Micromachined top-hat wavelength tuning filter

Abstract

The present application discloses a wavelength-tunable optical filter having a top-hat passband shape. The wavelength-tunable optical filter is adapted to be mounted on a top surface of a sensor and includes a set of at least three stacked dielectric mirrors adjacent and parallel to the sensor surface, the at least three stacked dielectric mirrors including a top mirror, a middle mirror, and a bottom mirror, the set of at least three stacked mirrors being spaced apart within a variable distance range to form at least two independent, adjacent, and equal-volume cavities between the top mirror and the middle mirror, and between the middle mirror and the bottom mirror.

Description

Detailed Description of the Invention

[0001] [Technical Field] The present application discloses a wavelength-tunable optical filter having a top-hat passband shape.

[0002] [Background technology] Currently, existing solutions include the use of low-angle shift (LAS) optical filters for LiDAR applications. These filters have an expanded angle of incidence range but are static filters that cannot be wavelength-tuned. In particular, document US2013329232A discloses a controllable Fabry-Perot interferometer fabricated with microelectromechanical system (MEMS) technology. The interferometer arrangement shown includes both an electrically tunable interferometer and a reference interferometer on the same substrate. Temperature drift is measured with the reference interferometer, and this information is used to correct the measurements made by the tunable filter interferometer. This improves measurement accuracy and stability and reduces packaging requirements.

[0003] To date, there has been some research and development in the field of wavelength-tunable Fabry-Perot optical filters, with a particular focus on portable spectroscopy applications and spectroscopic imaging. The developments demonstrated in these MEMS devices typically have a circular membrane with an annular ring around the membrane's edge for actuation. While this device achieves a large area for a MEMS filter, it suffers from the drawback of a very narrow angular range due to being a single-cavity Fabry-Perot filter.

[0004] In particular, this work focuses on the key findings of Avinash Parashar et al.'s "Three-cavity Tunable MEMS Fabry-Perot interferometer," which investigates the behavior of equally spaced cavities. By increasing or decreasing the mirror spacing by the same amount, these cavities can be tuned to the desired passband. Furthermore, document DE102019212986 discloses a Fabry-Perot interferometer (1a; 1b; 1c) comprising a first mirror (2a; 2b; 2c) and a second mirror (3a; 3b; 3c) spaced apart from the first mirror (2a; 2b; 2c) so that the first mirror (2a; 2b; 2c) and the second mirror (3a; 3b; 3c) form an optical cavity, and the first mirror (2a; 2b; 2c) and / or the second mirror (3a; 3b; 3c) comprises a plurality of layers (21a-25a; 31a-35a), the thicknesses of which are selected such that the optical path lengths corresponding to the thicknesses of the layers (21a-25a; 31a-35a) differ at least in part from one another.

[0005] 〔overview〕 The present disclosure discloses a wavelength tuning optical filter adapted to be placed on an upper surface of a sensor, the wavelength tuning optical filter including a set of at least three stacked dielectric mirrors adjacent and parallel to a surface on which the sensor is placed, the at least three stacked dielectric mirrors including a top mirror, an intermediate mirror, and a bottom mirror, the set of at least three stacked mirrors being spaced apart within a variable distance range to form at least two cavities, independent, adjacent, and having equal volumes, between the top mirror and the intermediate mirror, and between the intermediate mirror and the bottom mirror.

[0006] In the embodiment shown in this disclosure, the intermediate mirror is fixed in a plane perpendicular to the plane on which the sensor is mounted.

[0007] Furthermore, in another embodiment shown in the present disclosure, the top and bottom mirrors are movable in a plane perpendicular to the plane on which the sensor is mounted.

[0008] Furthermore, in another embodiment shown in the present disclosure, the movement of the top and bottom mirrors is actuated by actuation electrodes to ensure equal spacing between the top and middle mirrors, and between the middle and bottom mirrors.

[0009] Furthermore, in another embodiment shown in the present disclosure, the middle mirror has a higher reflectivity than the top and bottom mirrors.

[0010] Furthermore, in another embodiment shown in the present disclosure, the middle mirror has a lower reflectivity than the top and bottom mirrors.

[0011] Furthermore, in another embodiment shown in the present disclosure, the upper and lower mirrors are supported by independent inner movable frames, which are secured to an independent outer fixed frame by independent sets of springs.

[0012] Additionally, in another embodiment shown in the present disclosure, the peripheral gap between the inner movable frame and the outer fixed frame is filled with multiple sets of electrodes consisting of actuation electrodes and detection electrodes.

[0013] Furthermore, in another embodiment shown in the present disclosure, the actuation electrodes comprise a set of electrodes having asymmetric vertical heights and minimized lateral gaps between them, said electrodes being sequentially fixed relative to an outer fixed frame and an inner movable frame.

