Patent Application
20140198388
Disclosed embodiments relate to Fabry-Perot cavity devices.
The Fabry-Perot (FP) cavity or FP filter can be used to measure the spectral components of an optical signal. The most basic form of a FP filter is two partially reflecting mirror surfaces, placed so that the mirror surfaces are facing each other and arranged to be highly parallel to one another, at a distance commonly referred to as being “d” apart. The inner facing mirror surfaces (mirrors) are vacuum deposited or otherwise formed on substrates of a suitable optically transparent material for the wavelength range of interest, and are often in the form of identical discs or plates.
The transfer function as a function of wavelength (T(λ)) of a basic FP filter assuming incident on-axis rays of light is given by:
Where R is the reflectivity of the mirrors, d is the separation distance (or gap) of the mirrors, and λ is the wavelength of the light being processed. The peaks in transmission of the FP filter (passband wavelength) occur at the values of λ given by:
where n=1, 2, 3, . . . .
The transfer function of the FP filter repeats, with the biggest gap being between peaks in terms of wavelength occurring between λ1 and λ2, this being equal to one octave, i.e., λ1=2λ2. If the desired wavelength measurement range is restricted to the range λ1 to λ2 then the FP filter provides an unambiguous measurement (only one peak) within this range, which is one octave. This restriction in turn places a restriction on the allowed range of d, which is that d varies between λ1/2 and λ1. The gap (d) is far smaller than usually encountered in “classical” FP spectroscopy, and offers the possibility of constructing a simple tunable optical filter using few passive components.
The spacing and parallelism of the plates for FP filters is known to be important. Since as described above the wavelengths at which transmission occurs is determined by d, so that it is important to know what d is. In addition, parallelism of the mirrors is important for FP filters since any deviation from this condition will cause the FP filter's transmission peak to collapse, so that the device will not work correctly. In “classical” FP spectroscopy, piezoelectric actuators are often used coupled to one or both mirrors in order to be able to adjust the parallelism, and to allow wavelength scanning over a limited range by changing d. It is common to mount 3 actuators symmetrically around the periphery of the discs (at 120° with respect to one another) in order to allow full control of the tilt of one mirror surface relative to the other mirror surface.
The width of the transmission peaks is governed by the Finesse of the FP filter which in turn depends on the reflectivity R of the mirrors. It is known to use FP filters in series to extend the free spectral range and the resolution. For example, see J. E. Mack et al in an article entitled “The PEPSIOS Purely Interferometric High-Resolution Scanning Spectrometer. I. The Pilot Model”, Applied Optics, Vol. 2, Issue 9, pp. 873-885 (1963), the subject matter in Mack is hereby incorporated by reference into this application.
Disclosed embodiments include Fabry-Perot (FP) cavity devices including in one embodiment FP tunable filters (FPTFs) having a membrane “movable mirror” including both front side and back side flexible (membrane) portions and an inner (e.g., center) bend resistant thicker portion in the beam path, where the inner thicker portion in the beam path resists bending to preserve the finesse. Disclosed embodiments recognize having both front side and back side flexible portions allow a suitable actuator to provide essentially only linear movement of the inner bend resistant portion of the movable mirror in the direction of the light beam.
Another disclosed embodiment is a FP device including multiple FPTFs including some embodiments with different gaps and thus different free spectral ranges, where the respective gaps are configured to narrow the composite passband of the device while still retaining good out-of-band rejection. In one embodiment the FPTF includes built-in capacitive sensing elements by including electroded mirrors where one of the mirrors is an electrically floating mirror.
Reference will now be made to the accompanying drawings, which are not necessarily drawn to scale, wherein:
Example embodiments are described with reference to the drawings, wherein like reference numerals are used to designate similar or equivalent elements. Illustrated ordering of acts or events should not be considered as limiting, as some acts or events may occur in different order and/or concurrently with other acts or events. Furthermore, some illustrated acts or events may not be required to implement a methodology in accordance with this disclosure.
