be used as a one-dimensional spatially resolving spectral sensor. In combination with a one-dimensional displacement device (eg parallel to the cutting line 17 ) or above z. B. a conveyor belt thus two-dimensional, spectrally resolved images can be recorded.

Alternatively, the incident light can be widened so that all M ALVOFs are irradiated simultaneously. Each of the ALVOFs can then be used for a different wavelength range and / or a different central wavelength. For example, one ALVOF can be used for a wide spectral range and the others for detection within selected narrow spectral ranges. Also, different detectors that are optimized for specific wavelengths can be used. For example, the first and second semitransparent mirrors can be used for this purpose 1a and 1b for each of the four optical filters 10a - 10d be tilted separately. This makes it possible, for example, that the first optical filter 10a For example, performs a coarse frequency resolution, while the second optical filter 10b already provides an improved spectral resolution. The third and fourth optical filters 10c and 10d can then provide detail resolutions. This is possible for example because the first optical filter 10a has a larger tilt than the second optical filter 10b and the third and fourth optical filters 10c , d have a smaller tilt than that of the first and second optical filters.

In the embodiments as they are for example in the 3 . 4 . 5 , and 6 As shown, an integrated position sensor can be used, which allows the distance and the tilt angle of the mirror plates 1a . 1b to determine. This sensor can, for. B. from an integrated light source, which provides a reference light of known intensity, in combination with a photodiode or a photodiode array. For example, in the case of a single photodiode, the structure may be chosen so that the light source, as an LED or laser diode (both having a finite divergence), radiates from below onto the tilted mirror plate. The beam reflected from the tilted mirror plate strikes the photodiode, with a certain intensity striking the photodiode, depending on the tilt. The intensity is thus a measure of the tilt. When not tilted mirror plate, the intensity changes due to the finite divergence of the light source gelplatten as a function of the distance of the two Spie. Thus, for example, by means of the position sensors, first the distance and then the tilt can be adjusted. By means of a photodiode array, the determination of the tilt and the distance can be carried out in the same way, but with higher accuracy.

In principle, the actual detector line can also be used 9 used to determine the position of the mirrors. For this purpose, for example, the light source used for position determination is turned off when with the detector line 9 a measurement should be performed.

Alternatively, the light (reference light) for position determination can also be coupled from the outside. If the wavelength, or the wavelengths are known, so can over the detector line 9 directly a calibration can be performed.

Tilt angle and distance can also be detected capacitively. For this purpose, the capacity determined by a pair of electrodes (left and right in the component) is determined in each case. 8a , b or 8c , d), wherein one electrode is connected to the fixed mirror and the other electrode is connected to the tiltable mirror. From the capacity of the right electrode can be closed on the distance of the mirror plates right, from the capacity of the left electrode to the distance left. From the difference, the tilt angle can be determined.

In further embodiments, the tilt / deflection is caused by electrostatic forces. For example, the mentioned capacitive position selection can also be used as an actuator. When applying a suitably high voltage, the mirror plates can be adjusted to each other in the distance and tilt angle. One possible construction would be, for example, a deformable spacer 7 comprise, having a spring with a spring restoring force, so that the electrostatic force tical overcomes the spring restoring force. Alternatively, instead of the spring, another elastically deformable material (eg rubber, gel, foam or the like) can be used.

9I to 9VI show process steps based on cross-sectional drawings for the production of an optical filter according to embodiments.

In 9I initially becomes a transparent substrate 11 , which may for example comprise glass. On the transparent substrate 11 is on one or, as shown, on both gegenü Overlying pages a first optical remuneration 12a and a second optical remuneration 12b applied, which provides the desired reflection properties. The result is in 9II shown. The coating may consist, for example, of a thin metal layer (Au, Ag, Al,...), Of a dielectric layer stack (Bragg reflector) or of an antireflection coating.

As a next step, as in the 9III shown the first electrode 8a in a first contact area on the first optical remuneration 12a applied and analogously, the third electrode 8c in a second contact area on the first compensation 12a applied. For example, a conductive layer can be applied over the entire surface and structured in the form of electrodes.

The next step will be, as in the 9IV shown the first deformable spacer 7a on the first electrode 8a applied and the second deformable spacer 7b goes to the third electrode 8c applied. This can in turn be achieved by applying the actuator layer (eg, made of electroactive material) over the entire surface and also structuring it so that the deformable spacers are then formed.

9V shows how on the first spacer 7a the second electrode 8b is formed, and at the same time on the second deformable spacer 7b the fourth electrode 8d is trained. In a subsequent step, in the 9VI is shown is on the second electrode 8b and the fourth electrode 8d Again, a transparent substrate with optical compensation, as in the 9II shown is applied. The transparent substrate may optionally have only a one-sided or, as shown, a two-sided optical coating and may be applied to the device according to the state of 9V be bonded. Thus, the substrate provides 11 with the optical remuneration 12a and 12b as it is in the 9II shown is the first semitransparent mirror 1a and the substrate with the optical compensation, as in the 9VI on the second and fourth electrodes 8b . 8d is formed, the first semitransparent mirror 1a represents.