[0014] Furthermore, in another embodiment shown in the present disclosure, a plurality of actuating electrodes fixed to the inner movable frame share the same potential, and a plurality of actuating electrodes fixed to the outer fixed frame also share the same potential, and a potential difference is applied between the actuating electrodes fixed to the outer fixed frame.

[0015] Furthermore, in another embodiment shown in the present disclosure, the detection electrode has the same shape as the actuation electrode and is positioned perpendicularly alongside the actuation electrode.

[0016] Furthermore, in another embodiment shown in the present disclosure, the nominal distance between the upper mirror structure, the intermediate mirror structure, and the lower mirror structure is ensured by a set of spacers that facilitate mechanical support between the outer fixed frame of the lower mirror and the inner movable frame of the intermediate mirror, and between the inner movable frame of the intermediate mirror and the outer fixed frame of the upper mirror.

[0017] Furthermore, in another embodiment shown in the present disclosure, the outer surface of the bottom mirror and the outer surface of the top mirror, which are both outside the two adjacent cavities having equal volumes, include an anti-reflective coating.

[0018] [Summary of the Disclosure] This application discloses wavelength-tuning optical filters optimized for use in LiDAR systems in certain applications. The present disclosure improves upon the performance of the prior art, particularly by improving the acceptance angle, which is important for use in LiDAR systems, and by providing a rectangular aperture that can be adapted to the detector array.

[0019] Controlling and monitoring the operating wavelength is crucial in LiDAR systems and applications. In a LiDAR system's receiving unit (Rx), all radiation outside the operating wavelength must be filtered to maximize the signal-to-noise ratio (SNR). The light emitted from a LiDAR system's transmitting unit (Tx) must remain within the passband of the receiving unit (Rx) to avoid signal blocking by the filters embedded therein. Therefore, the full width at half maximum (FWHM) of these filters—the difference between the two values ​​of the independent variable where the dependent variable is equal to half of its maximum value—must be large enough to accommodate drift in the emitter wavelength due to temperature fluctuations and manufacturing tolerances in both the emitter and the filter. This makes the filter's FWHM very large compared to the laser's bandwidth, reducing the system's range because excess background light passing through the filter is reduced by the SNR.

[0020] For specific applications in automotive LiDAR systems, passband filters must have narrow bandwidths, large acceptance angles (low angular shifts), and high transmittance at the pass wavelengths. They must also have sufficient filter area; depending on the LiDAR structure, a large rectangular area is preferred for most designs. For automotive applications, the device must operate over a wide temperature range and be highly reliable. Static filters are useful for low angular shifts, but they must have wider passbands than desired to ensure the emission wavelength overlaps with the filter's passband due to manufacturing tolerances and temperature-induced laser wavelength shifts. This reduces the SNR and therefore the LiDAR's maximum range. This problem can be solved on the emitter side through the use of lasers with very precise emission wavelengths, but this is costly and lasers with the required performance are not widely available. Alternatively, on the receiver side, wavelength-tunable optical filters can be used to solve this problem, but existing technology does not offer the required performance.

[0021] While individual solutions to some of the identified problems may exist, there is currently no specific solution that can simultaneously solve them. For example, liquid crystal tuning filters are available with wide apertures and acceptance angles, but they cannot meet the requirement of a wide operating temperature range. MEMS Fabry-Perot (FP) tuning filters also have some drawbacks. In particular, they suffer from high angular shift, an inherent problem of having a single-cavity filter. This is due to the Lorentzian shape of the spectral passband, which exhibits a maximum at a very precise wavelength. As the angle of incidence of light changes, the peak wavelength of the passband shifts. With a Lorentzian shape, even a shift smaller than the FWHM significantly reduces the transmittance at the target wavelength.

[0022] A further issue with existing MEMS FP tuning filter designs is their aperture area and shape. The goal for these filters is to use this technology on detector chips. To be used in conjunction with detector arrays (compared to small, single-point detectors), the filters must have a relatively large area and ideally match the shape of the detector. Such detectors are typically rectangular, often with one dimension significantly longer than the other in the case of time-of-flight detectors. To create symmetrical actuation and thus maintain parallelism between the mirrors, most existing tuning filters are circular. However, such circular filters are not suitable because the surface area required for the membrane is too large, making it infeasible to fabricate a circular filter large enough for a long, rectangular detector using MEMS technology. However, because the detector surface area is relatively small, it is feasible to fabricate a filter with a matching shape.

[0023] These limitations can therefore be improved if a filter with a top-hat passband can be created, as shown and disclosed in this disclosure. The technical solution relies on the use of a set of mirrors with wavelength-dependent phase-of-reflection (PUR) that enhances the spectral stability at different angles.