A first disclosed embodiment comprises a FPTF having a flexible membrane mirror referred to herein as a “movable mirror” and a conventional non-movable mirror generally referred to herein as a top mirror. These mirrors together with a structure for moving or deforming the movable mirror define the tunable optical cavity of the FPTF device. The movable mirror includes an inner bend resistant portion in the path of the light beam, and a front side flexible portion and a back side flexible portion both outside the inner bend resistant portion. Because the front side and back side flexible portions are near both ends of the movable mirror and extend outward in two (2) directions from the inner bend resistant portion of the movable mirror, disclosed embodiments recognize that arrangement provides the advantage of constraining the inner bend resistant portion of the movable mirror to move essentially only in a highly one dimensional motion in the direction of the light beam (perpendicular to the mirror planes).
In one implementation, suitable for visible light applications, the top mirror 115, movable mirror 110 and back ring 120 are all made of fused silica. For infrared devices (wavelength>2200 nm with transmission spectrum data shown in
Regarding dimensions, for example, in one particular embodiment the front side flexible portion 110a is 0.027″ (685 μm) to 0.031″ (787 μm) thick, and the protrusion provided by the front side outer bend resistant portion 110b and bend resistant portion 110c extending beyond the front side flexible portion 110a is from 0.05″ (1270 μm) to 0.09″ (2286 μm) thick. Although bend resistant portion 110b can extend as shown in
The structure and formation of the back side flexible portion 110d is somewhat more complex. One example way to form back side flexible portion 110d is to first drill from the backside of the movable mirror 110 a plurality of holes 110f. A suitable tool can them be inserted into the holes 110f to remove the material (e.g. fused silica) in between and below adjacent back side flexible portions 110d. Such a tool can be used to define the back side flexible portion 110d thickness. The stiffness of back side flexible portion 110d is largely governed by its thickness, which can be 0.032″ (812 μm) to 0.038″ (965 μm), and the length and minimum width of the outer back side portion 110e which functions as bridges between the holes 110f. A similar suitable tool can be used to form the backside of the front side flexible portion 110a.
As noted above, the front side outer bend resistant portion 110b shown in
The spacer layer 119 can also be deposited using thin film selective deposition techniques onto the perimeter area of the second surface 115b of the top mirror 115. In yet another implementation, the spacer layer 119 can be deposited onto both the front side of the outer bend resistant portion 110b of the movable mirror 110 as well as on the matching area on the second surface 115b of the top mirror 115. After bonding the two surfaces together, the spacer layer 119 will separate the second surface 115b of the top mirror 115 and outer bend resistant portion 110b by an amount equal to the total spacer thickness for both mirrors (115/110). Thin film deposition techniques are well suited to deposit a uniform spacer layer (over its area) with a selectable thickness of a few hundred to a few thousand nm. This method ensures that the mirror surfaces forming the FP cavity are essentially parallel within the required accuracy with no adjustment needed. The thickness for the spacer layer 119 is chosen to provide sufficient separation of the two FP mirror surfaces such that the device can be operated in push-only configuration for the wavelength range of interest. The spacer layer 119 thickness can be selected based on the region of the spectrum targeted by the application. Example regions of the spectrum that can be targeted include visible light (e.g., 400 nm to 750 nm), and the mid-infrared region of the spectrum (e.g. 3-12 microns).
Now referring to
The flexible portions 110a and 110d of the respective sides of the movable mirror 110 can be formed in a variety of ways. In one embodiment, as noted above, the front side flexible portion 110a is formed by selectively removing material from the front surface to form a recessed annular depression, and the flexible back side portion 110d can be formed by selectively removing internal material from the body of the movable mirror 110 via access provided by the holes 110f in the back side of the movable mirror 110. The result of this process is the formation of a thin annular membrane to provide front side flexible portion 110a on the front surface of the movable mirror 110 and a back side surface having another flexible back side portion 110d containing the holes 110f.
The movable mirror 110 can be a single piece (e.g., a single glass piece) or a composite multipart assembly of two (2) or more pieces of material (e.g., glass) glued or otherwise optically contacted together. Thus, instead of forming the back side (bottom surface) of the flexible back side portion 110d from a single plate as described above, the back side flexible back side portion 110d can be a thin plate of separate piece of material (e.g., glass) having the desired thickness for flexible back side portion 110d glued or otherwise optically contacted to the bottom surface of the movable mirror 110 which has a top surface flexible membrane analogous to the front side flexible portion 110a formed by selectively removing material from recessed annular depressions from the top side, and a back side surface flexible back side portion similarly formed by selectively removing material from recessed annular depressions from the back side surface. This embodiment avoids the need for holes 110f (for access) and machining to selectively remove internal material from the body of the movable mirror 110.