Among other of the many production alternatives, the following may be mentioned: instead of individual structuring of the electrodes 8th and the actuator structure 7 At first all three layers (see 9III to 9V ) are applied over the entire surface and then in a process step (which may consist of several etching steps) are structured. The advantage is that all layers on a flat surface, ie without steps, can be applied, which offers great advantages especially for spin-on.

As a general rule the manufacturing process can also be done in the disk composite, which significant cost savings due to parallel litigation result, as per disk (wafer) more of these components can be made in parallel. Instead of a disc bonding process However, for the application of the upper mirror can also a pick-and-place procedure be used, which is a higher Flexibility for small quantities allowed.

additionally to the described layers it may be necessary to further To insert layers. For example, it may be necessary to use insulator layers to use or special layers, which the adhesion of layers to each other or from layers on one of the two mirrors.

process steps for producing or increasing the adhesion between individual Layers and structures can be, for example, conventional tempering methods, anodic, eutectic or glass solder bonding methods.

As well may in other embodiments, the wiring to the outside. This can be z. B. with the help of so-called Vias occur by means of special openings in the mirrors, to allow wire bonding, or by leading out the contacts - also on stages - by additional interconnects.

Further Embodiments also include combinations of ALVOF with a shutter. The shutter is used to set the integration time of the signal at the detector / detector elements. This is in the Spectroscopy especially for lock-in noise reduction used. The shutter can be used as external device (eg impeller) be realized or via pulsed readout of the detector / Detector elements.

QUOTES INCLUDE IN THE DESCRIPTION

This list The documents listed by the applicant have been automated generated and is solely for better information recorded by the reader. The list is not part of the German Patent or utility model application. The DPMA takes over no liability for any errors or omissions.

Cited patent literature

• - US 6909548 B2 [0005]
• US 2002/0131047 A1 [0006]

Cited non-patent literature

• A. Emadi, H. Wu, S. Grabarnik, G. de Graaf and RF Wollfenbuttel, FABRICATION OF TAPERED OPTICAL STRUCTURES USING RESIST REFLOW, Proc. of 19th MicroMechanics Europe Workshop, Aachen (2008) [0007]

Claims (23)