[0024] This is the basis of the optical concept of this disclosure, but this disclosure establishes it by modifying the optical design to improve the angle of incidence (AOI) response and providing a practical MEMS device implementation designed for LiDAR applications. Other methods demonstrated to improve the angular acceptance of these filters focus solely on adding phase to the mirror upon reflection. However, despite significant improvements, the acceptance angle still falls short of the required value. Furthermore, the proposed modifications effectively introduce a trade-off between the required operating range and the acceptance angle, meaning that even a relatively small change in the passband requires a large operating range.

[0025] Thus, the presented wavelength tuning filter provides the following technical features: low angular shift (within the passband wavelength), large area rectangular shape, top hat passband shape, narrow passband, high reliability, and wide operating temperature range, representing a significant improvement with respect to the current state of the art.

[0026] As a result, the wavelength-tunable top-hat profile filter using a phase-up-reflection (PUR) mirror has the following advantages over existing MEMS wavelength-tunable filters: 1. Reduced dependency on angle of incidence (AOI). 2. The spectral shape of the resulting passband filter resembles a top-hat shape rather than a Lorentzian shape. 3. The possibility of large-area devices with electrode actuation mechanisms strong enough to move large-area mirror membranes. 4. High wavelength accuracy: The electrodes for detecting the position of the large-area mirror membrane area can accurately control the plate distance. 5. High uniformity over a large area: Detection electrodes and multi-channel control allow correction of plate parallelism. 6. Narrow passband. 7.High transmittance. 8. Non-induced effect: improved electromechanical reliability.

[0027] The use of the shown wavelength-tunable filters (i.e., filters whose pass frequencies can be actively controlled) instead of the static filters shown earlier provides the opportunity to match the passband of the Rx according to drifts occurring in the Tx.

[0028] To achieve the above results and overcome technical limitations, the present disclosure discloses a dual-cavity MEMS wavelength tuning filter. This tuning filter was developed using a Fabry-Perot (FP) interferometer consisting of two highly reflective parallel mirrors spaced a fixed distance apart. The distance between the two mirrors determines the passband wavelength of the cavity. Therefore, a wavelength tuning filter can be obtained by actively varying the cavity thickness (the distance between the mirrors). Architecturally, the wavelength tuning filter device developed and disclosed herein consists of a stationary mirror surrounded by two movable mirrors that actuate to maintain equal cavity spacing between the stationary and movable mirrors. In one preferred embodiment of the present disclosure, the mirrors are modified Bragg reflectors, and the modification introduces a phase shift upon reflection of the reflected light. The reflectivity of the three combined mirror arrangements described above is designed to provide a top-hat shape for the filter's passband. The developed device is preferably rectangular, with actuation and detection along its edges.

[0029] BRIEF DESCRIPTION OF THE DRAWINGS For a better understanding of the present application, figures illustrating preferred embodiments are attached herewith, which are not intended to limit the technology disclosed herein.

[0030] FIG. 1 shows a possible embodiment of a multi-mirror Fabry-Perot (FP) cavity.

[0031] Figure 2 shows an example of a three-mirror FP cavity with equal mirror spacing (d). Numerical references are as follows: IS - interferometer spectrum of the two-cavity FP filter; T - transmittance; WL - wavelength; d1 - distance between mirrors, 300 nm; d2 - distance between mirrors, 400 nm; d3 - distance between mirrors, 500 nm; d4 - distance between mirrors, 600 nm.

[0032] 3 illustrates an embodiment of the present disclosure relating to the structure of the wavelength-tuning dual-cavity FP filter (10) of the present disclosure, mounted on top of a sensor (30). In the preferred embodiment shown, the three stacked mirrors—top mirror (101), middle mirror (102), and bottom mirror (103)—are equally spaced, since the distance (d5) between the top mirror (101) and middle mirror (102) is the same as the distance (d6) between the middle mirror (102) and bottom mirror (103), resulting in a wavelength-dependent PUR (dark gray surface). In one of the examples shown, incident light (20) traverses the filter (10) in the direction indicated by the arrow toward the sensor (30).

[0033] FIG. 4 shows the transmission spectrum of the dual cavity Fabry-Perot (DCFP) filter of the present disclosure when the reflectivity of the intermediate mirror (102) and the outer mirrors (101, 103) is varied, with the reflectivity of the inner mirror (sx) being varied and the reflectivity of the outer mirror (dx) being varied.