Top mirror 115 is essentially flat and can be about 1 inch in diameter in one particular embodiment with the first surface 115a receiving the image beam 130 including an anti-reflective (AR) coating, and a thin high reflectivity coating (e.g. 97% for the wavelength of interest) on the second surface 115b. The movable mirror 110 can be about 1 inch in diameter and in a typical embodiment having a thin high reflectivity coating (e.g. 97% for the wavelength of interest) on the cavity side and an anti-reflecting (AR) coating on the other side. In a typical embodiment, the inner bend resistant portion 110c and outer bend resistant portion 110b are 3 to 20 times thicker as compared to thin front side flexible portion 110a which as noted can be a thickness range from 0.027″ to 0.031″, but more generally can be from 0.017″ to 0.041″ thick. The thicker inner bend resistant portion 110c, such as 0.13″ to 0.25″ preserves the finesse of the FPTF 100 for processing the image beam, and can allow movement of the movable mirror 110 by a single actuator element.
The actuator is shown in
As configured for FPTF device 100, the piezoelectric elements are biased to operate in pull mode, to provide displacement which acts to reduce the length of the gap 118 relative to the nominal length of the gap 118, such as at a rate determined by the piezoelectric coefficient (d)) of about 0.22 μm of displacement/10 V for PZT actuators. The piezoelectric coefficient d can be either positive or negative. To provide pull mode operation, the piezoelectric elements can be biased to become smaller in size in the direction of the image beam 130 which forces the back ring 120 shown in
The piezoelectric actuators can be separate pieces as shown in
As described above, a spacer layer 119 having a desired protrusion length can be provided on outer bend resistant portion 110b by selectively defining a layer of a deposited material (e.g., silica) having a thickness equal to the desired protrusion length. Since in this embodiment the top mirror 115 is bonded to the protruding spacer layer 119 on the outer bend resistant portion 110b, the spacer layer 119 provides a nominal gap 118 with respect to the top mirror 115, such as a nominal gap 118 of 0.15 μm to 0.45 μm in some embodiments. The use of this type of structure allows relatively simple initial alignment of the complete FPTF and simple actuation with a single actuator or multiple actuators driven in synchronization without the need for complex sensing and control.
In another embodiment, the FPTF does not include a spacer layer 119 so that the inner bend resistant portion 110c and outer bend resistant portion 110b on the front side are the same thickness and are thus co-planar. The result is that when the top mirror 115 and movable mirror 110 are bonded together there is no nominal gap 118 in this embodiment, and the actuation control works in the pull mode to pull inner bend resistant portion 110c away from the top mirror 115 to provide a controllable gap. The pull mode can also include a spacer layer 119 so that the minimum gap 118 is non-zero.
As noted above, in an alternate embodiment the movable mirror 110 can comprise two thin (e.g., glass) flexible membranes supporting a central mirror (inner bend resistant portion 110c). In this embodiment the movable mirror 110 features inner bend resistant portion 110c as a central mirror supported by two thin (e.g., glass) flexible membranes, one membrane provided by thin front side flexible portion 110a near the cavity surface of the movable mirror 110 and another membrane provided by a separate plate providing flexible back side portion 110d on the back side surface of the movable mirror. This alternate arrangement also constrains the front side flexible portion 110a to move only in a highly linear (one-dimensional) motion. The FPTF 100 having the movable mirror 110 including the front side flexible portion 110a and inner bend resistant portion 110c solves problems experienced by broad bandwidth large area (10 mm) FPTFs which need precise parallelism (<0.02 arc seconds) between two very flat (<λ/200) mirror plates placed close together (e.g., 100-300 nm). Precise adjustment of the gap 118 between the top mirror 115 and the inner bend resistant portion 110c of movable mirror 110 is provided while maintaining precise parallelism.