Optical filter ( 10 ) comprising: a first partially transparent mirror ( 1a ); a second partially transparent mirror ( 1b ); and a deformable spacer ( 7 ; 15 ) between the first partially transparent mirror ( 1a ) and the second partially transparent mirror ( 1b ), wherein a distance ( 20 ) between the first partially transparent mirror ( 1a ) and the second partially transparent mirror ( 1b ) as a function of the optical filter ( 10 ) applied voltage (U) is variable. Optical filter ( 10 ) according to claim 1, wherein the deformable spacer ( 7 ; 15 ) an electroactive material that deforms at the applied voltage (U). Optical filter ( 10 ) Claim 2, wherein the electroactive material either dielectric, piezoelectric, electrostrictive, electro-chemo-mechanical properties or consists of an ionic polymer-metal composite, a gel, a liquid crystal elastomer, a shape memory polymer or an alloy. Optical filter ( 10 ) according to claim 2 or claim 3, wherein the electroactive material consists of polymers, copolymers, tripolymers, and polymer blends. Optical filter ( 10 ) according to any one of claims 2 to 4, wherein the electroactive material is a ceramic, a polymer-ceramic composite or a nanocomposite. Optical filter ( 10 ) according to any one of claims 2 to 5, wherein the electroactive material is a polymer nanocomposite. Optical filter ( 10 ) according to one of claims 2 to 6, further comprising a first electrode ( 8a ), a second electrode ( 8b ), a third electrode ( 8c ) and a fourth electrode ( 8d ) and in which the deformable spacer ( 7 ) a first spacer ( 7a ) and a second spacer ( 7b ), wherein the first electrode ( 8a ) between the first spacer ( 7a ) and the second partially transparent mirror ( 1b ) and the second electrode between the first spacer ( 7a ) and the first partially transparent mirror ( 1a ) and a first voltage between the first and second electrodes ( 8a . 8b ) can be applied, and wherein the third electrode ( 8c ) between the second spacer ( 7b ) and the second partially transparent mirror ( 1b ) and the fourth electrode ( 8d ) between the second spacer ( 7b ) and the first partially transparent mirror ( 1a ) and a second voltage between the third and fourth electrodes ( 8c . 8d ) can be applied, so that changes of the first and second voltage to a tilt and / or a shift of the first partially transparent mirror ( 1a ) relative to the second partially transparent mirror ( 1b ) leads. Optical filter ( 10 ) according to claim 7, further comprising a position sensor and the position sensor is adapted to detect from a capacitive measurement between the first and second electrodes ( 8a . 8b ) and / or another capacitive measurement between the third and fourth electrodes ( 8c . 8d ) the distance ( 20 ) and / or a tilt of the first relative to the second partially transparent mirror ( 1a . 1b ) to investigate. Optical filter ( 10 ) according to one of claims 2 to 6, which is a spectral filtering of incident light ( 3 ) and in which the electroactive material ( 15 ) between the first and second partially transparent mirrors ( 1a . 1b ) is formed so that the incident light ( 3 ) passes the electroactive material before spectral filtering. Optical filter ( 10 ) according to claim 9, wherein the deformable spacer ( 15 ) a first electrode layer ( 8e ) and a second electrode layer ( 8f ), which are separated by an actuator layer of electroactive material, wherein the first electrode layer ( 8e ) along the second partially transparent mirror ( 1b ) and the second electrode layer ( 8f ) along the first semitransparent mirror ( 1a ) are formed and the voltage (U) between the first and the second electrode layer ( 8f . 8e ) can be applied. Optical filter ( 10 ) according to claim 10, wherein the first electrode layer ( 8e ) has a first terminal for a first potential (U1), the second electrode layer ( 8f ) has a second terminal for a second potential (U2) and a third terminal a third potential (U3), and in which the second electrode layer ( 8f ), so that a difference between the second and third potentials (U2, U3) is a potential gradient in the second electrode layer (FIG. 8e ) and thus a tilting of the first relative to the second partially transparent mirror ( 1a . 1b ) causes. Optical filter ( 10 ) according to claim 10 or claim 11, wherein the actuator layer ( 15 ) is wedge-shaped deformable. Optical filter ( 10 ) according to one of claims 10 to 12, in which the first or second electrode layer ( 8e . 8f ) are structured in order to control the dispersion in the actuator layer ( 15 ). Optical filter ( 10 ) according to one of the preceding claims, in which the deformable spacer ( 7 ; 15 ) has an elastically deformable material and the optical filter ( 10 ) Contact areas ( 8th ), so that the applied voltage (U) causes an electrostatic attraction of the contact regions ( 8th ) and thus to a change in the distance ( 20 ) leads. Optical filter ( 10 ) according to claim 14, wherein the elastically deformable material comprises a spring. Optical filter ( 10 ) according to one of the preceding claims, in which the deformable spacer ( 7 ; 15 ) is formed so that the applied voltage (U) to a parallel displacement of the first partially transparent mirror ( 1a ) relative to the second partially transparent mirror ( 1b ) leads. Optical filter ( 10 ) according to one of the preceding claims, in which the deformable spacer ( 7 ; 15 ) is formed, so that the applied voltage (U) to a tilt ( 20v ) of the first semitransparent mirror ( 1a ) relative to the second partially transparent mirror ( 1b ) leads. Optical filter ( 10 ) according to any one of the preceding claims, further comprising an additional spacer ( 13 ) and the additional spacer ( 13 ) is adapted to form a Vorverkippung without a voltage (U) is applied. Optical filter ( 10 ) according to one of the preceding claims, further comprising a position sensor and the position sensor is adapted to detect from an intensity of a between the first and second partially transparent mirror ( 1a . 1b ) reflected reference light the distance ( 20 ) and / or a tilt of the first relative to the second partially transparent mirror ( 1a . 1b ) to investigate. Method for producing an optical filter ( 10 ), comprising the following steps: forming a first semitransparent mirror ( 1a ); Forming a second semitransparent mirror ( 1b ); Forming a deformable spacer ( 7 ; 15 ) between the first and second partially transparent mirrors ( 1a . 1b ), so that in dependence on the optical filter ( 10 ) applied voltage (U) is a distance ( 20 ) between the first and second partially transparent mirrors ( 1a . 1b ) is changed. The method of claim 20, wherein forming the deformable spacer comprises forming a first spacer (10). 7a ), a second spacer ( 7b ) and an additional spacer ( 13 ), which pre-ticks between the first and second partially transparent mirrors ( 1a . 1b ), and the method further comprises applying an adhesion-promoting layer ( 14 ) between the first partially transparent mirror ( 1a ) on the one hand and the first or the second spacer ( 7a . 7b ), wherein the adhesion-promoting layer ( 14 ) is formed, so that when heating the first partially transparent mirror ( 1a ) to the spacer ( 7 ) is pulled. A method according to claim 20 or claim 21, thus produced that realizes plane-parallel displacement and tilting become. Method according to one of Claims 20 to 22, in which a wedge-shaped deformable layer ( 15 ) is applied by spin coating, deposition or lamination.