[0034] Figure 5 shows a possible embodiment of the top mirror (101) dielectric structure. Relevant references are as follows: T1 - dielectric thickness (nm), distance d10 - 414.0 nm, distance d11 - 270.0 nm, distance d12 - 207.0 nm, distance d13 - 270.0 nm, distance d14 - 207.0 nm, distance d15 - 270.0 nm, distance d16 - 207.0 nm, distance d17 - 270.0 nm, distance d18 - 207.0 nm; SiN - silicon nitride dielectric layer; SiO2 - silicon dioxide dielectric layer. Additionally, the top mirror transmission and reflection spectra (UM TSR) and the reflection phase (UM PUR) of the top dielectric mirror (101) are shown. The unit of wavelength (WL) is nm, and the numerical references are as follows: 1011-R: p-pol - reflectance p-polarized or parallel polarized light; 1012-T: p-pol - transmittance p-polarized or parallel polarized light; 1013-p-pol - p-polarized or parallel polarized light.

[0035] Figure 6 shows a possible embodiment of the lower mirror (103) dielectric structure. Relevant references are as follows: T3 - dielectric thickness (nm), distance d30 - 414.0 nm, distance d31 - 270.0 nm, distance d32 - 207.0 nm, distance d33 - 270.0 nm, distance d34 - 207.0 nm, distance d35 - 270.0 nm, distance d36 - 207.0 nm, distance d37 - 270.0 nm, distance d38 - 207.0 nm; SiN - silicon nitride dielectric layer; SiO2 - silicon dioxide dielectric layer. Additionally, the lower mirror transmission and reflection spectra (LM TSR) and reflection phase (LM PUR) of the lower dielectric mirror (103) are shown. The unit of wavelength (WL) is nm, and the numerical references are as follows: 1031-R: p-pol - reflectance p-polarized or parallel polarized light; 1032-T: p-pol - transmittance p-polarized or parallel polarized light; 1033-p-pol - p-polarized or parallel polarized light.

[0036] Figure 7 shows a possible embodiment of the intermediate mirror (102) dielectric structure. Relevant references are as follows: T2 - dielectric thickness (nm); distances d200 and d220 - 242.0 nm; distances d201, d203, d205, d207, d209, d211, d213, d215, d217, and d219 - 158.0 nm; distances d202, d204, d206, d208, d210, d212, d214, d216, and d218 - 121.0 nm; SiN - silicon nitride dielectric layer; SiO2 - silicon dioxide dielectric layer. Additionally, the intermediate mirror transmission and reflection spectra (IM TSR) and reflection phase (IM PUR) of the intermediate dielectric mirror (102) are shown. The unit of wavelength (WL) is nm, and the numerical references are as follows: 1021-R: p-pol-reflectance p-polarized or parallel polarized light; 1022-T: p-pol-transmittance p-polarized or parallel polarized light; 1023-p-pol-p-polarized or parallel polarized light.

[0037] Figure 8 shows a possible embodiment of the dielectric structure of the shown double-cavity Fabry-Perot filter (10). Relevant references are as follows: T - total dielectric thickness (nm), distance d40, and d41 - 452.0 nm, the rest relate to the dimensions of the aforementioned mirrors (101, 102, 103). SiN - silicon nitride dielectric layer; SiO2 - silicon dioxide dielectric layer.

[0038] Additionally, the dual-cavity Fabry-Perot transmission and reflection (DCFP TSR) spectra and the phase in reflection (DCFP PUR) of the filter (10) are shown. Wavelengths (WL) are expressed in nm, and numerical references relate to the following: 1041-R: p-pol - reflectivity p-polarized or parallel polarization; 1042-T: p-pol - transmittance p-polarized or parallel polarization; 1043-p-pol - p-polarized or parallel polarization.

[0039] Figure 9 shows the transmission spectrum of a double-cavity Fabry-Perot (DCFP TS) filter for different angles of incidence on the filter in the wavelength range (WL (nm)). The relevant references are as follows: 1051 - incident angle 0°; 1052 - incident angle 10°; 1053 - incident angle 20°.

[0040] Figure 10 shows the transmission spectrum (TS) of the filter (10) as a function of the air gap separating the mirrors, i.e., the air distance between the mirrors, for different angles of incidence (AOI) of the incident light (20). Relevant references are: 1051 - 0° angle of incidence, 1052 - 10° angle of incidence, 1053 - 20° angle of incidence.

[0041] 11 is a top view showing the structure of the filter (10) with electrode actuation. Reference numbers are as follows: 101 - upper dielectric mirror, 201 - silicon outer frame, 202 - sensing electrode, 203 - spring, 204 - silicon inner frame, 205 - actuation electrode.