The protruding portion 124 of the back ring 120 attaches to the back side of the movable mirror 110 across the holes 110f. The holes 100f themselves are machined out holes used to form the flexible back side portion 110d so that the back ring 120 does not actually attach to them but to the material left between the holes 110f. Piezoelectric elements 123a, 123b and 123c can be glued to the outside region 126 of back ring 120. The length of protruding portion 124 is set to match the thickness of the piezoelectric elements 123. Protruding portion 124 is thus configured and attached (e.g., bonded) to the back side of the movable mirror 110 to leave an aperture 127 for light to exit the AR coating on the inner bend resistant portion 110c on the back side of the movable mirror 110 (shown as AR coating 136 in
In one example implementation, the tuning of the n=1 peak with force had a proportionality constant of 70 nm/N for a range of wavelengths covering the visible spectrum. The proportionality constant can be adjusted by choosing different thicknesses of the respective front side and back side flexible portions 110a and 110d and/or different distance between the inner bend resistant portion 110c and outer bend resistant portion 110b (such as by adding spacer layer 119 as described above). Portable devices which need low power consumption will use thinner and longer flexible portions to increase the proportionality constant, while a need for stability and immunity to vibration or acoustic noise will use a low proportionality constant.
Another embodiment comprises a FP device having built-in capacitive sensor(s). In this embodiment, two or more electrical capacitance gauges are used for in-situ measuring of the gap 118 and the parallelism of the FP filter plates (top mirror 115 and movable mirror 110). Although two capacitance gauges can be used, there are generally three or more gauges, such as three sets of capacitance gauges (e.g., 120 degrees apart), each gauge associated with a different actuator, such as a piezoelectric (e.g., PZT) actuator.
To reduce complexity of the FP device, one plate (the “base” plate) can be used for all the electrical connections, while only an isolated region of metallization (or other electrically conductive material) is used (one per finger pair) on the other upper electrically “floating” plate with no electrical connections to the upper plate (thus being an electrically floating electrode). The floating electrode bridges (is over) the two metal-coated regions on the lower base-plate in order that two air-spaced apart capacitors are formed in series, where the total capacitance between the finger pairs is based dependent on the spacing (d) between the base plate and the floating plate corresponding to gap 118. The capacitance between the fingers is negligible since the distance between the fingers in the finger pairs 219 is generally on the order of mms compared to the capacitance between the fingers and the floating electrode which is larger due to the distance (d) of the gap 118 between the plates (mirrors) being on the order of 1 μm (e.g., 0.15 μm to 0.45 μm). This is illustrated in the plate arrangement shown in
In
In
In
Three regions 291, 292, and 293 spaced at 120 degrees are shown non-metalized, except for a single metalized “finger” 285a, 285b, 285c present on each other non-metalized region 291, 292, and 293 (substrate 290 itself), respectively, which can be optionally folded over the edge of the substrate 290a in order to facilitate electrical connections. Alternatively, the bottom plate (or mirror) 280 can simply be made larger than the top plate (or mirror) 265, so that the fingers 285a, b, and c can extend radially beyond the region covered by the top plate (or mirror) 265, again, to facilitate electrical connection. The metalized region 284 on the bottom plate (or mirror) 280 is shown electrically grounded as it will be during operation.
When the top plate (or mirror) 265 is placed over the bottom plate (or mirror) 280, the capacitance between the metalized region 284 on the bottom plate and the metalized region 274 on the top plate (or mirror) 265 (C4 in the electrical equivalent figure shown in
Because the “sensor” capacitances C1, C2, C3 are far smaller than C4 (in the ratio of the areas of the fingers 285a, 285b, 285c as compared to the overlapping metalized region 274/284, they will dominate the sensitivity in total series capacitance to any changes in x1, x2, x3, where x1, x2 and x3 are the distances between the plates (or mirrors) 265 and 280 measured at fingers 285a, 285b, 285c, respectively. Measurement of the total series capacitance of C1 and C4 is possible between terminal (1) and ground, for example, and similarly for (C2, C4), (C3, C4) at terminals (2) and (3) respectively. These measurements can be used to monitor distances x1, x2, x3 and by using feedback to micropositioners (e.g., PZTs), to maintain parallelism and spacing of the top mirror 215 and movable mirror 210, explained in more detail below.
Regarding use of disclosed capacitive sensors to control mirror parallelism and thickness of the FP cavity, one example method control of the gap 118 is to employ 3 disclosed capacitive sensors at 120 degrees round the periphery of the FP cavity. The purpose of these capacitive sensors is to provide three independent distance measurements. The measurements can then be used to control the spacing and parallelism of the FP cavity by means of a look-up table and controller (e.g., processor-based) that outputs a voltage to each of the piezoelectric actuators in order to yield the gap 118 desired in order to tune to a given passband wavelength.