[0042] Figure 12 shows a cross-sectional view of the assembly of the filter (10) shown. In one preferred embodiment of the assembly of the filter (10), the reference numbers are as follows: 101 - upper dielectric mirror; 102 - middle dielectric mirror; 103 - lower dielectric mirror; 201 - silicon outer frame; 203 - spring; 204 - silicon inner frame; 206 - spacer.

[0043] The illustrated embodiment of the filter assembly uses two structures with the backside of the dielectric structure removed.

[0044] 13 shows a cross-sectional view of a second assembly of the filter (10) shown. In one preferred embodiment of the assembly of the filter (10), the reference numerals are as follows: 101 - upper dielectric mirror; 102 - middle dielectric mirror; 103 - lower dielectric mirror, 201 - silicon outer frame, 203 - springs, 204 - silicon inner frame, 206 - spacers, 207 - anti-reflection coating. In the shown embodiment of the filter assembly, the back surface of the filter is not removed, and an anti-reflection (AR) coating is applied to these surfaces.

[0045] 14 shows a third cross-sectional view of the structure of the filter (10). Reference numerals are as follows: 101 - upper dielectric mirror, 201 - silicon outer frame, 204 - silicon inner frame, 205 - detection electrode. In one embodiment, a predetermined voltage V is applied between the fixed electrode frame (204), the movable electrode (203) and the movable silicon substrate (201) frame.

[0046] [Description of the embodiment] Next, some embodiments will be described in more detail with reference to the figures, which, however, are not intended to limit the scope of the present application.

[0047] The present disclosure allows for the creation of top-hat filters, i.e., filters with a section exhibiting maximum transmittance instead of a single wavelength filter (e.g., an FP filter). Such filters can be achieved by dual-cavity FP filters. As shown simply in Figure 1, modeling of a multi-cavity FP filter involves the field reflectance r i (i=1...N-1) and transmittance t i (i=1···N-1) and are connected (sandwiched) by mirrors with refractive index n i We use a series of dielectric materials with intensity reflectance and transmittance R i =|r i | 2 and T i =|t i | 2 At each interface, the electric field that passes through the next dielectric material and the electric field that is reflected back by the same dielectric material are E i+ and E i- Figure 1 is an explanatory diagram. Each interface corresponds to a transfer matrix, and by multiplying all the matrices in a chain, the transfer matrix for the entire structure is obtained.

[0048] Figure 2 shows a specific example of a two-cavity filter (with mirror field reflectivities r_ext, r_int, r_ext = 0.87, 0.99, 0.87). While this shows a large spectral change, decreasing the mirror spacing (d) from d4 to d1 reduces the bandpass window width and center frequency (WL). This FWHM change can be barely noticeable in the region of interest (i.e., adjustments of less than 50 nm at key wavelengths such as 905 nm and 1550 nm).

[0049] The optical structure (10) of the two cavity filters disclosed herein consists of the following elements: three mirrors with matched reflectivity, namely a top mirror (101), a middle mirror (102) and a bottom mirror (103), as shown in Figure 3.

[0050] As shown in Figure 4, and in one of the disclosed embodiments, the two outer mirrors, the top mirror (101) and the bottom mirror (103), have the same reflectivity, but the inner one, the middle mirror (102), has a higher reflectivity. This results in: a. If the mirrors (101, 102, 103) are "matched" in terms of reflectivity, a top-hat shaped, flat transmittance profile is obtained; b. If the mirror reflectivities are "mismatched" in terms of reflectivity, such that the middle mirror (102) reflectivity is higher than the matched equilibrium of the remaining set of mirrors (101, 103), the profile will show a peak with sub-unity transmittance; c. If the reflectivity of the intermediate mirror (102) is lower than the equilibrium of the mirror set, the transmission spectrum will show two peaks with a non-zero valley between them. The depth of the valley increases with distance from the equilibrium point. At the same time, since the outer cutoff is sharper, allowing for a small central valley (e.g., less than 10% transmission drop) can achieve a sharper cutoff.

[0051] The use of two or more cavities can be used to obtain sharper cutoffs, increasing the squaring of the passband filter.

[0052] The structure of the dielectric mirror filter (10) is as follows: a. Outside mirrors (101, 103)-(HL) with symmetrical structures Next ×HH; b. Intermediate mirror (102)-H(HL) Nint ×HH

[0053] The middle mirror (102) is stationary, while the remaining two mirrors (101, 103) are actuated by out-of-plane MEMS actuated electrostatic electrode drive. In a stacked dielectric structure, two or more dielectric materials are used. In this case, focusing on the two dielectric materials, H denotes the material with a higher refractive index and L denotes the material with a lower refractive index. The thickness is expressed as a quarter wavelength in the material (λ / 4 / refractive index of the material). Thus, HL refers to the basic structure (stack) where a quarter wavelength thick "H" dielectric layer is followed by a quarter wavelength thick "L" dielectric layer. Furthermore, for short, if the structure is repeated, (HL) 3 Power elevation is used so that N = HLHLHL. ext and N int is the number of layers in the outer and inner mirrors. The mirror design can also look at thicknesses other than the optical quarter wavelength. In other embodiments, one can focus on dielectric thicknesses other than the quarter wavelength.

[0054] The mechanical structure (10) of the wavelength tuning optical filter described herein is formed by three mechanical structures, one of which is static (stationary, fixed) in the center (102), and the remaining two (101, 103) are MEMS actuated structures on each side (top and bottom) of the central structure (Figure 12). The central fixed structure has a static single membrane mirror (102), and the upper and lower structures also have their own mirror membranes (101, 103) supported by movable out-of-plane electrostatic mechanisms. In the middle region of the membranes, there is an overlap area that defines the optical aperture of the device (10), where an unobstructed optical path exists. The mirrors located on the upper structure (101) and lower structure (103) are positioned directly opposite the inner membrane (102) (meaning the upper structure device is inverted), reducing the initial distance between these elements. The actuation mechanism of this filter can be explained as follows: the mirror membranes (101, 103) are supported on a central movable frame (204) machined into a dielectric layer, such as the device layer of a silicon or SOI (Silicon on Insulator) wafer, which is fixed to an outer fixed frame (201) by a set of springs (203), as shown in FIG. 11. In the gap between the central movable frame (204) and the outer fixed frame (201), a set of mutually inductive electrodes (202, 205) is located. The structures of the electrodes (202, 205), the movable frame structure (204), and the fixed frame (201) are separate. To provide out-of-plane electrostatic actuation, the actuation electrode (205) structure has asymmetric vertical heights, minimizing the lateral gap between the electrodes fixed on the fixed frame (201) side and the electrodes (205) fixed on the inner movable frame structure (204) (FIG. 11). The actuation electrodes (205) fixed to the movable frame (204) can share the same potential, opposite to that of the electrodes connected to the fixed frame (201). Each set of electrodes (205) can be independently actuated to correct misalignment and improve parallelism. An additional set of electrodes can be used to provide sensing feedback for closed-loop operation.

[0055] The top and bottom mirror (101, 103) films can improve the overall transmission of the filter (10) by avoiding losses in the dielectric layer (201) that supports the mirror stack. This can be done by opening windows in the dielectric layer defined by the optical path region and applying an anti-reflection coating (207) on the dielectric layer on the side opposite the dielectric mirror.

[0056] All three structures must be assembled into a single filter device (10) while ensuring their correct relative positioning and maintaining the appropriate distance (height) between them (Figures 12 and 13). The nominal distance between the top mirror (101), middle mirror (102), and bottom mirror (103) structures is achieved using micromachined spacers (206). These spacers (206) set the nominal position and distance between the mirrors (101, 102, and 103) that form the filter (10) within the wavelength-tunable displacement range. The top, middle, and bottom elements (101, 102, and 103) are micromachined separately, avoiding the complexity of a monolithic process. The top and bottom mirror structures (101 and 102) can be evenly spaced, except for the top structure (101), which is flipped during assembly. The top and bottom structures (101 and 103) each have micromachined spacers (206) within the tolerance of the out-of-plane actuator adjustment range. Once all three parts are complete, they must be assembled in a specific order: a. positioning the lower element so that the inner movable structural frame (204) having the lower mirror (103) faces upward, i.e., facing away from the sensor (30); b. bonding the static intermediate mirror (102) structure to the lower spacer (206); c) placing the upper mirror element (101) inverted relative to the lower mirror (103) so that the inner movable frame (204) with the mirror (101) faces the static intermediate mirror membrane (102) and the sensor (30), with the spacers (206) of the upper element glued to the intermediate frame.

[0057] Alignment marks are used to align all components of the filter (10) and ensure that the light path is not obstructed.

[0058] The outside of both the upper and lower mirror elements (101, 103) of the filter (10) require electrical connections to apply electrostatic actuation to the actuation electrodes (205), but because they are electrically decoupled, these connections are one of the following: - a large flat surface applied to the inner electrode (connected to the inner frame (204)); - Different actuation surfaces for independently actuating different sets of electrodes (connected to the outer frame (201)) to achieve parallelism.

[0059] The sensing electrode (202) can be input to, for example, a capacitance to voltage (C / V) converter. Thus, based on the above, the sensing electrode (202) and the actuation electrode (205) can be made of the same material, but they behave differently in response to and respond to an electrical bias.