In one embodiment, each capacitor is used as the frequency-determining element in an electronic oscillator circuit. The measurement of the frequency of oscillation of each oscillator is can be used to determine the distance between the top mirror 115 and movable mirror 110 (=gap 118) at the location of that capacitor, by means of a previously compiled and stored look-up table. Assuming that all the capacitive sensors are of the same area, the parallelism condition will be achieved when all three oscillator frequencies are equal and no look-up table will then be needed. A feedback loop could be used during operation to make the mirrors maintain parallelism at all times by ensuring that at all times the voltages fed to the 3 piezoelectric s actuators caused the frequencies from the 3 respective oscillators to become equal.
Adding a constant bias voltage to each of the 3 voltages fed back to the piezoelectric actuators allows the gap thickness to be varied while simultaneously allowing the feedback loop to enforce the parallelism condition. The bias voltage used to determine the average gap thickness can be determined from a previously compiled look-up table. If the mirror parallelism condition is maintained, and the thicknesses measured at each sensor are the same, then the average gap thickness will be simply equal to the thickness of the gap 118.
HPF 170 is useful for avoiding transmission peaks for short wavelengths outside the target tuning range of FP filter 140 (as for example shown in
This embodiment recognizes improved FPTF optical performance can be obtained by using two (or more) FPTF's in series. However building and controlling FPTF devices having 2 gaps 118′, 118″ becomes significantly more complex as compared to building and controlling (e.g., tuning) an FPTF having a single gap 118, where tuning is facilitated by the FTFPs having disclosed movable mirrors. Embodiments including multiple FP filters in series solve the problem experienced by conventional single gap FPTFs which have limited optical performance in terms of a desired high ratio of maximum transmission on resonance versus minimum transmission of resonance. Single gap FPTF designs generally limit this ratio to ˜20%/0.2%=˜100:1. Disclosed dual gap FPTFs such as dual-cavity FP filter 140 can achieve a ratio of maximum transmission on resonance versus minimum transmission of resonance of ˜60%/0.06%=1000:1, this being about an order of magnitude (10×) better as compared to a single gap FPTF design.
The FPTF device having two or more FP filters in series can include different gaps for the respective filters. In this embodiment the respective FP filters each provide different free spectral ranges (FSRs) and the gaps are arranged so that overlap occurs between the transmission peaks of the lowest order (can call it n=1) of a first FP filter and a higher order peak (e.g., 2nd or 3rd, . . . ) of another (a second) FP filter. The higher order second FP has narrower transmission peaks but more peaks in a given wavelength band, so that this design ensures that its other peaks “miss” (i.e. do not overlap with) a transmission peak in the first FP filter. As shown in the data described below, this embodiment narrows the composite passband of the FP device while still retaining good out-of-band rejection. An advantage of this embodiment is that the composite peak resulting from the series combination of FP cavities is significantly narrower than would be obtained by having two identical FPTFs on the n=1 resonance in series, but retain the unambiguous 1 octave range. An issue for this embodiment can arise when working with a metal such as silver as the reflective surfaces in the visible light range because dispersion results in the blue end of the tuning range gets compressed making it difficult to keep spurious peaks out. However, by adding a suitable sharp cut-off HPF 170 with an edge near 450 nm to stop wavelengths <about 450 nm, the filter performance can be better than an identical pair of n=1 FP cavities in series in terms of overall passband width.
Disclosed embodiments are further illustrated by the following specific Examples, which should not be construed as limiting the scope or content of this Disclosure in any way.
Applications for disclosed FPTF devices include both imaging and non-imaging systems. Non-imaging systems include, for example, spectrometers and tunable laser filters.
Those skilled in the art to which this disclosure relates will appreciate that many other embodiments and variations of embodiments are possible within the scope of the claimed invention, and further additions, deletions, substitutions and modifications may be made to the described embodiments without departing from the scope of this disclosure.
Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
This application claims the benefit of Provisional Application Ser. No. 61/751,315 entitled “FABRY-PEROT CAVITY DEVICES”, filed Jan. 11, 2013, which is herein incorporated by reference in its entirety.
| Number | Date | Country | |
|---|---|---|---|
| 61751315 | Jan 2013 | US |