[0060] With respect to the mirrors shown in Figures 5, 6, and 7, one illustrated embodiment of the present disclosure illustrates the use of dielectric structures in the mirror set (101, 102, 103) constructed from a stacked combination of layers. This layer combination can be constructed from a wide range of materials. In one disclosed embodiment, it can also be constructed from a SiN / SiO2 stack, as shown throughout the figures. However, additional materials can be used, such as Ge (germanium), MgF2 (magnesium fluoride), Si (silicon (amorphous or crystalline)), AlN (aluminum nitride), Al2O3 (aluminum oxide), ITO (indium tin oxide), TiO2 (titanium dioxide), and ZnO (zinc oxide). Furthermore, stacked materials can be used in combination. It is important to ensure that the materials involved have negligible absorption and differing refractive indices in the operating wavelength range. The greater the difference in refractive index between adjacent materials, the fewer layers are required for each stack. However, design features require careful planning to achieve an optimal solution.

[0061] Returning to the actuation electrodes (205), they are configured to induce movement in the upper and lower mirrors (101, 103), which are supported by the inner frame (204) and further supported on the outer frame (201) by means of a set of springs (203). The detection electrodes (202) have the same shape as the actuation electrodes (205), but are arranged perpendicular to them. In the embodiment shown, the actuation electrodes (205) are arranged in at least two parts along the long sides of the filter (10), arranged between the outer and inner frames (201, 204) and between the springs (203) that ensure the stabilization of the mirrors (101, 103), said springs (203) being located in the center of the structure of the filter (10). The sensing electrode (202) is positioned on the shortest side of the filter (10), perpendicular to the actuation electrode (205), and is arranged in at least two sections between the outer and inner frames (201, 204) and between the springs (203), again to ensure stabilization of the mirrors (101, 103). In the illustrated embodiment, the positions and dimensions of the actuation and sensing electrodes (202, 205), the filter (10), and the mirrors (101, 102, 103) should not be understood as limiting features. The sensing electrode (202) is not voltage-driven; it merely probes capacitance to measure the position of the mirrors (101, 103). The springs (203) are mechanical elements that connect the floating mirror membranes (101, 103) to the silicon outer frame (201).

[0062] In both cases, the mirror membrane (101, 102, 103) is deposited on a silicon substrate identified as the inner frame (204). To improve light transmission, the silicon is etched down to the membrane; in the second case, the membrane is maintained but an anti-reflection coating is deposited on the other side of the substrate. This is an important choice depending on the target design wavelength. Silicon strongly absorbs at wavelengths of 905 nm, so at this target wavelength, the silicon substrate below the mirror must be etched away. For a wavelength of 1550 nm, silicon is essentially transparent, so the second option may be more mechanically robust. The anti-reflection coating (207) improves transmission. For 1550 nm, this leads to improved mechanical robustness.

[0063] In the illustrated embodiment of the filter assembly shown in Figure 13, the filter (10) includes an anti-reflection (AR) coating (207) on the rear / exterior surface of the bottom mirror (103) and on the front / exterior surface of the top mirror (101). As can be seen from this figure, the anti-reflection (AR) coating (207) is located on the outside of both cavities, i.e., the outside of the upper cavity formed between the top mirror and the middle mirror (101, 102) and the outside of the lower cavity formed between the middle mirror and the bottom mirror (102, 103). This anti-reflection coating (207) can improve the transmittance and mechanical robustness of the filter at a wavelength of 1550 nm.

[0064] In one embodiment shown in Figure 14, a voltage signal (V) is applied between a silicon inner frame (204), a movable actuation electrode (205), and a silicon outer frame (201). To maintain the distance that controls the mirror, a signal is applied to the actuation electrode (205), and the sense electrode (202) is used to probe the mirror position, which in a feedback loop mode can act on the input signal (V) to correct any misalignment. [Brief explanation of the drawings]

[0065] [Figure 1]1 shows a possible embodiment of a multi-mirror Fabry-Perot (FP) cavity. [Figure 2] An example of a three-mirror FP cavity with equal mirror spacing (d) is shown. [Figure 3] 1 shows an embodiment of the present disclosure regarding the structure of the wavelength-tuning dual-cavity FP filter (10) of the present disclosure placed on top of a sensor (30). [Figure 4] 1 shows the transmission spectrum of a dual-cavity Fabry-Perot (DCFP) filter of the present disclosure when the reflectivity of the intermediate mirror (102) and the outer mirrors (101, 103) is varied, with the reflectivity of the inner mirror (sx) being varied and the reflectivity of the outer mirror (dx) being varied. [Figure 5] Possible embodiments of the top mirror (101) dielectric structure are shown. [Figure 6] Possible embodiments of the bottom mirror (103) dielectric structure are shown. [Figure 7] Possible embodiments of the intermediate mirror (102) dielectric structure are shown. [Figure 8] 1 shows possible embodiments of the dielectric structure of the double cavity Fabry-Perot filter (10) shown. [Figure 9] 1 shows the transmission spectra of a double-cavity Fabry-Perot (DCFP TS) filter for different angles of incidence onto the filter in the wavelength range (WL (nm)). [Figure 10] 1 shows the transmission spectrum (TS) of the filter (10) as a function of the air gap separating the mirrors, i.e., the air distance between the mirrors, for different angles of incidence (AOI) of the incident light (20). [Figure 11] FIG. 1 is a top view showing the structure of a filter (10) with electrode actuation. [Figure 12] 1 shows a cross-sectional view of the assembly of the filter (10) shown. [Figure 13] 1 shows a second assembly cross-sectional view of the filter (10) shown. [Figure 14]A third cross-sectional view of the structure of the filter (10) is shown.

Claims

1. A wavelength tuning optical filter (10) adapted to be placed on top of a sensor (30), a set of at least three stacked dielectric mirrors adjacent to each other and parallel to a plane on which the sensor (30) is mounted; The at least three stacked dielectric mirrors include a top mirror (101), a middle mirror (102), and a bottom mirror (103); The set of at least three stacked mirrors is spaced apart within a variable distance range to form at least two independent, adjacent, and equal-volume cavities between the top mirror (101) and the intermediate mirror (102), and between the intermediate mirror (102) and the bottom mirror (103).

2. 2. The wavelength tuning optical filter (10) of claim 1, wherein the intermediate mirror (102) is fixed to a plane perpendicular to a plane on which the sensor (30) is mounted.

3. 2. The wavelength tuning optical filter (10) of claim 1, wherein the upper mirror (101) and the lower mirror (103) are movable in a plane perpendicular to a plane on which the sensor (30) is mounted.

4. 4. The wavelength tuning optical filter (10) of claim 1, wherein the operation of the upper mirror (101) and the lower mirror (103) is actuated by an actuation electrode (205) to ensure equal spacing between the upper mirror (101) and the intermediate mirror (102), and between the intermediate mirror (102) and the lower mirror (103).

5. The wavelength tuning optical filter (10) of claims 1, 2 and 4, wherein the intermediate mirror (102) has a higher reflectivity than the top mirror (101) and the bottom mirror (103).

6. 6. The wavelength tuning optical filter (10) of claims 1, 2, 4 and 5, wherein the intermediate mirror (102) has a lower reflectivity than the top mirror (101) and the bottom mirror (103).

7. 7. The wavelength tuning optical filter (10) of claims 1, 3, 4, 5 and 6, wherein the upper mirror (101) and the lower mirror (103) are supported by independent inner movable frames (204), which are fixed to an independent outer fixed frame (201) by a set of independent springs (203).

8. 8. The wavelength tuning optical filter (10) of claim 7, wherein a peripheral gap between the inner movable frame (204) and the outer fixed frame (201) is filled with a set of electrodes consisting of actuation electrodes (205) and detection electrodes (202).

9. 9. The wavelength tuning optical filter (10) of claims 4 and 8, wherein the actuation electrodes (205) comprise a set of electrodes having asymmetric vertical heights and having minimized lateral gaps between them, the electrodes being sequentially fixed to the outer fixed frame (201) and the inner movable frame (204).

10. The actuation electrodes (205) fixed to the inner movable frame (204) share the same potential, and the actuation electrodes (205) fixed to the outer fixed frame (201) also share the same potential; 10. The wavelength tuning optical filter (10) according to claims 4, 7, 8 and 9, wherein a potential difference is applied between the actuation electrode (205) fixed to the outer fixed frame (201) and the actuation electrode (205) fixed to the inner movable frame (204).

11. 11. The wavelength tuning optical filter (10) of claims 8, 9 and 10, wherein the detection electrode (202) has the same shape as the actuation electrode (205) and is arranged perpendicularly to and aligned with the actuation electrode (205).

12. 12. The wavelength tuning optical filter (10) of claim 1, wherein the nominal distance between the upper mirror (101) structure, the intermediate mirror (102) structure and the lower mirror (103) structure is ensured by a set of spacers (206) that facilitate mechanical support between the outer fixed frame (201) of the lower mirror (103) and the inner movable frame (204) of the intermediate mirror (102), and between the inner movable frame (204) of the intermediate mirror (102) and the outer fixed frame (201) of the upper mirror (101).

13. 13. The wavelength tuning optical filter (10) according to any one of claims 1 to 12, wherein the outer surfaces of the lower mirror (103) and the upper mirror (101), which are located outside two adjacent cavities having equal volumes, are provided with an anti-reflection coating (207).

Images